Search

• This paper presents a Bayesian multi-task learning model for pest detection. • The model estimates size of aphids, which is an indicator of infestation severity. • It also quantifies the predictive uncertainties enabling confident decisions. • It performs thorough audit studies that offer insight for future model refinements. Aphids are persistent insect pests that severely impact agricultural productivity. The detection of aphid infestations is critical for mitigating their effects. This paper presents an artificial intelligence approach to detect aphids in crop images captured by consumer-grade RGB imaging cameras. In addition to detecting the presence of aphids, the size of the aphid is an important indicator of infestation severity. To address these, we present a Bayesian multi-task learning model to detect the presence of aphids and estimate their size simultaneously. Our model employs a joint loss function, combining a classification loss and a customised size loss. The classification component aims to identify images containing aphids, whilst the customised size loss function estimates the size of the aphids. The latter is specifically designed to account for discrepancies between the estimated and actual ground truth sizes, enhancing the accuracy of the size estimation. The model utilizes a ResNet18 backbone, ensuring robustness and adaptability across various conditions. The proposed model was evaluated using an agricultural pest dataset consisting of images of corn, rape, rice, and wheat crops. It achieved aphid presence detection accuracies of 75.77%, 66.39%, 70.01%, and 59% for corn, rape, rice, and wheat images, respectively. An in-depth evaluation of predictive uncertainties revealed areas of high confidence and potential inaccuracies for both size and presence of aphids in images, offering insight for future model refinement. We also conducted an ablation study to thoroughly analyse the contributions of each component in proposed model. Our model offers a valuable tool that can be used in pest management strategies for facilitating more sustainable and efficient agricultural practices. [ABSTRACT FROM AUTHOR]

Iraq's population will have reached 50 million by 2030, according to the Ministry of Planning. This population growth must be handled by ensuring food security stability in the face of expected obstacles resulting from global warming. The agricultural sector encounters limitations due to ineffective detection of persistent agricultural obstacles like pathogens, flawed soil analysis, insect infestations, inadequate irrigation, and drainage. More proven farming strategies must be transferred using modern technological leaps from this vantage point. It is clear from the outset that this paper aims to discuss how artificial intelligence might aid in the growth of conventional methods in agriculture. This involves tracking and guiding each stage of agriculture through the adoption of an intelligent system, which in turn includes managing a set of intelligent systems that control and supervise the rest of the processes (System of Systems or SoS), beginning with ploughing, crop pattern, type of fertilizer, and the appropriate type of irrigation. On the other hand, this technology collection could give farmers an accurate view of market trends and customer preferences, resulting in greater output. This paper gives an insight into the concept of a SoS and the application of artificial intelligence in various fields of agricultural engineering in Iraq. It also proposes an intelligence framework model for self-cultivation and incorporates competitive strategies to develop a decent ecosystem to transform traditional agriculture into smart and accurate agriculture. [ABSTRACT FROM AUTHOR]

The development of practical and robust automated diagnostic systems for identifying plant pests is crucial for efficient agricultural production. In this paper, we first investigate three key research questions (RQs) that have not been addressed thus far in the field of image-based plant pest identification. Based on the knowledge gained, we then develop an accurate, robust, and fast plant pest identification framework using 334K images comprising 78 combinations of four plant portions (the leaf front, leaf back, fruit, and flower of cucumber, tomato, strawberry, and eggplant) and 20 pest species captured at 27 farms. The results reveal the following. (1) For an appropriate evaluation of the model, the test data should not include images of the field from which the training images were collected, or other considerations to increase the diversity of the test set should be taken into account. (2) Pre-extraction of ROIs, such as leaves and fruits, helps to improve identification accuracy. (3) Integration of closely related species using the same control methods and cross-crop training methods for the same pests, are effective. Our two-stage plant pest identification framework, enabling ROI detection and convolutional neural network (CNN)-based identification, achieved a highly practical performance of 91.0% and 88.5% in mean accuracy and macro F1 score, respectively, for 12,223 instances of test data of 21 classes collected from unseen fields, where 25 classes of images from 318,971 samples were used for training; the average identification time was 476 ms/image. • Development of an accurate and practical framework for plant pest identification. • Evaluation of the model requires images, such as those collected elsewhere. • Foreground extraction significantly improves pest discrimination performance. • Class integration and cross-crop training boosts model performance. [ABSTRACT FROM AUTHOR]

• The ISP method allows for gathering a large number of historical training samples. • The SRVI values of the same crop planted in neighboring years range from 0.91 and 1.2. • A distinct northward expansion trend is evident for all four major crops, in Heilongjiang Province, China. Mapping agricultural information such as cropping area and type is of great significance for land use and food security and a common method to retrieve such information is satellite imagery classification. The current image classification techniques require a statistically large number of training samples to obtain representative spectral signatures, ideally collected during the satellite overpass time/date. Although this prerequisite sounds simple and easy in principle, it is extremely challenging and sometimes impossible when classifying historical satellite images, as one simply cannot go back in time to collect training datasets or ground truthing. In this paper, we introduce an iterative and subsequent proximation (ISP) method to circumvent this problem for historical image classifications for historical crop mapping. This method assumes that the same crops grown two years apart have similar spectral properties within the same growing season, with limited variation. This allows the use of the crop information classified in any given year (n) to be used as training samples for the next year (n + 1) or the previous year (n-1). Iteratively repeating n-1, n-2... process leads to the mapping of historical cropping information without a priori ground truthing data. To demonstrate the ISP feasibility, we first used the historical Landsat time-series data to map four major stable crops (rice, maize, soybean, and wheat) in Hailun County, Heilongjiang Province, China, and then further expanded the ISP application to the entire Heilongjiang Province to map cropping areas from 1982 to 2020, using the classification and regression trees (CART) algorithm. The results were validated for those years when actual cropping measurement data were available and further verified with statistical data for other years. The results indicated that the proposed ISP method was appropriate for the historical mapping of four major crops, with a mapping accuracy of approximately 80 % when validated with field data, and correlation coefficients of 0.86, 0.85, 0.91, and 0.94, respectively when compared with historical statistical cropping data. The crop mapping results showed distinct trends of northward expansion in each of the four crops in Heilongjiang Province, which agreed well with previous studies. In conclusion, the ISP method is quick, easy, and convenient for historical crop mapping, which is important in understanding the agricultural production history. [ABSTRACT FROM AUTHOR]

[Display omitted] • Farmers must be an allrounder for plant growth, plant protection, legislation and different economic fields. • AI service integration holds great potential for assistance, documentation, education, interpretation forecasts or data-driven predictions. • LLMs depict a fundamental step to reduce the gap between AI-driven data analysis and common user. • Reproduction of processed data would increase confidence in the model support farmer's interpretation and guide the LLM for more precise results. • Introducing these technologies requires not only new training but also significant effort in structural transformation. Since the launch of the "Generative Pre-trained Transformer 3.5", ChatGPT by Open, artificial intelligence (AI) has been a main discussion topic in public. Especially large language models (LLM), so called "intelligent" chatbots, and the possibility to automatically generate highly professional technical texts get high attention. Companies, as well as researchers, are evaluating possible applications and how such a powerful LLM can be integrated into daily work and bring benefits, improve their business or to make the research outcome more efficient. In general, underlying models are trained on large datasets, mainly on sources from websites, and online books and articles. In combination with information provided by the user, the model can give an impressively fast response. Even if the range of questions and answers look unrestricted, there are limits to the models. In this paper, possible use cases for agricultural tasks are elucidated. This includes the textual preparation of facts, consulting tasks, interpretation of decision support models in plant disease management, as well as guides for tutorials to integrate modern digital techniques into agricultural work. Opportunities and challenges are described, as well as limitations and insufficiencies. The authors describe a map of easy-to-reach topics in agriculture where the integration of LLMs seems to be very likely within the next few years. [ABSTRACT FROM AUTHOR]

Highlights • In this paper, a new method is proposed for estimating soybean defoliation. • Regression and convolutional neural networks were employed in our method. • We also presented three approaches to generate synthetic images. • The proposed method was trained only with synthetic images. • Experiments on real images showed excellent results. Abstract Agriculture plays an important role in the economy of several countries, contributing from the production of food and income to the generation of jobs. To improve productivity in agriculture, proper crop management should be accomplished through pest control. One approach is to monitor the defoliation level, that is, the percentage of leaf damaged by insects. Despite the importance, this monitoring is performed manually most of the time, which affects reliability as well as it is considered to be a time-consuming task. In this paper, we propose a new fully-automatic method to estimate the defoliation level based on convolutional neural networks (CNN). The main CNN architectures (AlexNet, VGGNet and ResNet) were adapted from classification to regression by replacing the softmax layer with a fully-connected layer with one neuron and a sigmoid activation function. Since CNNs require a large number of training examples, this paper also proposes approaches for generating synthetic defoliation images. In this way, our method is trained only with synthetic images and evaluated using real images. In the experiments, we obtained a root mean square error of only 4.57 even for images with severe defoliation. Additionally, we presented experimental evidence that the proposed method reconstructs the damaged leaf parts to then estimate the defoliation level. [ABSTRACT FROM AUTHOR]

