Search

The Fruit Harvest Helper, a mobile application developed by Northwest Nazarene University's (NNU) Robotics Vision Lab, aims to assist farmers in estimating fruit yield for apple orchards. Currently, farmers manually estimate the fruit yield for an orchard, which is a laborious task. The Fruit Harvest Helper seeks to simplify their process by detecting apples on images of apple trees. Once the number of apples is detected, a correlation can then be applied to this value to obtain a usable yield estimate for an apple tree. While prior research efforts at NNU concentrated on developing an iOS app for blossom detection, this current research aims to adapt that smart farming application for apple detection across multiple platforms, iOS and Android. Borrowing ideas from the former iOS app, the new application was designed with an intuitive user interface that is easy for farmers to use, allowing for quick image selection and processing. Unlike before, the adapted app uses a color ratio-based image-segmentation algorithm written in C++ to detect apples. This algorithm detects apples within apple tree images that farmers select for processing by using OpenCV functions and C++ code. The results of testing the algorithm on a dataset of images indicate an 8.52% Mean Absolute Percentage Error (MAPE) and a Pearson correlation coefficient of 0.6 between detected and actual apples on the trees. These findings were obtained by evaluating the images from both the east and west sides of the trees, which was the best method to reduce the error of this algorithm. The algorithm's processing time was tested for Android and iOS, yielding an average performance of 1.16 s on Android and 0.14 s on iOS. Although the Fruit Harvest Helper shows promise, there are many opportunities for improvement. These opportunities include exploring alternative machine-learning approaches for apple detection, conducting real-world testing without any human assistance, and expanding the app to detect various types of fruit. The Fruit Harvest Helper mobile application is among the many mobile applications contributing to precision agriculture. The app is nearing readiness for farmers to use for the purpose of yield monitoring and farm management within Pink Lady apple orchards. [ABSTRACT FROM AUTHOR]

Using modern machines like robots comes with its set of challenges when encountered with unstructured scenarios like occlusion, shadows, poor illumination, and other environmental factors. Hence, it is essential to consider these factors while designing harvesting robots. Fruit harvesting robots are modern automatic machines that have the ability to improve productivity and replace labor for repetitive and laborious harvesting tasks. Therefore, the aim of this paper is to design an improved orange-harvesting robot for a real-time unstructured environment of orchards, mainly focusing on improved efficiency in occlusion and varying illumination. The article distinguishes itself with not only an efficient structural design but also the use of an enhanced convolutional neural network, methodologically designed and fine-tuned on a dataset tailored for oranges integrated with position visual servoing control system. Enhanced motion planning uses an improved rapidly exploring random tree star algorithm that ensures the optimized path for every robot activity. Moreover, the proposed machine design is rigorously tested to validate the performance of the fruit harvesting robot. The unique aspect of this paper is the in-depth evaluation of robots to test five areas of performance that include not only the accurate detection of the fruit, time of fruit picking, and success rate of fruit picking, but also the damage rate of fruit picked as well as the consistency rate of the robot picking in varying illumination and occlusion. The results are then analyzed and compared with the performance of a previous design of fruit harvesting robot. The study ensures improved results in most aspects of the design for performance in an unstructured environment. [ABSTRACT FROM AUTHOR]

In order to meet the growing agricultural demands, modern machinery needs to be deployed and cheaper energy supply needs to be ensured. With advancements in artificial intelligence, fruit harvesting robots can improve both the quality and productivity of fruit picking and increase orange fruit exports. However, to ensure effective working of the fruit harvesting robot, an energy source that is both efficient and cost-effective is necessary. This paper aims at designing of a solar energy system for a lightweight fruit harvesting robot for orange orchards in Pakistan and conducting a feasibility study for the deployment of the robot for remote agricultural land. The site for fruit harvesting robot employment is decided by irradiance using random forest regression. Solar system sizing is done based on the design and energy requirements of the fruit harvesting robot. The Homer Pro software is used for simulation of the system to analyse the potential of using solar system for fruit harvesting robot in Sargodha, Pakistan. The results show that compared to hybrid system, a stand-alone system is a more cost effective, reliable, and efficient option with a payback time of 3.36 years and levelized cost of energy being $0.085 per unit kWh. This study proves that solar energy is a viable and cheaper solution for using modern agricultural machineries, like fruit harvesting robot, in remote areas in developing countries, like Pakistan, to enhance productivity and improving quality of the produce. [ABSTRACT FROM AUTHOR]

