Search

Pesticide spray is a widely used chemical control method to minimize biological disasters in the agriculture industry. It is important to evaluate the efficacy and the quality of a pesticide sprayer, however, it cannot be conveniently achieved due to the lack of accessibility to the sprayed droplets and the intensity of the associated work. This paper proposes a novel method, based on an image processing-based approach, to assess the spray quality. The proposed method uses a combination of algorithms and criteria functions; such as, the maximum between-cluster variance algorithm, area threshold criteria, roundness factor, mathematical morphology operations, and the optimized watershed algorithm, to segment and assess the dark blue adhesive droplet images on a water-sensitive paper, placed in the crop field. In this work, three kinds of evaluation experiments are considered: (i) the manual analysis via droplet counting and processing, (ii) the automatic analysis by the commercially available droplet analyzer by Shenzhen DJI Co. Ltd., and (iii) the novel image processing based method introduced in this paper. The experimental results show a better consistency between the introduced novel method and the manual method. The former, however, provides a convenient and rapid way to assess the spray quality. It comes with an assessment algorithm along with an embedded device that provides a hardware view of the spray quality.

Deep learning (DL) based instance segmentation has attracted a growing research interest in the scientific community to tackle precision agriculture problems over the past few years. However, accurate crop detection and localization in complex environments pose a significant challenge. Instance segmentation is considered as a promising DL technique that expands on object detection to perform pixel-wise image instance segmentation and address pattern recognition problems efficiently. In this review, we identify 77 relevant studies on DL-based instance segmentation implementations in agriculture and thoroughly investigate them from the following perspectives: i) the specific architecture employed; ii) the data type and availability, the data annotation process and the data pre-processing techniques; iii) the performance metrics used; and iv) hardware, inference time and GPU requirements. Our findings indicate that crop detection (48 papers) constitutes a fundamental task in a DL-based instance segmentation pipeline to enable crop growth monitoring (19 papers) and plant health analysis (10 papers). Among them, 6 papers reported robotic manipulation and other related automation tasks. Based on our findings we can conclude that there is a significant trend towards two-stage DL-based instance segmentation models i.e., Mask R-CNN baseline and customized architectures (69 papers). Limitations and challenges, such as availability of benchmark crop datasets, open-source codes for semi-automatic annotation tools, technical requirements and opportunities for future research are discussed.

Smart farming practices offer decision-making support as farmers navigate economic, social, and environmental challenges. However, smart farming adoption remains low in many farming contexts due to the perceived cost and skills required, and hesitancies about sharing agricultural data. Numerous studies have reviewed the factors that influence smart farming adoption within different agricultural scenarios, but to the best our knowledge, none have specifically reviewed the motivators and obstacles of agri-data sharing within smart farming. The objective of this research was to identify and classify the most prominent drivers and barriers for sharing agri-data across stakeholders, by examining the existing literature. A Systematic Literature Review was conducted using the PRISMA 2020 methodology. The query initially identified 491 papers from Scopus and Web of Science, and after screening the final number of papers for assessment was 59. Factors affecting the willingness and capability to engage in agricultural data sharing were categorised in socio-economic, systemic, technical, and legal categories. The most prominent drivers for agri-data sharing were systemic and technical, which were discussed in 58 % and 57 % of the papers, respectively. Technical factors were the most prevalent barriers, discussed in 68 % of the literature. Perceived knowledge gain leading to improved decision-making, collaboration across the agri-value chain, improved technologies, and clarity around data sovereignty were key enablers identified for sharing data. Lack of knowledge about the purpose and benefit of sharing data, mistrust of “who will benefit from my data”, data privacy and security, and lack clarity on data ownership and rights of use were the prominent concerns. The findings from this paper will help inform on-the-ground social science research across the EU focused on feasible options for promoting the benefits of agri-data sharing.

The article provides a comprehensive review of the use of the Internet of Things (IoT) in agriculture, along with its advantages and disadvantages. However, it's important to recognize that IoT holds immense potential for generating new ideas that could drive innovations in modern agriculture and address several challenges faced by farmers today. Applications such as smart irrigation, precision farming, crop and soil tracking, smart greenhouses, supply chain management, livestock monitoring, agricultural drones, pest and disease prevention, and farm machinery are among the areas considered for IoT implementation in agriculture by this paper. These innovative solutions have the potential to revolutionize farming practices, improve efficiency, reduce resource wastage, and ultimately enhance agricultural productivity and sustainability. The analysis examines each application in terms of its utility and outlines measures necessary to enhance its effectiveness. Key considerations include addressing connectivity issues, managing costs, ensuring data security and privacy, scaling solutions appropriately, effectively managing data, and promoting awareness and adoption of IoT tools. Despite these challenges, IoT offers numerous benefits to the agricultural sector. The paper underscores the importance of collaboration among farmers, IoT technology companies, academia, and policymakers to address these issues and fully harness the potential of IoT. To achieve this goal, ongoing research, development, and acceptance of IoT-driven solutions are essential to sustain agriculture as a viable option amidst emerging challenges such as climate change and resource scarcity.