With the competitive seeds market and development needs of seed companies, the production of crop seeds, including raw seeds procurement, processing, storage, packaging, and logistics, is complicated. Seeds packaging is one of the most important phases, including automatic and manual packaging. Automatic packaging is highly efficient with high setup costs, whereas manual packaging is the opposite. It is difficult for the managers to make the packaging planning, i.e., the number of automatic and manual packaging, based on the forecasted demand. This problem is considered to be a special dynamic lot-sizing problem but it is more complex compared to the general ones. To deal with this problem, this paper proposes a mathematical programming model with the objective of minimizing the total costs. Because the problem is Non-Deterministic Polynomial (NP) hard, the computational time for solving this model is very long. Therefore, this paper proposes a heuristic algorithm (HA) to solve the problem. Numerical results showed that the total costs calculated by the proposed HA are close enough (8.12%) to the optimal total costs of the proposed model and 22.6% lower than the traditional planning. At different scenarios, the HA can keep beneficial and reliable to a certain extent. The proposed model and HA provided detailed illustration for understanding the complexity of the seeds packaging planning and were able to deal with real situations in the seed company. [ABSTRACT FROM AUTHOR]

Land consolidation is the current issue occurring around the world. In the course of consolidation works the state of fragmentation of farms and its impact on their productivity is analyzed. Various methods of fragmentation assessing and variants of implementation of consolidation procedures are being developed. There are many factors of a different nature affecting the need for consolidation. One of the basic elements needed in these works is a quantitative description of flawed shape of plots based on the parameters of their spatial shape. GIS systems allow quick collection of spatial data of the plots or any other surface structures, including parcels. In this paper we focus only on the geometric shape of a parcel as a condition to carry out the consolidation. The paper proposes a methodology for obtaining parameters of the geometry of parcel description, which supports the process of decision-making in the field of land consolidation works. In this paper we proposed an algorithm to replace parcel with equivalent rectangle ( ER ). The presented solution enables quick processing of the large amounts of data. The results of algorithm use are shown on the example of the fragment of village in the Southern Poland. [ABSTRACT FROM AUTHOR]

This paper deals with a vegetable production-planning problem in which a farmer has to simultaneously determine the crop rotations and harvest schedules for a number of vegetables in his farmlands. The crop rotation decisions include the determination of crop sequence, cropping times, and cropping size in each farmland. The harvest decisions include the collection times and harvesting size for each planted crop. By taking into account the growing season and technical-ecological constraints, this paper formulates a mixed integer linear programming model to develop vegetable production decisions for organic farmers. Due to the complexity of the model, we proposed a two-phase heuristic approach to solve the model. We applied the proposed model to analyze an organic farm in Taiwan. Computational results show that the proposed heuristic is superior to CPLEX in terms of solution quality and computational times for large scale problems. In addition, some sensitivity analyses were also conducted to evaluate the impact of parameters in the model. [ABSTRACT FROM AUTHOR]

In recent years, the escalating labor costs in agricultural production have emerged as a major concern. The use of inspection robots to achieve automated inspection of fruit and fruit bunches for ripeness not only enhances production efficiency and cost savings, but also simplifies the tasks for workers. To address this issue, an improved YOLOv7-based multi-task deep convolutional neural network (DCNN) detection model, called MTD-YOLOv7, is proposed in this paper. Initially, the dataset labels were expanded to meet the requirements of multi-task classification. Two additional decoders were then added on the basis of YOLOv7 to detect tomato fruit clusters, fruit maturity and cluster maturity. Subsequently, the loss function was designed based on the characteristics of multi-task and the Scale-Sensitive Intersection over Union (SIoU) was used instead of Complete Intersection over Union (CIoU) to improve the model's recognition accuracy. Finally, to verify the effectiveness of the algorithm, tests were conducted on the cherry tomato dataset, and comparisons were made with common target detection algorithms, classification models, and cascade models. The experimental findings reveal that MTD-YOLOv7 achieved an overall score of 86.6% in multi-task learning, with an average inference time of 4.9 ms (RTX3080). It excels in simultaneous detection of cherry tomato fruits and bunches, fruit maturity, and bunch maturity, offering exceptional precision, rapid detection, and robust generalization capabilities. Its suitability extends to various applications, notably in inspection tasks. • A multi-task detection network for cherry tomato based on YOLOv7 was proposed. • Improving the cherry tomato dataset format for multitasking experiments. • Achieving three tasks: cherry tomato detection, fruit and bunch ripeness grading. • The proposed model has high speed while maintaining high accuracy. [ABSTRACT FROM AUTHOR]

Integrated planning of agricultural and industrial productions in an agrichain may increase the competitiveness of the entire chain and bring benefits to all of its agents. In turn, systems dynamics models could be used to explore and assess quantitatively these benefits. The main objective of this paper was to build a system dynamics model to assess whether integrated mechanisms of agricultural and industrial production planning can improve the competitive performance of the citrus agrisystem in Brazil. The scenarios tested by the model used the gradual introduction of new orange varieties and technological changes in the citrus production. Our paper considers that changes in these parameters have the potential to provide gains to the system by using the expanded capacity implemented by the industries throughout the year, by reducing inventories and blending operations, reducing costs by eliminating seasonal workers, among other factors. The results achieved from the model indicate that integrated production planning mechanisms can effectively improve the coordination of agro-industrial production systems. Moreover, it was demonstrated that models based on Systems Dynamics are suitable to assess these improvements. The model application and assessment results showed that integrated planning can increase the income per hectare of agricultural producers by 70% and the agribusinesses EBITDA margin by 43%. Thus, this study has shown that the use of integrated planning mechanisms is an important strategy for the Brazilian citrus agribusiness system to exploit its growth potential and to remain competitive. [ABSTRACT FROM AUTHOR]

• Proposing a rapid detection model of green pepper based on Yolov4_tiny under complex background. • Improving the Yolov4_tiny network by modifying multi-scale detection, adaptive feature fusion and attention mechanism. • Realizing the rapid detection of green peppers under the conditions of large target scale span, severe occlusion and overlap. • The average precision is 95.11%, the model size is 30.9 MB, and the frame rate is 89 FPS. • The proposed model contributes to monitor the growth of green peppers and automatic picking. In agricultural production, the branches and leaves of green peppers are severely blocked due to the dense plant distribution. This makes the identification of green peppers difficult. Traditional green pepper detection methods entail the problems of low accuracy and poor robustness. This paper introduces a deep learning target detection algorithm based on Yolov4_tiny for green pepper detection. The backbone network in the classic target detection algorithm model is used to ensure classification accuracy. This paper introduces an adaptive spatial feature pyramid method that combines an attention mechanism and the idea of multi-scale prediction to improve the recognition effect of occluded and small-target green peppers. Finally, the method was applied to a test set of 145 images (the target number of green peppers was 602). The AP value of green peppers reached 95.11%; the precision rate was 96.91%, and the recall rate was 93.85%. In order to verify the effectiveness of the module in improving detection performance, we conducted independent combined experiments on the improved module and compared the results with five classic target detection algorithms: SSD, Faster-RCNN, Yolov3, Yolov3_tiny, and Yolov4_tiny. The comparisons verified that the detection rate of the model reached that of the current state-of-the-art technology (SOTA) green pepper detection models. In addition, this green pepper detection model is suitable for real-time detection and embedded development needs of agricultural robots. Such new methods are key components of the technology for predicting green pepper parameters and intelligent picking in unmanned farms. [ABSTRACT FROM AUTHOR]

• This paper reviews the latest research progress of image fusion technology in agriculture. • A variety of image fusion methods are introduced to improve the fusion image quality and monitoring accuracy. • The advantages and limitations of various image fusion techniques in crop and livestock monitoring were discussed. • The challenges and prospects of image fusion technology in agriculture are described. Crop and livestock monitoring are essential for guiding agricultural production and improving agricultural yield by facilitating the early prevention and treatment of diseases, yield prediction, and automatic harvesting. However, traditional monitoring methods have some shortcomings, including by humans and through a single sensor, manual monitoring method is time-consuming and labor-intensive, monitoring methods using a single sensor are inaccurate. Therefore, research on improving the methods of crop and livestock monitoring is essential. Image fusion technology can help provide important information from two or more images that can be used to obtain comprehensive agricultural information. After discussing the advantages and disadvantages of various of image fusion methods in agriculture, this paper reviews the application of image fusion in crop recognition and detection, planting area estimation, plant diseases and pest detection, and livestock health assessment and classification. It also highlights the main challenges to the successful application of image fusion technology to agriculture, and discusses directions of future research and development directions in the area. [ABSTRACT FROM AUTHOR]