Automatic apple harvesting robots have received much research attention in recent years to lower harvesting costs. A fundamental problem for harvesting robots is how to quickly and accurately detect branches to avoid collisions with limited hardware resources. In this paper, we propose a lightweight, high-accurate and real-time semantic segmentation network, Bilateral U-shape Network (BUNet), to segment apple tree branches. The BUNet consists mainly of a U-shaped detail branch and a U-shaped semantic branch, the former for capturing spatial details and the latter for supplementing semantic information. These two U-shape branches complement each other, keeping the high accuracy of the Encoder-decoder Backbone while maintaining the efficiency and effectiveness of the Two-pathway Backbone. In addition, a Simplified Attention Fusion Module (SAFM) is proposed to effectively fuse different levels of information from two branches for pixel-by-pixel prediction. Experimental results show (on our own constructed dataset) that BUNet achieves the highest Intersection over Union (IoU) and F1-score of 75.96% and 86.34%, respectively, with minimum parameters of 0.93M and 11.94G Floating-point of Operations (FLOPs) in branch segmentation. Meanwhile, BUNet achieves a speed of 110.32 Frames Per Second (FPS) with input image size of 1280 × 720 pixels. These results confirm that the proposed method can effectively detect the branches and can, therefore, be used to plan an obstacle avoidance path for harvesting robots. [ABSTRACT FROM AUTHOR]

The end effector plays a critical role in fruit picking by a robotic fruit harvesting system. A newly developed vacuum-based robotic apple harvesting system, using the twist-and-pull fruit picking method, has shown promise in picking fruit from clusters and navigating through the tree canopies. The robot's original thin foam end effector failed to achieve acceptable picking performance because it was unable to conform to fruit of different sizes and thus could not generate sufficient suction forces needed to detach fruit. This research was therefore aimed at developing new end effectors to greatly enhance the robot's fruit picking performance. Field manual pulling and twisting experiments for three varieties of apples were conducted, and the critical pulling or suction force and twisting torque needed to detach 95% of apples were determined to be 28.3 N and 0.257 N-m (equivalent to a pulling force of 21.0 N for the current robot's configuration), respectively. Three new silicone-based end effectors of different geometries (denoted as "Straight", "Bellow" and "Curved") were designed and fabricated, and they were evaluated in lab and field experiments. Results showed that the three new end effectors performed significantly better than the original, unconformable end effector based on multiple performance metrics, including vacuum pressure, overall picking success rate, picking rate by the rotation mechanism, and fruit attachment orientation. The "Straight" end effector performed consistently better than the other two new end effectors; it had 87% overall picking success rate, 65% success rate by the rotation mechanism alone, and 89% success rate when the middle section or cheek of fruit attached to the end effector. With further improvements, the "Straight" end effector should meet the apple picking performance requirement for the new vacuum-based robotic harvesting system. [ABSTRACT FROM AUTHOR]

Citrus harvesting is a labor‐intensive and time‐intensive task. As the global population continues to age, labor costs are increasing dramatically. Therefore, the citrus‐harvesting robot has attracted considerable attention from the business and academic communities. However, robotic harvesting in unstructured and natural citrus orchards remains a challenge. This study aims to address some challenges faced in commercializing citrus‐harvesting robots. We present a fully integrated, autonomous, and innovative solution for citrus‐harvesting robots to overcome the harvesting difficulties derived from the natural growth characteristics of citrus. This solution uses a fused simultaneous localization and mapping algorithm based on multiple sensors to perform high‐precision localization and navigation for the robot in the field orchard. Besides, a novel visual method for estimating fruit poses is proposed to cope with the randomization of citrus growth orientations. Further, a new end‐effector is designed to improve the success and conformity rate of citrus stem cutting. Finally, a fully autonomous harvesting robot system has been developed and integrated. Field evaluations showed that the robot could harvest citrus continuously with an overall success rate of 87.2% and an average picking time of 10.9 s/fruit. These efforts provide a solid foundation for the future commercialization of citrus‐harvesting robots. [ABSTRACT FROM AUTHOR]