Microgreens are the first leafy seedlings of edible plants. Microgreen farming is yet to be automated; the main challenge for automation is the lack of a sensory mechanism to detect and quantify microgreen phenotypes. This paper presents a novel automated microgreen phenotyping method targeting yield estimation. The paper demonstrates that phenotyping can be effectively performed using a consumer-grade RGB-D camera. First, the depth and RGB images are captured. Thereafter, the plant segments are filtered and the canopy is identified. Using image processing, the canopy height and density are calculated. Both yield prediction regression analysis and a TensorFlow learning algorithm are evaluated to estimate the yield as a function of height and canopy density. The authors believe the algorithm discussed in this paper is the first phenotyping algorithm combining RGB and depth data for microgreen yield estimation.

Despite the agribusiness industry's potential, it is still characterised by a high level of conventional means of production, denying it numerous opportunities along the value chain in Africa. Meanwhile, adopting modern means of production using digital innovations can boost farmers' efficiency. Given the relevance of digital innovations within the agribusiness industry, the current paper gives an overview of digital innovations in the agribusiness industry, particularly in Africa, by synthesising existing literature on digital innovations to consolidate scattered ideas and give recommendations for practice and future studies. The PRISMA technique was adopted to synthesise 47 existing relevant articles from the Scopus database. Bibliometric tools such as VOSviewer and R-Package-Bibliometrix were used to analyse the data. From the results, most papers on African digital innovations emanated from a journal with a broader scope—Sustainability Journal (Switzerland). Most of the funding for digital innovation studies in agribusiness was sponsored by foreign donors outside the African Continent, with the Consortium of International Agricultural Research Centers in France sponsoring the most research. Based on the publication synthesis and the trend of publications, five themes, namely, digital innovations in the agricultural value chain, training and skills development through digital innovations, digital finance innovations in agribusiness, precision digital innovations: agribusiness survival, and drivers of digital innovation adoption in agribusiness were identified. From the results, digital agribusiness innovations in marketing, production, and finance have been profound in Africa, though they are skewed towards a few South, West, and North African countries. Meanwhile, the drivers and challenges of digital innovations in Africa have been mainly social, economic, political, and institutional factors. Given the results, it is recommended that, since most of the studies were primarily sponsored by foreign governments, thus betraying the commitment of the African continent's readiness to transform the agribusiness sector via innovation and technology, African Governments, first, and second, non-governmental organisations and other agricultural donors should prioritise the digitalisation of the agriculture sector by integrating them into economic development plans and making relevant resources available to all actors from upstream to downstream.

In the context of increasing global population and climate change, modern agriculture must enhance production efficiency. Vegetables production is crucial for human nutrition and has a significant environmental impact. To address this challenge, the agricultural sector needs to modernize and utilize advanced technologies such as drones to increase productivity, improve quality, and reduce resource consumption. These devices, known as Unmanned Aerial Vehicles (UAV), with their agility and versatility play a crucial role in monitoring and spraying operations. They significantly contribute to enhancing the efficacy of precision farming.The aim of this review is to examine the critical role of drones as innovative tools to enhance management and yield of vegetable crops cultivation. This review was carried out using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) framework and involved the analysis of a wide range of research published from 2018 to 2023. According to the phases of Identification, Screening, and Eligibility, 132 papers were selected and analysed. These papers were categorized based on the types of drone applications in vegetable crop production, providing an overview of how these tools fit into the field of Precision Farming. Technological developments of these tools and data processing methods were then explored, examining the contributions of Machine and Deep Learning and Artificial Intelligence. Final considerations were presented regarding practical implementation and future technical and scientific challenges to fully harness the potential of drones in precision agriculture and vegetable crop production. The review pointed out the significance of drone applications in vegetable crops and the immense potential of these tools in enhancing cultivation efficiency. Drone utilization enables the reduction of input quantities such as herbicides, fertilizers, pesticides, and water but also the prevention of damages through early diagnosis of various stress types. These input savings can yield environmental benefits, positioning these technologies as potential solutions for the environmental sustainability of vegetable crops.

This paper presents the subsystem descriptions and testing of a small robotic platform intended to harvest strawberries that are grown on elevated beds in open field conditions. Agricultural robots can assist in the optimization of farm resources and help solve issues related to increasing farm costs and labor shortages. Monolithic, large size harvesters are already in development; however, they are susceptible to single-point-of-failure and lack the flexibility to adapt to varying field conditions and farm sizes. The proposed robotic harvester covers one row at a time and features a Delta arm configuration manipulator with a five-finger structure as the end-effector to individually pick strawberries. A deep neural network-based vision subsystem using a YOLOv4 model was adopted, which was configured to detect small objects and to locate and classify strawberries into five stages of maturity. During experiments on a commercial farm, the proposed platform, including vision, manipulation, and overbed navigation and control subsystems, achieved an overall success rate of 71.7% for five environment scenarios, with a minimum of 37.5% (the most complicated scenario) and a maximum of 94.0% (the easiest scenario). The average harvesting speed of the system was 7.5 s per strawberry.