• A variable-rate seeding control system based on controller area network bus. • A location model was built for calculating the position of all planter units at the plane coordinate system. • A compensation algorithm of seeding lag was designed to decrease seeding lag distance. • The performance of the control system was evaluated relying on artificial prescription map and GPS simulation software. Variable-rate seeding (VRS) has had a significant impact on agricultural production by varying the seeding rate according to field conditions, which can save seeds and increase yield and net farm returns. However, domestic VRS control systems are presently unavailable in China, and the high cost of the imported VRS control systems severely restricts their adoption by domestic farmers. In this paper, a VRS control system that offers great expandability by utilizing a controller area network (CAN) and low cost by employing off-the-shelf components was presented. The developed control system was validated by laboratory experiments that were conducted at three travel speeds (6, 8, and 10 km/h) and three GPS frequencies (1, 2, and 5 Hz) compounded with seven possible seeding rates of 64,103, 69,444, 75,758, 83,333, 92,593, 104,167, and 119,048 seeds/ha. The results demonstrate that the developed compensation algorithm of seeding lag (CASL) could decreased seeding lag distance (LD) immensely: when the system applied CASL, LD values were in range 0.27–2.82 m, and average value was 0.99 m, while LD values increased to 1.35–4.93 m with average value of 2.79 m when the system didn't use CASL. Both travel speed and GPS frequency had a significant influence (P < 0.01) on the LD, but the variation of seeding rate had no significant influence (P > 0.05), so a smaller LD could be obtained at a lower travel speed or higher GPS frequency. The largest seeding accuracy (SA) and seeding coefficient of variation (SCV) values, which assesses the control accuracy and stability of the VRS control system for varying seeding rate, were 99.68% and 0.85%, respectively, while the smallest SA and SCV values were 98.87% and 0.16%, respectively. Using the technologies of a CAN bus and algorithm for calculating the locations of planter units, the developed VRS control system was capable of flexibly expanding planter rows and independently controlling each row's seeding rate. Further field trials were conducted in the sequel to this study (Part II: Field performance). [ABSTRACT FROM AUTHOR]

Highlights • A decision support tool for farmer's strategic thinking. • An activity concept to multi-criteria and multi-scale analyses. • Long rotations increased income but also labor needs. • Crop contracts contributed to the appearance of weeds resistance. Abstract In an uncertain socio-economic and climatic context, sustainable farming is a major challenge for farmers as well as for their agricultural advisors. It is therefore essential to develop a decision support tool (DST) that is likely to be useful to establish and evaluate new production strategies, in accordance with farm sustainability and environmental protection. This paper aims, by using the Agricultural Activity concept, to put forward and test a DST based on mathematical programming used to evaluate strategic production decisions, in conjunction with farmers and agricultural advisors. In the test case, the decisions concern the crop activities and their spatio-temporal combinations in order to reduce both the use of herbicides and the risk of weed resistance to herbicides in cereal-based production systems. Moreover, the DST considers the availability of the workforce during the crop cycle to determine the periods which are likely to require the most significant increase in comparison to the current situation. One scenario showing the current situation (Sc_baseline) and two alternative scenarios have been defined to address the weed-herbicide issue. The comparison of the scenario promoting soil tillage and the introduction of spring crops with Sc_Baseline has shown that the adoption of long-term rotations, the increase in winter crop frequency and the return to deep soil tillage have contributed to an increase in farmer income, total labor and water consumption by 7, 21 and 22% respectively. However, the intensity of pesticide use and nitrate fertilization have dropped by 15% and 17% respectively. By allowing the farmer to establish specific contracts for certain crops, the average income as well as the use of pesticide and nitrate fertilization were increased by at least 10%. This situation is the result of a simplification of rotations with a predominance of winter cereals and the elimination of deep soil tillage. The analysis of these scenarios shows that the use of the DST has made it possible not only to put forward and evaluate alternatives that result in strategic decisions but also to understand, with the concept of Agricultural Activity, the biophysical and technical processes relating to farmer decisions and their impacts at field and farm level. Understanding and sharing this functional chain at farm level is expected to strengthen the farmer-advisor relationship in order to address the complex challenges of farming system sustainability. [ABSTRACT FROM AUTHOR]

Deep learning constitutes a recent, modern technique for image processing and data analysis, with promising results and large potential. As deep learning has been successfully applied in various domains, it has recently entered also the domain of agriculture. In this paper, we perform a survey of 40 research efforts that employ deep learning techniques, applied to various agricultural and food production challenges. We examine the particular agricultural problems under study, the specific models and frameworks employed, the sources, nature and pre-processing of data used, and the overall performance achieved according to the metrics used at each work under study. Moreover, we study comparisons of deep learning with other existing popular techniques, in respect to differences in classification or regression performance. Our findings indicate that deep learning provides high accuracy, outperforming existing commonly used image processing techniques. [ABSTRACT FROM AUTHOR]

To tackle the increasing challenges of agricultural production, the complex agricultural ecosystems need to be better understood. This can happen by means of modern digital technologies that monitor continuously the physical environment, producing large quantities of data in an unprecedented pace. The analysis of this (big) data would enable farmers and companies to extract value from it, improving their productivity. Although big data analysis is leading to advances in various industries, it has not yet been widely applied in agriculture. The objective of this paper is to perform a review on current studies and research works in agriculture which employ the recent practice of big data analysis, in order to solve various relevant problems. Thirty-four different studies are presented, examining the problem they address, the proposed solution, tools, algorithms and data used, nature and dimensions of big data employed, scale of use as well as overall impact. Concluding, our review highlights the large opportunities of big data analysis in agriculture towards smarter farming, showing that the availability of hardware and software, techniques and methods for big data analysis, as well as the increasing openness of big data sources, shall encourage more academic research, public sector initiatives and business ventures in the agricultural sector. This practice is still at an early development stage and many barriers need to be overcome. [ABSTRACT FROM AUTHOR]

Crop diseases are responsible for the major production reduction and economic losses in agricultural industry worldwide. Monitoring for health status of crops is critical to control the spread of diseases and implement effective management. This paper presents an in-field automatic wheat disease diagnosis system based on a weakly supervised deep learning framework, i.e. deep multiple instance learning, which achieves an integration of identification for wheat diseases and localization for disease areas with only image-level annotation for training images in wild conditions. Furthermore, a new in-field image dataset for wheat disease, Wheat Disease Database 2017 (WDD2017), is collected to verify the effectiveness of our system. Under two different architectures, i.e. VGG-FCN-VD16 and VGG-FCN-S, our system achieves the mean recognition accuracies of 97.95% and 95.12% respectively over 5-fold cross-validation on WDD2017, exceeding the results of 93.27% and 73.00% by two conventional CNN frameworks, i.e. VGG-CNN-VD16 and VGG-CNN-S. Experimental results demonstrate that the proposed system outperforms conventional CNN architectures on recognition accuracy under the same amount of parameters, meanwhile maintaining accurate localization for corresponding disease areas. Moreover, the proposed system has been packed into a real-time mobile app to provide support for agricultural disease diagnosis. [ABSTRACT FROM AUTHOR]

Many relevant studies about screening of agriculture materials have been conducted from different perspectives such as mechanism design, optimization of parameters, and particle motion. Some studies suggest that a screen driven by parallel mechanisms is more adaptable than a traditional one while the traditional reciprocating screen still plays a positive role in agricultural production. Based on previous research regarding the variable-amplitude screening method, 4 indices of particle movement have been defined and computed in this paper to investigate the quantitative dispersion and migration characteristics of agricultural particles by using DEM (Discrete Element Method) simulation data. It shows that the turning angle of the front swing bar has a significant effect on the horizontal expansion coefficient δ x ( t ) , and the particles will be thin quickly at first and be stabilize within the next process. It also shows that the increase in the turning angle of the front swing bar leads to a negative stratification effect, but results in a positive migration effect of all particles on the screen. This research could provide a useful reference for solving the retention problem of agricultural particles in any position on the screen. [ABSTRACT FROM AUTHOR]

The use of low cost cameras has been extended in all fields of technology in general, and agricultural applications in particular. Images provide useful information on the growing state of horticultural crops, which allow an accurate estimation of water balance and, hence, precise irrigation scheduling. In all of these cases, the temporary images of a crop can provide the percentage of green cover (PCG). This data is calibrated with the irrigation water amount that the crop needs for growing. Therefore, the use of visual monitoring systems in agriculture may reduce water consumption and increase productivity. In this paper, a novel system is presented using low cost cameras and a client-server architecture. It is composed of a set of inexpensive camera modules which communicate with a cloud computing server. Camera modules have been developed using open standard Arduino components; they are able to work independently, with their own connectivity, storage and power supply. On the other hand, the server is responsible for configuring these modules, performing computer vision algorithms and water balance estimation, storing all data in a secure database, and interacting with the user interface using the web. The final result is a complete and inexpensive system that allows continuous monitoring of the state of the crops, providing the user with valuable information about water balance for irrigation management. The proposed method achieves a high accuracy in the estimation of PGC, with an average error below 5%, requiring less than 2 s of processing time per image in the server. This is transformed into an error in the computation of the crop coefficient below 1%. Technical details on the hardware and software components of the system are presented. Finally, advantages and weaknesses of the proposed solution are discussed, drawing new lines for future research. [ABSTRACT FROM AUTHOR]