Abstract. Organic apple (Malus ×domestica Borkh.) growers lack effective strategies to manage preharvest drop. For susceptible cultivars, yield losses caused by preharvest drop can exceed 30% at the beginning of harvest. To address this issue, a formulation of aminoethoxyvinylglyine (AVG) designed for organic use was developed and compared with a commercially available AVG formulation. We evaluated the effects and interactions of the AVG formulation and application number on preharvest drop and fruit maturity in 2017 and 2018. We selected 30 pairs of mature ‘Oregon Spur II Red Delicious’/‘M. 111’ trees planted at the Mountain Horticultural Crops Research and Extension Center in Mills River, NC, USA. Trees were planted with spacing of 2.7 × 6.1 m, trained to a central leader, and received plant protectant sprays that adhered to local recommendations throughout the growing season. Both AVG formulations were applied at 132 mg.L-1 at 3 or 3 + 1 weeks before the anticipated harvest. An untreated control was also included for comparison. The experiment had six replicates and a randomized complete block design with a 2 × 2 augmented factorial treatment structure. A one-way analysis of variance was performed and single degree of freedom contrasts were used to compare treatment groups of interest. During both years, organic and conventional AVG were equally effective for reducing preharvest drop and delaying fruit softening and starch hydrolysis at harvest. During one year, increasing the number of applications of AVG reduced cumulative fruit drop, delayed fruit softening at harvest, and reduced internal ethylene concentrations. Inconsistencies in responses across years may be explained, in part, by abnormally warm temperatures observed in 2018. AVG approved for organic use appears to be a promising preharvest drop management technology with efficacy similar to that of conventional AVG. [ABSTRACT FROM AUTHOR]

Trunk recognition is a critical technology for Camellia oleifera fruit harvesting robots, as it enables accurate and efficient detection and localization of vibration or picking points in unstructured natural environments. Traditional trunk detection methods heavily rely on the visual judgment of robot operators, resulting in significant errors and incorrect vibration point identification. In this paper, we propose a new method based on an improved YOLOv7 network for Camellia oleifera trunk detection. Firstly, we integrate an attention mechanism into the backbone and head layers of YOLOv7, enhancing feature extraction for trunks and enabling the network to focus on relevant target objects. Secondly, we design a weighted confidence loss function based on Facol-EIoU to replace the original loss function in the improved YOLOv7 network. This modification aims to enhance the detection performance specifically for Camellia oleifera trunks. Finally, trunk detection experiments and comparative analyses were conducted with YOLOv3, YOLOv4, YOLOv5, YOLOv7 and improved YOLOv7 models. The experimental results demonstrate that our proposed method achieves an mAP of 89.2%, Recall Rate of 0.94, F1 score of 0.87 and Average Detection Speed of 0.018s/pic that surpass those of YOLOv3, YOLOv4, YOLOv5 and YOLOv7 models. The improved YOLOv7 model exhibits excellent trunk detection accuracy, enabling Camellia oleifera fruit harvesting robots to effectively detect trunks in unstructured orchards. [ABSTRACT FROM AUTHOR]

Depth estimation is one of the bottleneck parts for harvesting robots to determine whether the operation of grasping or picking succeeds or not directly. This paper proposed a novel disparity completion method combined with bilateral filtering and pyramid fusion to improve the issues of incorrect outputs due to the missed or wrong matching when achieving 3D position from 2D images in open-world environments. Briefly, our proposed method has two significant advantages in general. Firstly, occlusion between leaves, branches, and fruits is a universal phenomenon in unstructured orchard environments, which results in the most depth estimation algorithms facing great challenges to obtain accurate outputs in these occluded regions. To alleviate these issues, unlike other research efforts that already exist, we optimized the semi-global matching algorithm to obtain high accuracy sparse values as an initial disparity map; then, an improved bilateral filtering algorithm is proposed to eliminate holes and discontinuous regions caused by occlusion to obtain precise and density disparity outputs. Secondly, due to taking the practical high-efficiency requirements of the automated harvesting robot in its working status into consideration, we attempted to merge multiple low-resolution bilateral filtering results through the pyramid fusion model which goes beyond the trade-off mechanism to improve the performance of both accuracy and time cost. Finally, a prototype harvesting robot was designed to conduct experiments at three kinds of different distances (0.6~0.75 m, 1~1.2 m, and 1.6~1.9 m). Experiment results showed that our proposed method achieved density disparity maps and eliminated holes and discontinuous defects in the disparity map effectively. The average absolute error of our proposed method is 3.2 mm, and the average relative error is 1.79%. In addition, the time cost is greatly reduced more than 90%. Comprehensive experimental results demonstrate that our proposed algorithm provides a potential possibility for designing harvesting. [ABSTRACT FROM AUTHOR]