Soluble solids content (SSC) and pH of red globe grapes are crucial measures of quality. In this paper, we used hyperspectral imaging technology to achieve nondestructive detection and distribution visualization of SSC and pH of red globe grapes. First, the hyperspectral images of samples were collected. Then, CARS, SPA, GA, IRIV were used to extract feature variables from raw spectral (RAW) information. The PLSR prediction models of samples were developed. By comparing the different prediction models, RAW-IRIV-PLSR was selected as the optimal model. Finally, the SSC and pH of the samples were calculated to obtain a grayscale image and perform a pseudo-color transformation to visualize the distribution of SSC and pH. By studying the classification of the maturity of samples, it was concluded that the best discriminant classification model of maturity was RAW-IRIV-ELM. Hyperspectral also provided a new method for maturity stage classification of red globe grapes.

This paper presents an autonomous multi-agent drone system utilized for an artificial pollination task to address the declining bee population and its impact on natural pollination. The main goal is to visit all the targets (flowers) by drones. The system receives the position of the targets located on a vertical fruiting wall from a single sensor mounted on an unmanned ground vehicle (UGV). Then, it determines the sequence of target allocation to minimize the task time and to minimize the accumulated distance each drone travels. The limited payload capacity of each drone and the need to refill stock carried by the UGV are also considered. After allocating the targets, a safe trajectory is generated for each drone to avoid collision with other drones, considering uncertainty in the drones’ positions. When a drone reaches a target, the system allocates a new target, and the cycle repeats until the task is completed. The system is reactive to the evolution of the task and adjusts accordingly. An additional tested factor is the impact of the positioning error on the task time. The system was evaluated through dynamic simulations using real datasets of peach and pear orchards and finally in a laboratory proof-of-concept experiment using a micro-drone system.

Software architecture forms the cornerstone for achieving and ensuring various software quality attributes. It encompasses the collected requirements of the product, serving as a blueprint that delineates quality features for all project stakeholders, along with methods for measurement and control. Despite the significant increase in IoT-based agricultural systems, there is a dearth of studies on the quality elements of their software architecture. To address this need, this study offers an overview of components and services tailored to address specific quality attributes pertinent to agriculture systems. It identifies, investigates, and presents quality attributes influencing the design of software architecture for IoT-based agriculture systems. This paper identified and discussed several quality attributes, including performance, scalability, flexibility, interoperability, productivity, extensibility, and security, and mapped them to corresponding components of the IoT-based agriculture software architecture. Also, several issues were identified and discussed for the software architecture quality of IoT-based agriculture systems, such as real-time processing and interoperability due to the various devices and protocols utilized in these systems. The findings of this study offer valuable insights for developing, executing, and refining IoT-based agricultural systems to fulfill the changing requirements of the agriculture industry.

The research results presented in this paper were obtained as part of the D2KAB project (Data to Knowledge in Agriculture and Biodiversity) which aims to develop semantic web-based tools to describe and make agronomical data actionable and accessible following the FAIR principles. We focus on constructing domain-specific Knowledge Graphs (KGs) from textual data sources, using Natural Language Processing (NLP) techniques to extract and structure relevant entities. Our approach is based on the formalization of a semantic data model using common linked open vocabularies such as the Web Annotation Ontology (OA) and the Provenance Ontology (PROV). The model was developed by formulating motivating scenarios and competency questions from domain experts. This model has been used to construct three different KGs from three distinct corpora: PubMed scientific publications on wheat and rice genetics and phenotyping, and French agricultural alert bulletins. The named entities to be recognized include genes, phenotypes, traits, genetic markers, taxa and phenological stages normalized using semantic resources such as the Wheat Trait and Phenotype Ontology (WTO), the French Crop Usage (FCU) thesaurus and the Plant Phenological Description Ontology (PPDO). Named entities were extracted using different NLP approaches and tools. The relevance of the semantic model was validated by implementing experts questions as SPARQL queries to be answered on the constructed RDF knowledge graphs. Our work demonstrates how domain-specific vocabularies and systematic querying of KGs can reveal hidden interactions and support agronomists in navigating vast amounts of data. The resources and transformation pipelines developed are publicly available in Git repositories.

This paper introduces a precision agriculture application aimed at mitigating the excessive utilization of agricultural chemicals, including pesticides and fungicides during crop spraying. The prevailing spraying techniques face two principle challenges: first, the indiscriminate dispensation of chemicals irrespective of plant size and requirements and second, the farmer's exposure to health hazards. To tackle these issues, a detection and segmentation model employing both YOLOv5 and YOLOv6 architectures is proposed and a comparative assessment of their accuracies within the same model category is conducted. The training dataset originates from a subset of the TobSet dataset, while the evaluation of the trained models is executed using publicly accessible aerial videos/images from available repository. The best detection accuracy achieved for the tobacco plant model size is observed with YOLOv6s and the YOLOv5-segmentation model, yielding accuracies of 95% and 94.8%, respectively. Additional performance metrics such as precision, recall, area under the PR-curve, inference time, and NMS per image are also compared between the two models. The YOLOv5-segmentation model excels by outperforming the YOLOv6s model in precision, recall score, and area under the PR-curve whereas slightly extended inference time and NMS per image duration are noted for the YOLOv5-segmentation model and the speed performance is comparable for the two models. Subsequently, the evaluation of these two models is conducted on the drone videos, which were recorded during drone traversal at a speed of 2 km/hr. The results demonstrate superiority of YOLOv5-segmentation model over the YOLOv6s model, with detection accuracies of 98.1% and 97.3%, respectively. These findings indicate the potential of integrating YOLOv5 segmentation models in precision spraying applications and contribute in improving the overall agricultural practices.