• There is not enough literature on optimization in crop maintenance tasks. • The cost of waste in crop tasks can be reduced by optimization and lean principles. • This model is a first approach for integrating lean principles to agricultural tasks. • Response surface is used as tool for optimizing results of MINLP problem. The operational planning for agricultural production systems become an important tool for decision making for farmers. This paper shows a mathematical model as a decision support tool for crop maintenance planning problems. The model seeks the minimization of waste generated by the use of resources necessary for the task of crop maintenance. The model is inspired on the types of wastes from Lean Manufacturing (LM), which includes labor, use of machinery and operations in time windows that preserve the quality of harvest. The problem is modeled as a Mixed Integer Nonlinear Program (MINLP) in a case study on banana farms in Colombia. The model was tested with different scenarios, operations for crop maintaining, and number of workers. Then, we used the Response Surface Methodology (RSM) for sensitivity analysis, and obtain a significant reduction in cost, resources optimization, wastes, and schedule of crop maintenance activities: a reduction of 59% on the total cost was obtained, as well as a new approach for activities and task scheduling. The contributions of this paper to the body of knowledge in agricultural planning are at least twofold. Firstly, in contrast with traditional modeling approaches, this paper integrates the principles of lean manufacturing (LM) into the definition of parameters and constraints in a mathematical model for crop maintenance. Secondly, the paper couples nonlinear mathematical programming with response surface methodology (RSM) to carry out sensitivity analysis on the results obtained by the mathematical model. It is to note also that, although the modeling and solution methodology follow a case study research approach, it can be applied to other agricultural planning problem at operational level, being an economic option to minimize wastes and resources in others crops. [ABSTRACT FROM AUTHOR]

•. In-depth systematic review of computational methods to detect the diseases in major grains. • Review of 120 articles related to disease detection in major gains rice, wheat, maize, etc. since 2007. • Pros and cons of each method used in various phases of plant disease detection were discussed. • Findings and discussions, further direction will be helpful for researchers in future research. • The challenges in pathogens identification through image processing approach are deeply discussed. Agricultural productivity significantly contributes to every country's economy. The grain plants such as wheat, rice, corn (maize), barley, oats, rye, millet, and sorghum are commonly grown across the world. The pest and various diseases in the grain plants severely affect major production and cause heavy losses in the global economy. Monitoring of health and early diagnosing of diseases in grains plants is a critical task for sustainable agriculture. The information on early diagnosis of several diseases can facilitate the control of diseases through proper selection of pest control techniques to improve the grains productivity. The manual identification of the disorders in grains plants can lead to inaccurate measurements of pesticides. While several papers on grain diseases identification through intelligent techniques have been published in recent years, there has been no clear attempt to study these papers systematically to describe various phases of diagnosis system such as image preprocessing, segmentation, feature extraction, features selection, and classification methods. In this context, a total of 109 peer-reviewed articles reporting to identify the diseases at the early stage to increase the production of the five most-produced grains in the world such as maize, rice, wheat, soybean, and barley are reviewed, ranging in publication date from 2001 to 2020. The article also presents a detailed taxonomy of grain plant leaf diseases. The study found that there are still many issues that's need to be addressed in each phase of the automated disease detection system. The pros and cons of reviewed computational method is explored and future directions are highlighted. The survey outcomes reveal that the existing automated detection and classification methods for grain plants diseases is still infancy. Hence novel fully automated tools are necessary for the process of detection and classification diseases in grain plants. [ABSTRACT FROM AUTHOR]

• There is a need for public image datasets to enable precision agriculture tasks. • The 34 public image datasets collected under field conditions are surveyed. • The practices on facilitating public image dataset creation are discussed. Computer vision technologies have attracted significant interest in precision agriculture in recent years. At the core of robotics and artificial intelligence, computer vision enables various tasks from planting to harvesting in the crop production cycle to be performed automatically and efficiently. However, the scarcity of public image datasets remains a crucial bottleneck for fast prototyping and evaluation of computer vision and machine learning algorithms for the targeted tasks. Since 2015, a number of image datasets have been established and made publicly available to alleviate this bottleneck. Despite this progress, a dedicated survey on these datasets is still lacking. To fill this gap, this paper makes the first comprehensive but not exhaustive review of the public image datasets collected under field conditions for facilitating precision agriculture, which include 15 datasets on weed control, 10 datasets on fruit detection, and 9 datasets on miscellaneous applications. We survey the main characteristics and applications of these datasets, and discuss the key considerations for creating high-quality public image datasets. This survey paper will be valuable for the research community on the selection of suitable image datasets for algorithm development and identification of where creation of new image datasets is needed to support precision agriculture. [ABSTRACT FROM AUTHOR]

• The proposed method H2K is a fusion of Harris corner detector, HOG (Histogram on Oriented Gradient) and KNN classifier for accurate detection and classification of groundnut leaf diseases. • The H2K method works well on sample images and detects and classifies 5 major groundnut leaf diseases. • It includes late spot detection which is difficult to control if it is not identified in the early stages. • The H2K is robust and optimum in classifying groundnut leaf diseases with improved accuracy of 97.67%. One of the greatest key factors contributing to low yield and destruction of groundnut plant growth is leaf disease attack. The groundnut plants are prone to the diseases such as fungi, soil borne and viruses. The leaf disease detection and identification in the early stage is needed to control the spread of infection and also helps to get the maximum yield. The detection and classification of groundnut leaf diseases through naked eye observation by an expert is expensive that too developing countries. So, providing a software based solution for the above task is of great significance. In this paper, an image processing based approach that automatically identifies and categorizes the groundnut leaf diseases has been presented. The proposed method H2K is a fusion of Harris corner detector, HOG (Histogram on Oriented Gradient) and KNN classifier for accurate detection and classification of groundnut leaf diseases. It comprises of number of steps viz. image acquisition, image pre-processing by applying binary mask, HSV segmentation to segment the disease affected part, features detection and extraction using H2K (Harris, HOG and K-Nearest Neighbor) based classification of groundnut leaf diseases. The existing works concentrates on leaf diseases that are commonly occurring in any crops, but this paper proposed a robust and optimum approach for detection and classification of all major leaf diseases for Groundnut crop which is first of its kind. Thus the H2K method aids to improve the crop production and maximizes the yield. The proposed method H2K works well on sample images and detects and classifies 5 major groundnut leaf diseases including late spot which is difficult to control if it is not identified in the early stages. The existing Multiclass SVM is taken for comparison and the examined results shows that the H2K is robust and optimum in classifying groundnut leaf diseases with improved accuracy of 97.67%. [ABSTRACT FROM AUTHOR]

The world of viticulture connected to wine making has become a very important activity in many inland areas permitting both the generation of important income and the sustaining of agriculture systems. Recent progress in both crop modeling and Decision Support Systems (DSS) applied to viticulture promises important changes that combine both high quality production and environmental sustainability. However, most of this progress is only addressed at the farm level and does not challenge the viticulture landscape, which is a key issue when facing DOC, DOCG areas, wine growers' cooperatives and consortiums and strategic viticulture planning. Thus, this paper aims to demonstrate that a new type of DSS, which is developed on a Geospatial Cyberinfrastructure (GCI) platform, may provide an important web-based operational tool for high quality viticulture as it connects farm and landscape levels better. The GCI platform supports acquisition, management, processing of both static and dynamic data (e.g. pedological, daily climatic, and vineyard distribution), data visualization, and on-the-fly computer applications in order to perform simulation modeling (e.g. grapevine water stress, evaluation of ecosystem services, etc.). These are all potentially accessible via the Web. This is possible thanks to the implementation of a set of modeling clusters that is strongly rooted in soil-plant-atmosphere and physically based simulation modeling. The DSS tool, applied to an area of 20,000 ha in Southern Italy, is designed to address viticulture planning and management by providing operational support for farmers, farmer associations and decision makers involved in the viticulture landscape. Output of the system includes viticulture planning and management scenario analysis, maps and evaluation of potential and current plant water stress. The tool will also be demonstrated through a short selection of practical case studies. [ABSTRACT FROM AUTHOR]