Fruit detection and recognition has an important impact on fruit and vegetable harvesting, yield prediction and growth information monitoring in the automation process of modern agriculture, and the actual complex environment of orchards poses some challenges for accurate fruit detection. In order to achieve accurate detection of green fruits in complex orchard environments, this paper proposes an accurate object detection method for green fruits based on optimized YOLOX_m. First, the model extracts features from the input image using the CSPDarkNet backbone network to obtain three effective feature layers at different scales. Then, these effective feature layers are fed into the feature fusion pyramid network for enhanced feature extraction, which combines feature information from different scales, and in this process, the Atrous spatial pyramid pooling (ASPP) module is used to increase the receptive field and enhance the network's ability to obtain multi-scale contextual information. Finally, the fused features are fed into the head prediction network for classification prediction and regression prediction. In addition, Varifocal loss is used tomitigate the negative impact of unbalanced distribution of positive and negative samples to obtain higher precision. The experimental results show that the model in this paper has improved on both apple and persimmon datasets, with the average precision (AP) reaching 64.3% and 74.7%, respectively. Compared with other models commonly used for detection, themodel approach in this study has a higher average precision and has improved in other performance metrics, which can provide a reference for the detection of other fruits and vegetables. [ABSTRACT FROM AUTHOR]

Mango, the king of fruits, is one of the most important fruit crops of Pakistan. In the country, Sammar Bahisht Chaunsa is a major cultivar grown for domestic consumption as well as exporting to international markets; however, the cultivar has a strong disorder of biennial bearing. The main cause of the disorder is lack of immediate, sufficient and vigorous postharvest vegetative growth, which is essential for next year flowering/fruiting. The current study was designed to overcome this fruiting pattern by inducing postharvest vegetative growth immediately after fruit harvesting as this growth is must required for flowering. The trial was conducted in the experimental orchard of Mango Research Station, Shujabad on 25 years old trees of cultivar Sammar Bahisht Chaunsa, planted at 12m apart rows and plants. The experiment was performed during 2015-2020. One level of urea (2%) and two levels of KNO3 (1 % and 2%) alone and in various combinations were applied to mango trees immediately after the fruit harvesting to study their effects on the induction of vegetative growth. It was found that KNO3 at 1% ensured the maximum postharvest growth (67.11%) with maximum flowering (39.49%), and fruit yield (241 kg/plant). However nonsignificant difference (25 to 26 OBRIX) in total soluble solids (TSS) and titratable acidity (TA) was noted(0.7 to 0.9%) in all the treatments. It was concluded that application of KNO3 twice at 10 days interval alone or in combination with urea 2% after the harvesting of fruits was found beneficial for getting proper postharvest growth in mango trees to support subsequent fruit crop. [ABSTRACT FROM AUTHOR]