Assessing the phenotype of cattle through human visual inspection is a very common and important practice in precision cattle breeding. This paper presents the results of a correlation analysis between scores produced by humans for Nelore cattle and a variety of measurements that can be derived from images or other instruments. It also presents a study using the k-means algorithm to generate new ways of clustering a batch of cattle using the measurements that most correlate with the animal's body weight and visual scores.

The automation of all-terrain vehicles (ATVs) through the integration of advanced technologies such as machine learning (ML) and artificial intelligence (AI) vision has significantly changed precision agriculture. This paper aims to analyse and develop trends to provide comprehensive knowledge of the current state of ATV-based precision agriculture and the future possibilities of ML and AI. A bibliometric analysis was conducted through network diagram with keywords taken from previous publications in the domain. This review comprehensively analyses the potential of machine learning and artificial intelligence in transforming farming operations through the automation of tasks and the deployment of all-terrain vehicles. The research extensively analyses how machine learning methods have influenced several aspects of agricultural activities, such as planting, harvesting, spraying, weeding, crop monitoring, and others. AI vision systems are being researched for their ability to enhance precise and prompt decision-making in ATV-driven agricultural automation. These technologies have been thoroughly tested to show how they can improve crop yield (15-20%), reduce overall investment (25-30%), and make farming more efficient (20-25%). Examples include machine learning-based seeding accuracy, AI-enabled crop health monitoring, and the use of AI vision for accurate pesticide application. The assessment examines challenges such as data privacy problems and scalability constraints, along with potential advancements and future prospects in the field. This will assist researchers and practitioners in making well-informed judgments regarding farming practices that are efficient, sustainable, and technologically robust.

Plant diseases can significantly reduce crop yield and product quality. Visual inspections of plants by human observers for disease identification are time-consuming, costly, and prone to error. Advances in artificial intelligence (AI) have created opportunities for the rapid diagnosis and non-destructive classification of plant pathogens. Several machine vision techniques have been developed to identify and classify plant diseases automatically based on the morphology of specific symptoms. The use of deep learning models has achieved acceptable disease classification results, but they require large datasets for training, which can be labor-intensive, time-consuming, and computationally costly This problem can be solved, to a point, by using data augmentation techniques and generative AI in order to increase the size of the datasets. Furthermore, a combination of deep feature extraction and classification by machine learning was used for accurate disease detection and classification. In some cases, traditional base classifiers trained with small datasets including basic shape, color, and texture features can be feasible for the efficient identification of plant diseases. The performance of such classifiers depends primarily on the features extracted from images; therefore, feature extraction plays a vital role in identifying diseases. Feature engineering, a process to identify the most relevant variables from raw data in order to develop an efficient predictive model, is explored in this paper.

The transition from traditional agriculture to Digital Agriculture involves various elements that make it challenging. Given the different contexts and particularities, this paper aims to identify and analyze the primary factors and barriers to adopting Digital Agriculture, to promote an understanding of this transition. By conducting a bibliometric assessment and analyzing case studies, the results show that the economic condition, availability of technological infrastructure, technical knowledge, age of farmers, type of organization, reliability of technology, and concerns about security and privacy are important elements in adopting Digital Agriculture. The study also reveals a limitation linked to the data sources used in their preparation by analyzing the case studies. This finding reveals the existence of a gap in the literature concerning the scarcity of indicators capable of measuring the adoption of digital agriculture while at the same time providing a perspective devoid of producer bias. Furthermore, by considering the insights provided by the identification and analysis of those factors and barriers, policymakers can tailor policies to address specific challenges and promote the widespread adoption of digital technologies in agriculture.

This research explores the predictive capabilities of random forests algorithm on datasets coming from standard experiments on crop management systems in soybeans. This is a secondary analysis of a dataset from a project evaluating the relationship of cover crop systems to soybean yield prediction. The purpose of this paper is to compare a random forest algorithm to standard statistical techniques such as linear regression on a clean information rich agronomic experiment. The main findings include an estimate of the hyperparameters for optimal predictions using random forests, a threshold for data for optimal results and a general description of comparison methodologies for AI based techniques.

This paper presents a low-cost (locally-assembled) drone system with a sprayer arm and a remote control for monitoring and precisely applying pesticides to coconut trees infested by the Rhinoceros Beetle. A modified RGB IP camera continuously monitors potential coconut beetle infestation points, capturing real-time spray images, and displaying the results through a smartphone interface. The system provides real-time monitoring, allowing immediate remote control of the spray operation from the ground upon detecting beetle infestations. The drone-mounted sprayer arm is controlled in response to the ground-based signals, to accurately extend or contract while effectively dispensing spray chemicals onto the target areas at the upper canopy of coconut trees. In our tests using Nam-Hom coconut trees, which averaged a height of 10.4 m, the system operated at a speed of 1.25 km/h, achieving a spraying flow rate of 0.333 l/h and a working capacity of 0.352 ha/h. The controlled spraying setup emits a solid jet, ensuring precision and effective atomization of chemicals within a distance range of 1.0 to 4.0 m from the target. The spray quality attributes (droplet size, pattern, and coverage) were not significantly affected by the wind speed. The system could potentially reduce labor costs, chemical usage, and health risks associated with pesticide application.