Agricultural production has become a key factor for the stability of the world economy. The use of pesticides provides a more favorable environment for the crops in agricultural production. However, the uncontrolled and inappropriate use of pesticides affect the environment by polluting preserved areas and damaging ecosystems. In the precision agriculture literature, several authors have proposed solutions based on Unmanned Aerial Vehicles (UAVs) and Wireless Sensor Networks (WSNs) for developing spraying processes that are safer and more precise than the use of manned agricultural aircraft. However, the static configuration usually adopted in these proposals makes them inefficient in environments with changing weather conditions (e.g. sudden changes of wind speed and direction). To overcome this deficiency, this paper proposes a computer-based system that is able to autonomously adapt the UAV control rules, while keeping precise pesticide deposition on the target fields. Different versions of the proposal, with autonomously route adaptation metaheuristics based on Genetic Algorithms, Particle Swarm Optimization, Simulated Annealing and Hill-Climbing for optimizing the intensity of route changes are evaluated in this study. Additionally, this study evaluates the use of a ground control station and an embedded hardware to run the route adaptation metaheuristics. Experimental results show that the proposed computer-based system approach with autonomous route change metaheuristics provides more precise changes in the UAV’s flight route, with more accurate deposition of the pesticide and less environmental damage. [ABSTRACT FROM AUTHOR]

This paper examined the determining factors in decisions of citrus farmers on adoption of computers and Farm Management Information Systems (FMIS). Primary data were collected from a random representative sample of 98 citrus farmers from the state of São Paulo, Brazil. The data was analyzed using logit and count data (Poisson regression) models, which enabled testing hypotheses on the effect of ten variables on the decisions of farmers. The results of the logit model showed that education and production size had a positive and statistically significant effect on the adoption of computers, while experience had a negative effect. The adoption and intensity of use of FMIS were influenced positively by overconfidence in management, production size and use of technical assistance. Contract adjustments and farmers’ experience have a negative impact on the adoption of FMIS. The results confirmed the main hypotheses and can contribute to the development of new strategies for greater diffusion of FMIS in Brazilian citrus industry, which is relevant to increasing farm efficiency. [ABSTRACT FROM AUTHOR]

Plant diseases is one of the major bottlenecks in agricultural production that have bad effects on the economic of any country. Automatic detection of such disease could minimize these effects. Features selection is a usual pre-processing step used for automatic disease detection systems. It is an important process for detecting and eliminating noisy, irrelevant, and redundant data. Thus, it could lead to improve the detection performance. In this paper, an improved moth-flame approach to automatically detect tomato diseases was proposed. The moth-flame fitness function depends on the rough sets dependency degree and it takes into a consideration the number of selected features. The proposed algorithm used both of the power of exploration of the moth flame and the high performance of rough sets for the feature selection task to find the set of features maximizing the classification accuracy which was evaluated using the support vector machine (SVM). The performance of the MFORSFS algorithm was evaluated using many benchmark datasets taken from UCI machine learning data repository and then compared with feature selection approaches based on Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) with rough sets. The proposed algorithm was then used in a real-life problem, detecting tomato diseases (Powdery mildew and early blight) where a real dataset of tomato disease were manually built and a tomato disease detection approach was proposed and evaluated using this dataset. The experimental results showed that the proposed algorithm was efficient in terms of Recall, Precision, Accuracy and F-Score, as long as feature size reduction and execution time. [ABSTRACT FROM AUTHOR]

Leaf diseases have become more prevalent in recent years due to climate change, increased growth of outdoor air pollutants, and global warming. They may severely damage crop yield, leading to detrimental effects on global food security. The timely and precise detection of leaf diseases is thus crucial for preventing their spread and ensuring the sustainability of agricultural production. In this paper, we introduce a transformer-based leaf disease detection model, namely FormerLeaf along with two optimization methods to enhance the model performance. In more detail, we propose the Least Important Attention Pruning (LeIAP) algorithm to select the most important attention heads of each layer in the Transformer model. It could reduce the model size up to 28% and accelerate the evaluation speed by 15% with about 3% accuracy enhancement. In addition, we employ the sparse matrix-matrix multiplication (SPMM) to calculate matrix correlation in the model. This reduces the model's complexity from O (n 2) to O ( n 2 p) , resulting in lowering training time by 10% while keeping a similar performance. The evaluation results on the Cassava Leaf Disease Dataset show that our proposal outperforms the state-of-the-art models in most cases. • Proposing an efficient transformer-based model to precisely detect leaf diseases. • Designing a pruning algorithm to optimize attention heads of each model layer. • Employing sparse matrix-matrix multiplication to reduce the training time. [ABSTRACT FROM AUTHOR]

Determining the shape of agricultural products is important in many areas of the food industry, such as for classification or quality inspection. The characterization methods used to describe the behaviour of particulate systems during handling may also benefit from an accurate description of the particle shapes. Several methods for particle shape representation have been proposed, including those using super-quadric equations, polygon formulations, or composite particles. However, it has been proved that these methods are not accurate enough to represent complex-shaped particles. The use of spherical harmonics has recently received increasing attention for this purpose. In this paper, spherical harmonics are used to obtain the shapes of three agricultural grains, namely bean, chickpea, and maize, which differ in complexity. Once it has been proved that spherical harmonics can accurately describe the shapes of agricultural grains, the advantages and disadvantages of this technique are discussed. Furthermore, the relationship between spherical harmonics and the discrete element method for the simulation of particle systems is also discussed. [ABSTRACT FROM AUTHOR]

Site-specific crop management (SSCM) is a part of precision agriculture which is helping increase production with minimal input. It has enhanced the cost-benefit scenario in crop production. The main goal of this paper was to use advanced geospatial techniques in data acquisition, remote sensing (RS), image processing, geographic information systems (GIS), global positioning systems (GPS) and statistical modeling to determine the correlation between image digital information and healthy olive trees growth and production management characteristics. It is to be noted that assumptions were based on the overall canopy greenness of the olive trees as healthy trees. This research was carried out during 2012–2014 in an irrigated olive orchards located in the Tarom region, Zanjan province of Iran. The following data were gathered: fruit set percent in shoot, canopy volume (CV), shoot length (SL), trunk diameter (TD), trunk height (TH), soil plant analysis development (SPAD), leaf area index (LAI), leaf dry matter percent (LDMP), leaf properties like nitrogen (N) and potassium (K) content in leaves, soil properties/characteristics like amount of Clay, Silt, Sand, Sodium adsorption rate (SAR), organic matter (OM), available phosphorous (P av ), available potassium (K av ), boron (B), total neutralizing value (TNV), electrical conductivity (EC), chloride (Cl), available iron (Fe av ). Advanced land observing satellite-Advanced visible and near infrared radiometer type 2 (ALOS-AVNIR-2) image was used in this experiment. A set of six clusters of olive trees existing in a compacted parcel of olive orchards were chosen. The image indices developed for this study were the normalized digital vegetation index (NDVI), newly developed vegetative vigor index (VVI) and the soil adjusted vegetation index (SAVI). Multivariate regression models were developed using remotely sensed image digital values in relation to the site specific crop growth parameters as mentioned above. As is stated above, individual band DN value statistics as input parameters and plant growth characteristics such as CV, SL, TD, TH, SPAD, LAI, and LDMP as output parameters were used in the multivariate regression models development. Multicollinearity analyses were completed on the input parameters to reduce redundancy of data usage. Multicollinearity analysis of the image related variables shows that VVI and b1 are highly correlated with other variables. It was also observed that NDVI – b3 and b2-b4 are highly correlated and hence omitted from the input parameter list. The multivariate regression models developed with NDVI and SAVI along with individual band (Green, Red and Infrared bands) as input parameters for olive crop growth parameters like TD, TH and SPAD provided excellent coefficient of determination (R 2 ) values of 0.98, 0.99 and 0.99, respectively. SAVI, Red-, Green-, and Blue-band image information together best estimated the olive tree canopy volume with R 2 value of 0.84. Similarly, SAVI, Red-, Green-, and Blue-band image information together also best estimated the olive tree SL and LA with R 2 value of 0.88 and 0.96, respectively. SAVI, NDVI, Red-, Green-, and Blue-band image information as input parameters estimated the olive tree trunk diameter with R 2 value of 0.98. The same SAVI, NDVI, Red-, Green-, and Blue-band image information together best predicted the olive tree trunk height with maximum correlation (R 2 = 0.99). Similarly SPAD and LDMP were estimated with excellent correlations (R 2 = 0.99 and 0.79, respectively) using image related input parameters of SAVI, NDVI, Red-, Green-, and Blue-band. Algorithms developed with this study could be used by farmers or orhard managers for estimating the olive tree physical characteristics in similar environmental conditions that prevailed in our study area using remotely sensed imagery in a non-invasive, economic, and efficient manner. [ABSTRACT FROM AUTHOR]