Fixed plot survey was carried out at fortnightly intervals from March to July, 2021 and 2022 jamun fruiting season at six locations of Belagavi district of Karnataka, India. Survey locations comprised in the study were fruit orchard of KRC College of Horticulture Arabhavi, Dupadal, Ankalagi, Kolavi, Kaitanal and Hidkal Dam with six available varieties viz., Dupadal, Kolavi local, Kaitanal local and Hidkal Dam local, Konkan Bahadoli and AJG-85. Pooled data of 2021 and 2022 for fruit damage due to fruit and seed weevil, Curculio c-album revealed that pest incidence starts with an onset of fruit formation and continues till harvesting of fruits. The overall mean fruit damage percentage ranged from 12.3–41.5% with peak incidence during second fortnight of June. Across all the locations highest (50.1%) fruit damage was noticed in Dupadal variety at Dupadal and at the same time less (19.0%) fruit infestation was observed in AJG-85 at Arabhavi. The overall mean of jamun fruit damage due to Conogethes punctiferalis ranged from 0–3.6% with peak incidence during II fortnight of June. Correlation analysis between percentage fruit infestation due to Curculio c-album and Conogethes punctiferalis with prevailing major weather parameters were studied in cv. Konkan Bahadoli. The fruit infestation due to C. c-album and Conogethes punctiferalis was negatively correlated with temperature, positively correlated with relative humidity. C. c-album had positive correlation with rainfall whereas C. Punctiferalis was negatively correlated with rainfall. [ABSTRACT FROM AUTHOR]

Persimmon is a typical fruit crop of the Mediterranean region and, since it is considered a minor species, little information is available on its nutrients need. In the present experiment, it was estimated the quantity of removed and recycled nutrients by Kaki Tipo and Rojo Brillante, the two main varieties of persimmon grown in Emilia-Romagna region (Po Valley, Italy). Plants from ten mature orchards were selected and harvested; organs (leaves in summer, fruits at harvest, abscissed leaves, roots and skeleton) biomass and mineral composition were determined. The yearly uptake of macronutrients was similar for the 2 varieties, accounting for (kg ha−1): N 89-91, P 10-11, K 79-91, Ca 132-162, Mg 22-26 and S 9. While K was mostly found in fruits, Ca and Mg were mainly partitioned to leaves. Among micronutrients, Mn and Fe showed the highest values (1.1–1.3 and 1.2–2.1 kg ha−1, respectively), followed by B (370 g ha−1), while Cu and Zn showed the smallest amounts (less 100 g ha−1). Nitrogen, Ca, Mg and S were those more recycled than removed, while K showed an opposite trend; among the microelements, the annual recycled fractions of B and Mn were higher than that removed. [ABSTRACT FROM AUTHOR]

The olive tree is an important crop in Portugal, where different levels of intensification coexist. The traditional olive orchards present profitability problems, mainly due to harvesting, so there has been a drastic reconversion towards high-density or super-high-density olive orchards. The latter present major constraints due to very specific needs for their use, being practically destined for new orchards. Consequently, the possibility of using systems based on canopy shakers in high-density olive orchards with local varieties is promising. The objective of this work is to evaluate a prototype canopy shaker for the harvesting of high-density olive orchards of the Portuguese variety 'Cobrançosa'. The evaluation is based on the study of canopy shaking in order to adapt canopy training and the adaptability of the machine. For this purpose, the vibration of 72 points of the tree canopy was recorded and a qualitative assessment of the harvest was carried out. Differences were found between the different zones according to the direction of the forward movement of the harvester and the distance to the trunk. These differences were associated with the values obtained for fruit detachment, and a greater quantity of fruit was harvested in the areas of the canopy in contact with the rods. [ABSTRACT FROM AUTHOR]

Accurate models for early harvest estimation in citrus production generally involve expensive variables. The goal of this research work was to develop a model to provide early and accurate estimations of harvest using low-cost features. Given the original data may derive from tree measurements, meteorological stations, or satellites, they have varied costs. The studied orchards included tangerines (Citrus reticulata x C. sinensis) and sweet oranges (C. sinensis) located in northeastern Argentina. Machine learning methods combined with different datasets were tested to obtain the most accurate harvest estimation. The final model is based on support vector machines with low-cost variables like species, age, irrigation, red and near-infrared reflectance in February and December, NDVI in December, rain during ripening, and humidity during fruit growth. [ABSTRACT FROM AUTHOR]