This comprehensive review delved into the economic and environmental benefits of Digital Agricultural Technologies (DATs) in crop production, synthesising data from 136 peer-reviewed papers and 28 documents with empirical data from relevant EU projects. This analysis highlighted the substantial contribution of DATs across five key categories: Recording and Mapping Technologies (RMT), Guidance and Controlled Traffic Farming (CTF) Technologies, Variable Rate Technologies (VRT), Robotic Systems or Smart Machines (RSSM), and Farm Management Information Systems (FMIS). Specifically, it provided an overview of the various benefits that these technologies can deliver with the most significant ones revealing reductions of up to 80 % in fertiliser usage with RMT and CTF applications, while VRT demonstrated a 60 % decrease in fertiliser usage and up to 80 % reduction in pesticide use. VRT also showed an increase in yield by 62 %. RSSM was able to reduce labour by 97 % and diesel consumption by 50 %. FMIS improved yield by 10 % to 15 %, facilitating simultaneous reductions in labour and input costs, illustrating the critical role of integrated digital solutions in enhancing agricultural efficiency and sustainability. From an environmental point of view, VRT has emerged as a major factor in environmental sustainability, demonstrating water savings of 20 % to 50 % in vineyards and pear orchards and a significant reduction in greenhouse gas emissions. These findings highlighted the significant benefits of DATs on enhancing productivity and promoting environmental sustainability. They provided a compelling case for further investment and research in DATs through quantifiable benefits in crop production.

Combine harvester is one of the most crucial pieces of harvesting equipment in modern agriculture. Farmers and entrepreneurs employing a combine harvester need to focus on two pivotal factors: its profitability and optimal replacement time. Proper performance in this scenario can lead to prompt, and high-quality farm operations, enhancing income and simultaneously reducing product expenses significantly. This paper proposes a web-based application that would allow estimating the economic life and feasibility of combine harvesters in Bangladesh. The minimum cost value determination process has been used to estimate economic life by applying the double declining depreciation method. A high-level Python web framework has been used for the design and development of the software. The combine harvester exhibits a consistent economic life of 10 years for 300 to 450 hrs of annual use, with an increase in average yearly use from 300 to 450 hrs. As average yearly usage increases, the combine harvester's economic life declines drastically after 500 hrs of operation. Comparison with manual estimation shows that economic feasibility estimated by the web-based app provides more relevant data. The web-based app provides a useful tool for farmers, and entrepreneurs in Bangladesh to estimate the economic life and feasibility of combine harvesters, enabling better decision-making and improved managerial policies.

This paper presents the design and development of an artificial incubator for poultry eggs. The incubator is designed to maintain the optimal temperature and humidity conditions for embryo development, and it is also equipped with an egg-turning mechanism. It was designed using Solid Edge software, and it was constructed using plywood, ceramic bar heaters, electric fans, ballcock valves, egg turning steel mechanism, and a CO2 meter. The temperature and humidity conditions inside the incubator were maintained within the desired ranges, and the egg-turning mechanism worked effectively. The results of this study show that this artificial incubator could be a reliable and effective tool for hatching poultry eggs. The incubator is easy to use and maintain, and it is affordable for local poultry farmers. Future studies should focus on determining the conditions under which the incubator will yield the best results in terms of hatching efficiency.

The adoption of Fourth Industrial Revolution (4IR) technologies has revolutionized agricultural practices worldwide. However, their application in the context of sub-Saharan Africa remains a critical challenge. This study presents a systematic review that investigates the potential of 4IR technologies in smart irrigation. Sub-Saharan Africa faces multiple agricultural challenges, exacerbated by climate change, water scarcity, and inefficient irrigation practices. The need for sustainable, water-efficient, and data-driven irrigation systems is urgent to ensure food security and economic development in the region. This study addresses a crucial knowledge gap by assessing the constraints, opportunities, and prospects of implementing 4IR technologies for smart irrigation in sub-Saharan Africa. A systematic review methodology was employed, utilizing reputable databases including Web of Science, Google Scholar, Science Direct, Institute of Electrical and Electronics Engineers (IEEE) Xplore, and PubMed. A comprehensive search strategy was designed to identify peer-reviewed articles, conference papers, and reports related to the application of Internet of Things, Artificial Intelligence, Unmanned Aerial Vehicles, Big Data, and Blockchain technology in smart irrigation. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach was used to arrive at 95 articles included in this research. The review reveals promising trends in the integration of 4IR technologies for smart irrigation. 4IR technologies including Artificial Intelligence, Internet of Things, Big Data technology, Blockchain and drones are widely used in smart irrigation. These technologies have facilitated real time monitoring of soil moisture and weather conditions, precision irrigation scheduling, water allocation optimization, rapid data collection, insights in real time crop water use and transparency and trust in agricultural water management among others. While the potential of 4IR technologies is evident, challenges including limited infrastructure, access to technology, and technical expertise pose significant barriers for the adoption in sub-Saharan Africa. Additionally, the high initial costs associated with these technologies can impede widespread adoption. The study highlights the potential to leverage existing mobile phone penetration for IoT data collection, collaborative partnerships, and innovative financing models to overcome financial constraints. Capacity-building initiatives and knowledge transfer can empower local communities to embrace these technologies. The future of smart irrigation in sub-Saharan Africa relies on policy support, investments, and technology diffusion strategies. It is imperative to create an enabling environment that fosters innovation and addresses the unique needs of the region. Collaboration between governments, academia, industry, and international organizations can catalyze the transformative impact of 4IR technologies, ultimately enhancing irrigation sustainability and food security.