• A one-stage deep object detector U3YOLOXs for detecting subhealth regions in rape. • Pre-train the model with a self-supervised strategy on the self-collected dataset. • Three improvements in the backbone, neck and head network for solving the U3 problem. • The comparative study between average accuracy, model size and inference latency. Agricultural production in high latitudes could be limited by cold climate. Plant factory allows continuous production all year round, where the detection of plant growth is one of the most important tasks. To achieve non-destructive detection of rape in our plant factory, a feasible approach is to automatically detect subhealth areas from the rape images. However, this task faces the following challenges: (1) U ncommon problem: the subhealth regions on rape are the domain-specific objects, so the mainstream transfer learning-based detector is unreliable; (2) U nregular problem: the subhealth regions are difficult to detect due to their variable shapes, sizes and colors. (3) U nbalance problem: localization and classification of subhealth regions each have low-/high-quality bounding box unbalance and easy/hard sample unbalance. In this paper, a novel deep object detector based on the YOLOXs, called U3YOLOXs, is proposed for the detection of subhealth regions on rape at the bolting stage. Specifically, a domain-specific self-supervised pre-training strategy in the backbone is developed for the Uncommon problem; next, a coordinate attention mechanism in the multi-scale neck network is built for the Unregular problem; finally, the focal EIoU and the focal loss in the decoupled head are designed for the Unbalance problem. The experimental results show that the mAP of our U3YOLOXs is 94.38 % with a latency of 20.4 ms per image, which achieves an optimal accuracy-speed tradeoff. Compared to the YOLOXs it achieves a significant improvement of 9.27 % on mAP at the cost of only 3.55 % increase in latency. Experimental analysis further shows the effectiveness of each improvement, and the reliability of porting U3YOLOXs to edge devices for agricultural production. [ABSTRACT FROM AUTHOR]

• Four requirements of Agriculture 4.0 are concluded. • Thirteen agricultural decision support systems are presented in a systemic view. • Selected systems are evaluated from eight aspects. • Seven challenges for new decision support systems are detected. Undoubtedly, high demands for food from the world-wide growing population are impacting the environment and putting many pressures on agricultural productivity. Agriculture 4.0, as the fourth evolution in the farming technology, puts forward four essential requirements: increasing productivity, allocating resources reasonably, adapting to climate change, and avoiding food waste. As advanced information systems and Internet technologies are adopted in Agriculture 4.0, enormous farming data, such as meteorological information, soil conditions, marketing demands, and land uses, can be collected, analyzed, and processed for assisting farmers in making appropriate decisions and obtaining higher profits. Therefore, agricultural decision support systems for Agriculture 4.0 has become a very attractive topic for the research community. The objective of this paper aims at exploring the upcoming challenges of employing agricultural decision support systems in Agriculture 4.0. Future researchers may improve the decision support systems by overcoming these detected challenges. In this paper, the systematic literature review technique is used to survey thirteen representative decision support systems, including their applications for agricultural mission planning, water resources management, climate change adaptation, and food waste control. Each decision support system is analyzed under a systematic manner. A comprehensive evaluation is conducted from the aspects of interoperability, scalability, accessibility, usability, etc. Based on the evaluation result, upcoming challenges are detected and summarized, suggesting the development trends and demonstrating potential improvements for future research. [ABSTRACT FROM AUTHOR]

• Practical approach to measure & monitor the contribution of IoT to sustainability. • We developed SDG-based typology to measure impact using key performance indicators. • The novel approach enables actors to test and evaluate IoT i.r.t. sustainability. • We show evidence how IoT can contribute to sustainable agricultural production. • Supports communication between multiple stakeholders incl. policy makers. Sustainability is a major challenge in agri-food systems. Digital technologies, such as Internet of Things (IoT) hold substantial promises for attaining the sustainability goals of the economy, environment and society at large. However, in practice it is difficult to evaluate to which extent these technologies contribute to sustainable development raising doubts about their impact. This paper demonstrates a stepwise approach that allows for measuring and monitoring IoT contribution to sustainability in a real-life context. The UN sustainable development goals (SDGs) underpin the principles of the approach by a typology and by framing the sustainability impact in terms of business opportunities. The approach has been developed and evaluated by 33 use cases in the EU-funded IoF2020 project. The research illustrates how the measurement and monitoring tool is applied in 5 of these use cases from different agricultural subsectors showing how the approach is applied and validated. The results indicate an overall positive impact of IoT on improving sustainability, although these results are also partly determined by other influential external factors that cannot be easily discerned in a practical situation. The main contribution of this approach is the set of instruments for practitioners to measure and monitor the impact of fast-changing technologies such as IoT to sustainability in a real-life context. This set of instruments can also be used by other stakeholders in large IoT projects where strategic sustainability objectives should be supported by IoT solutions. The stepwise approach is easy to communicate and supports stakeholders such as farmers in decision-making, but also policy makers and investors in funding projects. [ABSTRACT FROM AUTHOR]

The forest industry is a capital intensive business and therefore high efficiency in management and forest operations is a must. Maine has millions of acres of forest stands with tree diameters smaller than 30 cm. The harvest productivity in such stands is low compared to stands with larger diameter trees. A recent harvest productivity study in Maine identified operational constraints for whole tree harvest systems, but efforts to improve active operations would be expensive and time consuming. A common practice to reduce costs and time consumption is to develop simulation models and implement new ideas within them. We developed a production efficiency model that leverages an agent-based modeling approach. The model is based on the interaction of two common forest machines (grapple skidder and stroke delimber) and incorporates empirical cycle time estimates from research in Maine. Three scenarios have been developed to investigate baseline conditions, and two GPS and GIS aided communication improvements. The goal of this paper is to document a new agent based model and to investigate the effect of hardwood component on machine idle time and productivity. Results showed that system productivity was affected by skidding distance, bunch spacing, and removal intensity. An increase in hardwood component led to a decrease in stroke delimber idle time but did not affect grapple skidder idle time. Further, hardwood component did not affect system productivity, and none of the three single-skidder scenarios tested performed any better than another. We validated the model by conducting a sensitivity analysis to confirm previous research results. The modeled waiting times are well within the range of observed values and therefore suggest that this model is accurate and well calibrated. Our conclusions are that when operating under average harvesting conditions there is no loss in productivity due to a change in hardwood component and that a stroke delimber idle time of 40% or more is unavoidable unless the stroke delimber can work independently. Future applications of this model may target specific production forestry conditions. [ABSTRACT FROM AUTHOR]

Due to the increasing demands of a growing world population, farmers are tasked with increasing productivity without increasing resource use. This can be achieved by deploying resources in a more efficient manner and by ensuring operations will produce maximum results. However when contemplating the problem of scheduling multiple machines, executing multiple operations at multiple fields, it is a cumbersome task to find an optimal solution without the aid of a decision support tool. A field’s readiness, in terms of its trafficability and workability, is an important factor to consider when scheduling operations. Executing operations on an unready field can lead to additional operational costs, such as drying grain harvested at an unready moisture level or removing subsoil compaction caused by machinery driving on unready soils. To ensure a schedule is actionable, the fields’ readiness must be accounted for by the scheduling process, further increasing the complexity of the problem to be solved. In this paper, a novel scheduling algorithm is presented which creates individual machine work plans for multiple machines to execute multiple consecutive operations at multiple fields, accounting for the field’s readiness for the specified operation. Two optimisation algorithms are utilised to find near optimal solutions of predefined scenarios, a standard Tabu Search and a modified Tabu Search, producing optimised work schedules. The functionality of the optimisation algorithms are tested both for a real world scenario and a large set of numerical examples. The results of the optimisation methods are assessed by their computational time and their relative error compared to the optimal solution, where available. The combination of the presented scheduling algorithm and optimisation algorithms could be used either as a stand-alone tool for farm managers and agricultural contractors as part of their logistic planning, or as a component of an overall Farm Management Information System. [ABSTRACT FROM AUTHOR]

• Proposing a robust and effective DSE-YOLO network. • Proposing a detail-semantics extraction module for feature enhancement. • Proposing enhanced loss functions for class imbalance. • Validation of an effective framework for multi-stage strawberry fruits detection. Multi-stage strawberry fruits detection is one of the important clues to estimate crop yields and assist robotic picking in modern agricultural production. However, it is difficult for detecting strawberries due to their small size, foreground-foreground class imbalance, and complex natural environment. Many works focus on how to detect fruits while ignoring multi-stage fruit detecting problems. In this paper, we propose DSE-YOLO (Detail-Semantics Enhancement You Only Look Once) to detect multi-stage strawberries. In DSE-YOLO, DSE (Detail-Semantics Enhancement) module is designed for detecting small fruits and distinguishing different stages of the fruit with higher accuracy, which utilize pointwise convolution and dilated convolution to extract various detail and semantics features in the horizontal and vertical dimensions. Exponentially Enhanced Binary Cross Entropy (EBCE) and Double Enhanced Mean Square Error (DEMSE) loss function are constructed to focus on small fruits, which can deal with foreground-foreground class imbalance problem. Experiments conducted on datasets demonstrate the superiority of DSE-YOLO over state-of-the-arts. The detection results had a mAP value of 86.58% and an F 1 -Score value of 81.59%, which demonstrates the effectiveness of the proposed model. Especially, DSE-YOLO can almost detect every stage of strawberry fruit accurately in the natural scene, which can provide an important theoretical basis and premise for automatic picking and monitoring system. [ABSTRACT FROM AUTHOR]