In the hilly area of Gioia Tauro (Calabria, Southern Italy), lemon orchards are grown in consociation with centuriesold olive trees. Lemons are partially shaded by olive canopies and the microclimate at the level of their canopies is suitable for plants growth and quality productions. Under these conditions, lemon trees are grown even without irrigation, providing, despite this limitation, a quality product. This study aimed to i) investigate the qualitative characterisation of two clonal selections of the lemon cultivar Femminello, F. Siracusano (S) and F. Zagara bianca (ZB), from the described intercropping, on irrigated (I) and nonirrigated (NI) crops; ii) assess the quality preservation during cold storage, in order to evaluate the availability of lemons for marketing in a period of shortage such as the summer season. Fruits were harvested at commercial maturity, and cold stored at 10±1°C and RH 8590%, for 60 days. Decay incidence, physiological disorders, weight loss, and the main physicalchemical parameters were assessed at harvest (T0) and every 15 days (T15, T30, T45, T60). The absence of decay and physiological disorders was observed throughout the 60day storage period, in both clonal selections under the two management conditions. The weight loss was greater in fruits from irrigated lemon groves of both S and ZB. S_I showed significantly lower fruits weight and higher titratable acidity than S_NI. Total soluble solids and titratable acidity were statistically lower for ZB_I than for NI fruits. [ABSTRACT FROM AUTHOR]

Endocarp development in olive trees includes three periods: growth (Period I), massive sclerification (Period II) and maximum hardening (Period III). The two first are strongly related to yield and irrigation management. Period I was reported to coincide with mesocarp cell division and thus with final fruit size. Period II was considered to be the most drought-resistant phenological stage. However, little is known in olive trees about the length of these periods and their capacity for predicting fruit size at harvest. The aim of this work was to evaluate the length of both periods in different cultivars and different location of full irrigated orchards. We also aimed to study the fruit feature impact on harvest at the end of Period I. Data from full irrigated olive orchards of cv Cornicabra, Arbequina and Manzanilla in two different locations (Ciudad Real, Central Spain, and Seville, South Spain) were used. The pattern of pit-breaking pressure throughout the season was measured with fruit samples for several years (2006 to 2022). These data and climatic data were used to compare different estimation methods for the length of Period I and II of endocarp development. Then, fruit volume and dry weight at the end of Period I were used to estimate fruit features at harvest. Results suggest that the Period I length was less temperature- and cultivar-dependent than expected. The duration of this period was almost constant at around 49 days after full bloom. Thermal time was negatively correlated with fruit size at the end of Period I. On the contrary, a lineal thermal model presented the lowest variability when estimating the Period II length, which was also affected by the cultivar. The best fit between fruit dry weight and volume at Period I vs. harvest was unique for oil cultivars (Cornicabra and Arbequina), while cv Manzanilla presented a different relationship. A temperature increase in the future would not affect the Period I length but would reduce the fruit size at the end of this period and at harvest. [ABSTRACT FROM AUTHOR]

An important stage of the agri-food supply chain is production (agricultural practices and harvesting), where relevant decisions are made in a short period of intensive activity. Regarding the major fruit supply chain, the complexity of these decisions increases because most fruit export companies manage several orchards with different fruit species and varieties. Moreover, the amount of labor and resources must be estimated at least two months in advance of the harvest so that they can be relied upon during the harvest season. In this study, a multi-objective modeling approach for supporting tactical harvest planning decisions of major fruits is developed, aiming to evidence its adequacy and to identify different harvest plans according to the prioritized objectives. Therefore, a multi-objective linear programming (MOLP) model, which considers three conflicting objectives, is proposed. The three objectives correspond to minimizing harvest costs, fruit loss and harvest days. For solving the MOLP model, five a posteriori multi-objective exact methods are compared using a synthetic instance based on a real case. The Pareto frontiers obtained by these methods are analyzed using three metrics related to accuracy, cardinality, and diversity criteria, so that the most suitable method can be selected. Thus, the selected method is applied to an actual case study of a Chilean fruit export company. In addition, the ε -constraint method is used in this case study to compare the obtention of non-dominated solutions in the convex and non-convex regions of the Pareto frontier. The results show that harvest plans vary significantly according to the prioritization of the objective functions. Thus, the percentage difference between each objective function's worst and best values is around 497% when the harvest costs are minimized, around 1,448% when the fruit loss is minimized, and around 18% when the number of harvest days is minimized. In addition, the adequacy of the multi-objective approach rather than the mono-objective approach to support tactical harvest planning decisions is demonstrated. Furthermore, a scenario analysis for considering changes in the harvest periods is carried out by considering three possible scenarios. This analysis is required because these periods can change every year due to different factors, such as weather conditions and agricultural practices, among others. The results show that the harvest costs and fruit loss increase, on average, 87% and 100%, respectively, when the harvest period is short, that is, one day. Finally, managerial insights about the information obtained by the MOLP model are discussed. [ABSTRACT FROM AUTHOR]