In general, agriculture plays a crucial role in human survival as a primary source of food, alongside other sources such as fishing. Unfortunately, global warming and other environmental issues, particularly in less privileged nations, hamper the Agricultural sector. It is estimated that a range of 720 to 811 million individuals experienced food insecurity. Today's agriculture faced significant difficulties and obstacles, as do the surveillance and monitoring systems (climate, energy, water, fields, works, cost, fertilizers, diseases, etc.). The COVID-19 pandemic has exacerbated the susceptibilities and insufficiencies inherent in worldwide food systems. Current agricultural practices tend to prioritize productivity and profitability over environmental conservation and long-term sustainability. To establish sustainable agriculture capable of meeting the needs of a projected ten billion people in the next 30 years, substantial structural and automation changes are required. However, these obstacles can be overcome by employing smart technologies and advancing Artificial Intelligence (AI) in agricultural operations. AI is believed to contribute to global sustainability goals in multiple sectors, particularly in the incorporation of renewable energy. It is anticipated that AI will revitalize both existing and new agricultural fields by retrofitting, installing and integrating automatic devices and instruments. This paper presents a comprehensive review of the most promising and novel applications of AI in the agriculture industry. Furthermore, the role of AI in the transition to sustainability and precision agriculture is investigated.

Precision viticulture is increasingly being applied to automate and optimize grape production in the vineyard. This paper describes the development of a method for automatic selection of regions of interest from hyperspectral images obtained of a row of vines and intended for prediction of soluble solids content. For this purpose, a dataset consisting of hyperspectral images of a row of ‘Sangiovese’ wine grapes was adopted. Hyperspectral images were acquired directly in the field by means of a hyperspectral imaging Vis/NIR system (400–1000 nm) mounted on a ground-based vehicle. The analyses were carried out on 17 different days, under clear or partly cloudy conditions, in the period between post-veraison and harvest. The vineyard row of Sangiovese vines was divided into 11 sections and a hyperspectral image for each section for each day of analysis was acquired. The regions of interest of the hyperspectral images, comprising the areas representing the grapes, were selected using a PLS-DA-based method. The best PLS-DA model provided excellent results, with sensitivity and specificity values of 0.991 and 0.996, respectively. The mean spectra of the selected regions of interest (ROI) were finally used to predict the soluble solids content (SSC) of the grapes by PLS regression to a primary reference analysis. The results of SSC predictions using the automatic selection of ROIs (R2CV = 0.74 and RMSECV = 0.86 °Brix) were on par with similar regression based on carefully manual selection of ROIs (R2CV = 0.73 and RMSECV = 0.87 °Brix).

In the context of food security in mountain regions, this study explores the potential of a Smart Greenhouse (SGH) to address the challenges posed by climate change and limited resources. The SGH utilizes an 8 m2 solar water heating system, weather stations for external temperature and humidity monitoring, as well as internal light, humidity, and temperature sensors to create an optimized environment for crop cultivation. Experiments were conducted with various crops within the SGH during the harsh winter months, typically rendering the cultivable land fallow due to extreme cold conditions. The SGH is equipped with sensors and communication technologies, ensuring continuous data collection and the automated regulation of heating, ventilation, and air conditioning systems. Located in Genekha, Thimphu, this pilot SGH seamlessly integrates renewable energy technology to provide heat during the cold winter season, obviating the need for external heating sources in the summer. The study employed TRNSYS and OpenStudio simulation software to assess the SGH's performance, focusing on temperature and humidity control inside and outside the greenhouse. Additionally, soil pH and nutrient content were examined, and the time required for crop growth and harvesting was recorded. In conclusion, this paper acknowledges some of the current design's limitations and presents recommendations for future enhancements, with a focus on the specific context of improving crop production and food security in mountainous regions.