• Remote sensing technologies has greater application potential in N diagnosis. • Dilution curves are helpful for an in-depth understanding of G × E × M interactions. • The newly developed Bayesian method has the potential to minimize uncertainties. Nitrogen (N) is one of the decisive elements for plant growth, crop biomass accumulation, and yield formation of cereal crops. However, managing N in crop production and comparing N use efficiency are challenging without prior knowledge of in-season crop N status. The development of critical N (N c) dilution curves based on allometry between plant metabolic and structural compartments allows the estimation of crop N nutrition status by determining the N nutrition index (NNI). The purpose of this article is to review the research progress on the development of N c curves in three major cereal crops (wheat, maize, and rice). The focus of this review paper is to compare the N c curves of major cereals developed worldwide and to explain the differences in these N c curve parameters in light of genotype through the management of environmental interactions. Additionally, this review provides an update on the most relevant methods for non-destructive estimation of the NNI of major cereal crops using various remote sensing technologies for large-scale applications. This review also sheds light on the applicability of N c curve-based NNI for in-season crop N diagnosis, crop N requirement, crop grain yield, and quality prediction. Moreover, this review outlines future research directions to expand the impact of this approach on crop production. [ABSTRACT FROM AUTHOR]

• How and which quantitative methodologies based on Operations Research are implemented to solve problems in farms. • Response surface methodology is successfully used as optimization tool in decision making for agricultural production. • The lack of integration between the solution methods, the works only reach a real implementation near 17% of the cases. • The solutions are focused on planning problems in tactical and operational levels in farms. • Integrating lean to mathematical models for agricultural production contribute to the sustainability of food production. Due to the increase of global demand of agriculture and agro-industrial products, agriculture has undergone some changes in the stages of procurement, planning, planting, maintenance, harvesting, transport and distribution. The use of formal mathematical model has been employed to evaluate and implement improvement policies and strategies. In this paper, we present a Mixed Integer Nonlinear Programming Problem model to solve the planning of agricultural production systems in the stages of sowing, crop maintenance and harvesting. The objective function seeks to minimize the cost of resources and wastes generated in operations. The minimization is inspired by lean manufacturing wastes but newly applied in agricultural production. The proposed approach is validated using data for the planning of bananas farmers in the Caribbean region of Colombia. While a deterministic solution is obtained, sensitivity analyses are carried out to evaluate different scenarios for production based on surface response methodology. Model offers a reduction of cost near 59% of cost relationed with wastes of production, moreover, we can find an alternative for scheduling operations in agricultural production based in the most important resources and tasks required along of production in agriculture. [ABSTRACT FROM AUTHOR]

More efficient agricultural machinery is needed as agricultural areas become more limited and energy and labor costs increase. To increase their efficiency, trajectory tracking problem of an autonomous tractor, as an agricultural production machine, has been investigated in this study. As a widely used model-based approach, model predictive control is preferred in this paper to control the yaw dynamics of the tractor which can deal with the constraints on the states and the actuators in a system. The yaw dynamics is identified by using nonlinear least squares frequency domain system identification. The speed is controlled by a proportional–integral–derivative controller and a kinematic trajectory controller is used to calculate the desired speed and the desired yaw rate signals for the subsystems in order to minimize the tracking errors in both the longitudinal and transversal directions. The experimental results show the accuracy and the efficiency of the proposed control scheme in which the euclidean error is below 40 cm for time-based straight line trajectories and 60 cm for time-based curved line trajectories, respectively. [ABSTRACT FROM AUTHOR]

Farm Management Information Systems (FMIS) in agriculture have evolved from simple farm recordkeeping into sophisticated and complex systems to support production management. The purpose of current FMIS is to meet the increased demands to reduce production costs, comply with agricultural standards, and maintain high product quality and safety. This paper presents current advancements in the functionality of academic and commercial FMIS. The study focuses on open-field crop production and centeres on farm managers as the primary users and decision makers. Core system architectures and application domains, adoption and profitability, and FMIS solutions for precision agriculture as the most information-intensive application area were analyzed. Our review of commercial solutions involved the analysis of 141 international software packages, categorized into 11 functions. Cluster analysis was used to group current commercial FMIS as well as examine possible avenues for further development. Academic FMIS involved more sophisticated systems covering compliance to standards applications, automated data capture as well as interoperability between different software packages. Conversely, commercial FMIS applications targeted everyday farm office tasks related to budgeting and finance, such as recordkeeping, machinery management, and documentation, with emerging trends showing new functions related to traceability, quality assurance and sales. [ABSTRACT FROM AUTHOR]

• A dual-stream hierarchical bilinear pooling model was proposed. • Models using the approach of different bilinear pooling were investigated. • Optimizing the weights of two tasks using homoscedastic uncertainty. Plant diseases have an important impact on agricultural production and economic efficiency. Timely detection of crop diseases and accurate determination of diseases are important for protecting crop safety and controlling the spread of diseases. Although current plant disease recognition research using deep learning has yielded advanced results, it is difficult to produce good results when applied to the actual plant growth environment. The background of plant disease images acquired under field conditions is complex, and the same crop disease often varies widely due to many uncertainties such as pose, shading, and other factors. Models using fine-grained image recognition methods can extract discriminative fine-grained features, thus enhancing the representation capability of the model. Therefore, in this paper, a dual-stream hierarchical bilinear pooling model is proposed for the multi-task classification of crops and diseases under field conditions with a method for the independent identification of plants and diseases. After optimizing multi-task learning using homoscedastic uncertainty, the plant and disease accuracies of the dataset obtained under field conditions were 84.71% and 75.06%, respectively. [ABSTRACT FROM AUTHOR]

• The Burgers model was deduced to describe the constitutive model of corn grain. • The "corn-mechanical" interaction model and "corn-corn" model based on the Hertz contact method were established. • The construction methods of the corn ear through particle splicing method and particle filling method were proposed. • The particle filling method has been proved to be more accurate in explaining the mechanism of corn threshing. The constitutive model of corn has an intense influence on the mechanical structure design and machine performance. In this paper, a creep property formula of corn-based on the Burgers model was proposed, which described the change process of the stress–strain trait of corn grain with time. Then, the mathematical model of the "corn-mechanical" interaction system was then established based on the Hertz-Mindlin no-slip model. The kinematic and dynamic formulas of corn ear under the action of threshers were deduced, and the "corn-corn" contact characteristics based on Hertz-Mindlin with bonding model were analyzed. Moreover, a new discrete element model construction method of the corn ear was proposed through particle filling and splicing methods. The simulation experiment was carried out in EDEM to verify the constitutive model and discrete element model of corn ear. The results showed that the particle filling model could better explain the mechanism of corn ear threshing and obtain the force of each corn grain during threshing. The average velocity of corn ear was 55.2 m/s, the maximum value was 216.7 m/s, and the minimum value was 16.2 m/s, while the average normal force was 201.7 N, the maximum force was 342.0 N, and the minimum force was 104.2 N respectively, it verified the feasibility and accuracy of models. Finally, the field experiment was carried out to verify the simulation results, and the corn grain broken rate was 2.16%. These results fill the blank of fundamental theoretical research on the design of corn ear threshing devices and provide a reference for the optimal design of other crop production equipment. [ABSTRACT FROM AUTHOR]

• Empirical density functions are linked to other datasets for crop-yield modeling. • Artificial covariates can represent complex phenomena in predictive models. • The proposed approach provides measures of uncertainty along with yield predictions. Machine learning methods are increasingly used in analyzing remotely sensed data and studying different aspects of agricultural production. In particular, several of these flexible models are widely adopted to predict regional crop yield during or after the growing season. However, most existing models cannot be applied when dealing with functional covariates. In this paper, an approach based on multidimensional scaling is proposed to generate a set of artificial covariates from empirical density functions of different phenomena captured within specific administrative boundaries through satellites. In contrast to traditional aggregation methods, this approach is designed to reduce the loss of information associated with the use of summary statistics as covariates. The proposed methodology is applied to NASA remote sensing data, combined with information from surveys and USDA's end-of-season county estimates, to study the prediction accuracy of different crop-yield models for three major crops in North Dakota. [ABSTRACT FROM AUTHOR]