In the old course of the Yellow River area, most orchards are over-applied with nitrogen (N) fertilizers. To improve N management in this area, a 15N tracing experiment was conducted to investigate the absorption, distribution and loss of spring-topdressing urea in pear orchards from March to August 2019. The 7-year-old Sucui 1 pear was used as the test material, and 277.5 g of ordinary urea and 15 g of 15N-urea N were evenly applied to each plant. The N absorption, distribution and utilization efficiency of different organs from the flowering stage to the post-harvest stage were analyzed, and the residual and loss of N in the soil were also discussed. The N fertilizer utilization rate increased with the advancement of the phenological period. The N fertilizer utilization rate in the full bloom period is 10.39%, which is the fastest growing period, and reached a maximum of 23.62% in the post-harvest stage. In the young fruit stage, the amount of N derived from labeled fertilizer (%Ndff) of the fruit was only 1.02%, and most of the new vegetative organs were above 1%. Residual amount in the 20–40 cm soil layer was significantly higher than that in other soil layers. Direction of N fertilizer is N fertilizer loss>soil residue>tree absorption. N loss in the fruit expansion stage and the harvest stage is higher, which are 3.76 g and 3.74 g, respectively. N utilization rate in this area is low throughout the year. There is nutrient competition between reproductive growth and vegetative growth, which can be effectively alleviated by spring top-dressing. The N loss during fruit expansion and harvesting is serious. Attention should be paid to split fertilizer application and the timely supplementation of an appropriate amount of N fertilizer to improve N use efficiency. [ABSTRACT FROM AUTHOR]

Achieving rapid and accurate detection of tree fruits under natural environments is essential for many precision agriculture application (such as harvesting robots and yield estimation). A real-time citrus recognition method was proposed in this paper by improving the latest YOLOv4 (You Only Look Once version 4) detector for using in orchard environments. The Canopy algorithm and the K-Means + + algorithm were used to automatically select the number and size of prior frames corresponding to the image dataset. Then, an attention mechanism module is added in front of the output layer of each feature of different scales, and a depthwise separable convolution module is added before the upsampling of the network neck to better detect clementines in complex backgrounds. Finally, the network is pruned using the scientific control-based neural network pruning (SCOP) algorithm, and the parameters of the pruned model were fine-tuned to restore some recognition accuracy. Five common used deep learning algorithms including the Faster R-CNN, SSD, YOLOv3, YOLOv4 and Detectron2 were compared to verify the effectiveness of the proposed method. The experimental results showed that our improved YOLOv4 detector works well for detecting different growth periods of citrus in natural orchard environment. The average accuracy increase from 92.89 to 96.15%, the detection time for each image is 0.06s, both are superior the above five algorithms. While the model size was down from 250 MB to 187 MB. This would promise the proposed method being suitable for orchard yield estimation and the development of fruit harvesting robots. [ABSTRACT FROM AUTHOR]

The end effector plays a critical role in fruit picking by a robotic fruit harvesting system. A newly developed vacuum-based robotic apple harvesting system, using the twist-and-pull fruit picking method, has shown promise in picking fruit from clusters and navigating through the tree canopies. The robot's original thin foam end effector failed to achieve acceptable picking performance because it was unable to conform to fruit of different sizes and thus could not generate sufficient suction forces needed to detach fruit. This research was therefore aimed at developing new end effectors to greatly enhance the robot's fruit picking performance. Field manual pulling and twisting experiments for three varieties of apples were conducted, and the critical pulling or suction force and twisting torque needed to detach 95% of apples were determined to be 28.3 N and 0.257 N-m (equivalent to a pulling force of 21.0 N for the current robot's configuration), respectively. Three new silicone-based end effectors of different geometries (denoted as "Straight", "Bellow" and "Curved") were designed and fabricated, and they were evaluated in lab and field experiments. Results showed that the three new end effectors performed significantly better than the original, unconformable end effector based on multiple performance metrics, including vacuum pressure, overall picking success rate, picking rate by the rotation mechanism, and fruit attachment orientation. The "Straight" end effector performed consistently better than the other two new end effectors; it had 87% overall picking success rate, 65% success rate by the rotation mechanism alone, and 89% success rate when the middle section or cheek of fruit attached to the end effector. With further improvements, the "Straight" end effector should meet the apple picking performance requirement for the new vacuum-based robotic harvesting system. [ABSTRACT FROM AUTHOR]