The current era witnesses the advent of the Internet of Things (IoT), a transformative force giving rise to various technological innovations, most notably connected objects. This paradigm shift catalyzes the widespread adoption of autonomous decision-making systems, particularly in sectors like agriculture, where the aim is to amplify productivity. As a result of the agricultural domain, mushrooms have concurrently emerged as a significant component of daily diets, offering additional vitamins and flavor. Despite their popularity, cultivating mushrooms in open environments poses challenges in maintaining optimal environmental conditions, prompting numerous research efforts. Recognizing the inconsistencies in existing approaches to safeguard vital parameters in mushroom farms, this paper introduces an innovative system utilizing intelligent sensors whose real-time records are managed based on the fuzzy sets concept. These sensors, encompassing the Capacitive Soil Moisture Sensor v1.2, DHT22 Sensor, Light Dependent Resistor (LDR sensor), and Passive Infrared Receiver (PIR sensor), collectively capture essential data for decision-making in mushroom farming. Employing fuzzy logic, the system addresses pivotal aspects such as substrate watering, environmental control, light management, and pest detection. Through experimental results, it becomes evident that the proposed system not only exemplifies the potential of IoT technologies in agriculture but also offers a comprehensive and efficient approach to real-time decision-making. By aggregating sensor data, the system proves instrumental in enhancing the quality and yield of mushroom crops, showcasing a promising trajectory for sustainable and technologically driven agricultural practices.

This study aimed to deploy a deep learning model in a mobile application for the early identification of Fusarium Wilt and Black Sigatoka in bananas. In this paper, a Convolutional Neural Network (CNN) model for the classification of Black Sigatoka banana disease and Fusarium Wilt disease is assessed. A dataset of 27,360 images of diseased and healthy banana leaves and stalks that were collected from the farms using a mobile phone camera served as the training data for this model. An extra class of 407 images that are not of the banana plant downloaded from the internet was used to help the model detect other images not of the banana plant. The CNN model achieved an accuracy of 91.17 % and was deployed in a mobile application for the classification of the diseases. This study shows that deep learning can be implemented and assist in the early identification of banana diseases. The application could detect images of healthy and diseased banana leaves and stalks and images not of the banana plant with a confidence score of more than 90 % in less than five seconds per image and provide research-based mitigation recommendations.

Due to the constantly increasing shortage of labour availability in the agricultural sector, automation is on the rise. This trend is particularly evident in field vegetable production, where the majority of costs are associated with human labour. To automate labour-intensive tasks, intelligent camera systems are required. Typical state of the art learning algorithms requires large amounts of labelled data for a reliable detection and segmentation. The effort for recording and labelling is increased by the diverse conditions on a field that need to be included in the datasets. In this paper, an unsupervised approach is presented, which unifies varying environmental conditions in the field, as well as an analysis to transfer this unpaired deep learning based unification to unknown exposure scenarios, different growth stages and other varieties. For this purpose, a Cycle-Consistent Adversarial Network (CycleGAN) is used which balance varying exposure and background situations to the training data variance without additional labelling effort. By adapting the exposure with a CycleGAN, the required labelling effort could be reduced by up to five times in relation to a heterogeneous dataset, as fewer variations need to be covered for a robust segmentation. The transfer of this approach to other growth stages, as well as the transfer to another cabbage cultivar shows the potential of additional data generalisation.

Extreme weather events such as droughts have posed a significant risk to the agricultural economy of the semi-arid region of the Southwestern U.S. In this paper, a multistage stochastic optimization model is developed for harvest planning and machinery scheduling of guayule, a perennial woody shrub native to this region, and addresses the uncertainty posed by drought scenarios for harvesting in this region. We consider various aspects such as the yield of the crop as a stochastic parameter, time windows for crop quality, multi-machinery scheduling, the penalty for delayed harvesting, shortage, and inventory. The goal of the model is to minimize the expected total costs over the planning horizon. The results present long-term optimal harvest planning and machinery scheduling for crops, and optimal routes for harvesting based on the Geographic Information System (GIS). In addition, we perform analysis on the expected cost breakdown, the shortage cost, and planting plans for guayule, wheat, and cotton to offer a comprehensive comparison of the economic viability under various drought scenarios in this region. This study underscores incorporating drought scenarios into harvest planning and highlights the guayule's resilience for farmers in mitigating economic losses under drought scenarios.

Total acid is an important indicator of the internal quality of red globe grapes. This paper proposed a non-destructive method for the determination of total acid of red globe grapes based on hyperspectral fusion technique. A non-linear LSSVM prediction model based on spectral information, image information and the fusion of the two was built respectively for total acid. The results showed that the MSC-CARS-SPA-LSSVM model of samples about total acid built by fusing the spectra after feature wavelength extraction using the MSC-CARS-SPA algorithm and the image information after dimensionality reduction by the PCA algorithm using the graph fusion technique worked best. The correlation coefficients of the prediction sets of the optimal LSSVM model was 0.9907, which improved the accuracy over the unilateral models based on spectral or image information. A new method for nondestructive detection of total acid of red globe grapes by map fusion technique was discovered.