• A concise and accurate FoveaMask model is proposed for segmentation of green fruits. • A position attention module(PAM) is added into mask branch for improving robustness. • FoveaMask eliminates dependence on anchors. • It can be referred and even migrated directly to other fruit segmentation. In the process of agricultural automation production, efficient and accurate segmentation of target fruit is the basis and guarantee for numerous applications including crop growth monitoring, yield prognosis, and machine picking. Green fruit is apt to affect by complicated scenes such as occlusions and overlaps, as well as the homo-chromatic background, which brings great challenges to the recognition and segmentation. In this paper, we propose a novel methodology named FoveaMask for improving the robustness and generalization of green fruit segmentation framework. Features of input images are firstly extracted by ResNet and fused by FPN. The classification and bounding-box regression of each spatial position on the feature maps are then carried out directly by the way of full convolution. RoI Align layer is applied to fix feature region of proposals to the same size yet preserve exact spatial locations. Finally, instance-level fruit segmentation is realized by pixel-level classification on each proposal using embedded mask branches. The whole network architecture of the new model does not involve the related design and operation of anchor, which greatly improves the generalization ability for different shape fruits, alleviates the computing and storage resources, and balances the contradiction between accuracy and efficiency simultaneously. Additionally, a position attention module(PAM) is introduced into the embedding mask branch to aggregate the effective information pixels, which can improve the robustness of the segmentation model in the actual complex environment of the orchard. We test the model on green apple and immature persimmon data sets and the experimental results show that FoveaMask performs best in terms of both recognition accuracy and model complexity compared with other 11 different types of detection and segmentation models. [ABSTRACT FROM AUTHOR]

• Identification of optimal decisions for production planning and machinery scheduling to maximize the farmers' NPV. • Optimal rotation of perennial and non-perennial crops satisfying the irrigation water requirements and availability. • Optimal routes for machinery between the parcels determined based on ArcGIS Pro tools. • Two case studies for low-water-use crops (guar and guayule) and other competing crops. Scarce water resources have made production planning a key management decision in the agriculture sector, especially in arid and semi-arid regions. To address this issue, farmers can gradually adopt low-water-use crops, such as guar and guayule, which have great potential for the agricultural economy of the Southwestern U.S. The farmers' profits can be further increased by reducing machinery transportation costs through optimized scheduling. For such purpose, this paper introduces a farm production planning and machinery scheduling optimization model for perennial and non-perennial crops that maximizes the farmers' net present value. The optimization model allows for multi-crop production planning, multi-machinery scheduling, and crop rotation. To address the water scarcity problem in arid and semi-arid regions, perennial and non-perennial low-water-use crops can be considered in the crop rotation. Based on the Geographic Information System, the optimal scheduling routes are determined and the irrigation water requirements are analyzed. The resulting model yields a complex large-scale integer linear optimization problem. The model is assessed on two case studies, one for the guar crop (non-perennial) considering parcels located at the Cochran and Hockley counties in Texas and another case study for the guayule crop (perennial) considering parcels located in Pinal County in Arizona. [ABSTRACT FROM AUTHOR]

• Temporal estimation of biophysical parameters is vital for precision agriculture. • We estimated biophysical parameters of vegetable crops for different N levels. • LiDAR data is useful for plant level estimation of height, crown area and biomass. • LiDAR based estimates of plant height and crown area are good predictors of biomass. • Estimated plant parameters correlate strongly with reference data across N levels. Estimation of biophysical parameters at various crop growth stages is vital for precision agricultural crop production. Spatial delineation of crops' responses to various levels of nutrients helps optimise resources and reduce nutrient leaching. This paper explores the potential of 3D terrestrial laser scanning (TLS) for the estimation of plant height, crown area, and biomass of vegetable crops at various N levels. Experimental setup of growing three vegetable crops: tomato (Solanumlycopersicum L.), eggplant (Solanummelongena L.) and cabbage (Brassica oleracea L.) with three levels of N fertilization was laid out at the University of Agricultural Sciences, Bengaluru, India in 2017. LiDAR point clouds using a terrestrial laser scanner were collected at different growth stages. A methodology which included, among other processing steps, adaptive spatial filtering, canopy height modelling, watershed segmentation, and support vector regression has been adapted for the estimation of plant height, crown area, and biomass. Validation with ground measurements show high prediction accuracies for plant height (lowest coefficient of determination (R2), 0.96; highest symmetric mean absolute percentage error (SMAPE) of 3.18), and crown area (lowest R2, 0.82; highest SMAPE, 8.82) for all the three crops across growth stages. The combined use of plant height and the crown area has enabled accurate and consistent estimation of biomass (lowest R2, 0.92; highest SMAPE, 7.53) throughout the growing season. However, the mapping of a specific range of biomass to a specific N level is ambiguous due to wider variations in the crop growth due to rainfall, and wind interferences. [ABSTRACT FROM AUTHOR]

• City-level agricultural production efficiency is mapped by Super-Efficiency Data Envelopment Analysis model. • The agricultural production efficiency is featured as spatial agglomerations. • The OLS and GWR models are used to explore the spatial heterogeneity and driving mechanisms of agricultural production efficiency. • Series of policy recommendations are provided. The level of agricultural production efficiency (APE) affects farmers' livelihood decisions and the transformation and reconstruction of the rural territorial system. This paper constructs an evaluation index system of China's APE from the perspective of input–output. The spatial autocorrelation model and econometric model are used to calculate the spatio-temporal evolution characteristics and influencing factors of China's APE in 1990 and 2017. The results show that China's APE level is relatively low, and there are significant spatial differences. The extremely high efficiency regions are distributed in the eastern plains of China; the regions with extremely low efficiency are mainly distributed in the southwest and the western regions of China. The results of the spatial correlation effect show that the H-H agglomeration regions of APE are usually distributed in the eastern coastal areas and northeastern areas, while the L-L agglomeration regions are mainly distributed in the central mountainous areas and the border areas of western Yunnan. From 1990 to 2017, the influencing factors of APE have gradually changed, and showing a more complex trend, mainly including farmland management scale , multiple crop index, rural per capita net income, population density, and precipitation per unit area. Finally, the corresponding policy and suggestions are provided. [ABSTRACT FROM AUTHOR]

• A dense CNN model is presented for diseases recognition and classification in corn leaf. • A comprehensive step-by-step discussion of proposed model is given. • Rigorous experimental evaluation is presented. • Performance comparison is done with state-of-the-art models. An optimized dense convolutional neural network (CNN) architecture (DenseNet) for corn leaf disease recognition and classification is proposed in this paper. Corn is one of the most cultivated grain throughout the world. Corn crops are highly susceptible to certain leaf diseases such as corn common rust, corn gray leaf spot, and northern corn leaf blight are very common. Symptoms of these leaf diseases are not differentiable in their nascent stages. Hence, the current research presents a solution through deep learning so that crop health can be monitored and, it will lead to an increase in the quantity as well as the quality of crop production. The proposed optimized DenseNet model has achieved an accuracy of 98.06%. Besides, it uses significantly lesser parameters as compared to the various existing CNN such as EfficientNet, VGG19Net, NASNet, and Xception Net. The performance of the optimized DenseNet model has been contrasted with the current CNN architectures by considering two (time and accuracy) quality measures. This study indicates that the performance of the optimized DenseNet model is close to that of the established CNN architectures with far fewer parameters and computation time. [ABSTRACT FROM AUTHOR]

• An end-to-end two-stage framework is proposed for multi-classes agricultural pest detection. • A feature fusion module is introduced to enhance features representation of pest with small size. • We propose a new region proposal generation network without excessive hyper-parameters. • A matching algorithm based on the effective region of receptive field is used during training. The frequent outbreaks of agricultural pests have resulted in the reduction of crop production and seriously restricted agricultural production. And many kinds of agricultural pests bring challenges to the accurate identification of agricultural pests for agricultural workers. Currently, the traditional methods of agricultural pest detection cannot satisfy the needs of agricultural production because of low efficiency and accuracy. In this paper, we put forward an anchor-free region convolutional neural network (AF-RCNN) for precision recognition and classification of 24-classes pests. First, a feature fusion module is designed to extract effective feature information of agricultural pests, especially small pests. Then, we propose an anchor-free region proposal network (AFRPN) that is used for getting high-quality object proposals as possible pest positions based on the fusion feature maps. Finally, our anchor-free region convolutional neural network (AF-RCNN) is employed to detect 24-classes pest via an end-to-end way by merging our AFRPN with Fast R-CNN into a single network. We evaluate the performance of our method on the pest dataset including 20 k images and 24 classes. Experimental results demonstrate that our method is able to obtain 56.4% mAP and 85.1% mRecall on 24-classes pest dataset, 7.5% and 15.3% higher than Faster R-CNN, and 39.4% and 56.5% higher than YOLO detector. The running time could achieve 0.07 s per image, meeting the real-time detection. The proposed method is effective and applicable for accurate and real-time intelligent pest detection. [ABSTRACT FROM AUTHOR]