The digitalization of agriculture is on the rise, and while the dairy industry is not considered a leader in digitalization, more and more technologies have been developed in recent years. In particular, monitoring the health of dairy cows is becoming increasingly digital with Internet of Things (IoT) technologies. IoT-based sensors can potentially support efficient dairy production while enabling better animal welfare. Against this background, the question of technology acceptance among dairy farmers is becoming increasingly important. It is evident that acceptance, especially in the area of IoT-based dairy sensors, has only been sporadically researched, with behavioral factors rarely considered. Acceptance research has predominantly focused on a technological perspective or has been limited to operational and socio-demographic factors, thus failing to capture the entire acceptance process. Based on this, the aim of the paper is to identify prevailing acceptance barriers and examine whether cognitive and affective influencing factors, combined into a single behavioral model, can explain the acceptance, and therewith precondition for adoption, of IoT-based animal sensors. For this purpose, an online survey was conducted among 212 German dairy farmers. The analysis included a partial least squares estimation of the behavioral model. The results show that acceptance barriers differ between users and non-users of IoT-based sensors. However, both groups consider high investment costs as the most important barrier. Furthermore, the results indicate that both cognitive factors and positive anticipated emotions influence acceptance decisions. Additionally, the main determinants of the theory of planned behavior, namely attitude and behavioral control, have a positive influence on the desire to use IoT-based animal sensors. Ultimately, this has a positive impact on the acceptance. The results emphasise that, in addition to the technical requirements and cognitive attitudes, emotional factors are relevant for the acceptance of IoT-based animal sensors among dairy farmers. These findings are important for manufacturers, policy and agricultural and digital associations.

Agricultural products are crucial for the long-term sustainability of developing economies. Agriculture is a vital sector for most developing countries, including Ethiopia. The pumpkin crop is one of the most significant agricultural products globally in terms of human food security. However, it is susceptible to several diseases, including pumpkin common rust diseased leaf, pumpkin Downy mildew diseased leaf, and pumpkin fruit rot, among others. Traditionally, farmers identify diseases through visual observation, which is both inaccurate and time-consuming. Previous research has indicated that binary classification has to be improved because some classes were more difficult to detect. In this paper, we proposed a hybrid method that incorporates the extract features of the ResNet and LeNet networks to construct a pumpkin disease classification model. Dataset gathering, image preprocessing, segmentation, augmentation, feature extraction, and classification are all processes in the proposed hybrid CNN ResLeNet model. Finally, the suggested hybrid method CNN model was assessed, yielding 99.78 % training accuracy, 98.18 % validation accuracy, and 97.21 % testing accuracy1. The study's findings suggest that the proposed hybrid model is suitable for recognizing pumpkin leaf and fruit diseases from digital images of pumpkin.

Recent advancements in deep learning-based approaches have led to remarkable progress in fruit detection, enabling robust fruit identification in complex environments. However, much less progress has been made on fruit 3D localization, which is equally crucial for robotic harvesting. Complex fruit shape/orientation, fruit clustering, varying lighting conditions, and occlusions by leaves and branches have greatly restricted existing sensors from achieving accurate fruit localization in the natural orchard environment. In this paper, we report on the design of a novel localization technique, called Active Laser-Camera Scanning (ALACS), to achieve accurate and robust fruit 3D localization. The ALACS hardware setup comprises a red line laser, an RGB color camera, a linear motion slide, and an external RGB-D camera. Leveraging the principles of dynamic-targeting laser-triangulation, ALACS enables precise transformation of the projected 2D laser line from the surface of apples to the 3D positions. To facilitate laser pattern acquisitions, a Laser Line Extraction (LLE) method is proposed for robust and high-precision feature extraction on apples. Comprehensive evaluations of LLE demonstrated its ability to extract precise patterns under variable lighting and occlusion conditions. The ALACS system achieved average apple localization accuracies of 6.9 - 11.2 mm at distances ranging from 1.0 m to 1.6 m, compared to 21.5 mm by a commercial RealSense RGB-D camera, in an indoor experiment. Orchard evaluations demonstrated that ALACS has achieved a 95% fruit detachment rate versus a 71% rate by the RealSense camera. By overcoming the challenges of apple 3D localization, this research contributes to the advancement of robotic fruit harvesting technology.

Crop diseases and pests constitute significant causes of yield losses for crops. To limit the harm incurred by those events, farmers resort to plant protection products. Such products are known to have adverse effects both on the environment and on human health. Agronomists make continuous efforts to limit the usage of plant protection products to situations where those products are strictly necessary. To determine such situations, agronomists and policy-makers often rely on decision support tools to model and predict the dynamics of plant diseases. Decision support tools are based either on mechanistic models or on statistical approaches learned from large datasets of biotic (e.g., disease incidence, plant phenological stage) and abiotic (meteorological, soil characteristics) observations in cultures. The surge of powerful machine learning (ML) methods in the last decade makes such approaches a natural pathway to model the dynamics of plant diseases.Machine learning models can reveal the factors that contribute the most to disease and pests outbreaks, provided that those models are simple enough for human inspection. Simplicity, however, may come at the price of lower prediction performances when compared to more complex models.In this paper, we offer a deep look at the performance of ML models of different complexity when used on two use cases of crop disease prediction: downy mildew in the grapevine, and Cercospora leaf spot in the sugar beet.We compare model accuracy and complexity using a year-based cross-validation approach. Our results suggest that interannual meteorological variations are a very important factor in plant disease prediction. Moreover, in line with the observations of the research community in interpretable ML, model complexity stands in clear trade-off with accuracy. This makes models of intermediate complexity appealing for predicting the dynamics of crop diseases as they can provide explicit insights about the rationale of their predictions.