Fruit object detection is crucial for automatic harvesting systems, serving applications such as orchard yield measurement and fruit harvesting. In order to achieve fast recognition and localization of green apples and meet the real-time working requirements of the vision system of harvesting robots, a fast optimized Foveabox detection model (Fast-FDM) is proposed. Fast-FDM uses an optimized form of anchor-free Foveabox to accurately and efficiently detect green apples in harvesting environments. Specifically, the EfficientNetV2-S with fast training and small size is used as the backbone network, a weighted bi-directional feature pyramid network (BiFPN) is employed as the feature extraction network to fuse multi-scale features easily and fast, and then the fused features are fed to the fovea head prediction network for the classification and bounding box prediction. Furthermore, an adaptive training sample selection (ATSS) method is adopted to directly select positive and negative samples, allowing green fruits of different scales to obtain higher recall and achieve more accurate green apple detection. Experimental results show that the proposed Fast-FDM realizes a mean average precision (mAP) of 62.3% for green apple detection using fewer parameters and floating point of operations (FLOPs), achieving better trade-offs between accuracy and detection efficiency. [ABSTRACT FROM AUTHOR]

• Adaptive Zone-fusion Network for waxberry semantic segmentation was proposed. • We have captured and provided a comprehensive and complete dataset. • Improve segmentation accuracy and reduce the impact of various interferences. • Our research promotes the advancement of automated fruit harvesting technology. In China, waxberries are a specialty fruit that require large harvesting efforts each season. To ameliorate these issues, automated fruit-picking equipment has been widely developed within the last decade. However, most image analysis research used for waxberry segmentation contains issues such as incomplete datasets, insufficient segmentation accuracy, and susceptibility to interference. Here, we developed the Adaptive Zone-fusion Network (AZNet) that enables automated fruit harvesting and is designed to precisely segmentate waxberries in an orchard environment. Initially, we collected a diverse range of images to train and evaluate a better model, and found that AZNet demonstrated exceptional performance and achieved a mean Accuracy (mAcc) of 99.83%, mean Intersection over Union (mIoU) of 96.57%, and mean Dice coefficient (mDice) of 98.20%. To further validate AZNet's effectiveness, we compared several state-of-the-art (SOTA) models, including DeepLab V3, SegNet, PSPNet, U-Net, RefineNet, FCN, DANet, and FastSCNN. Along with these comparisons, we also performed an ablation study to substantiate the individual impact of each module within AZNet. Through these experiments, we found that AZNet exhibits the capability to achieve accurate semantic segmentation of waxberries in complex orchard environments, and is suitable for automated fruit harvesting devices. Ultimately, AZNet improves upon previous models by constructing information fusion modules through spatial attention, channel attention, Atrous Spatial Pyramid Pooling (ASPP), and graph convolutional methods, which can be utilized in the future to enhance other fruit-picking software to reduce fruit loss and the investment of human resources. [ABSTRACT FROM AUTHOR]

China has been known as the largest producer of forest and fruit in the world, the same as grain and vegetables. The forestry and fruit industry can gradually dominate the ecological and economic benefits for better living. Therefore, a timely, non-destructive and efficient harvesting of mature fruits has greatly contributed to the subsequent fruit storage, transportation, processing, and sales. Mechanical harvesting has also been one of the most promising operations in the whole process of fruit production. More importantly, fruit mechanical harvesting has been one of the important ways for the forest fruit industry to transform the development mode for cost saving, high efficiency, and competitiveness in the international market. An automatic, intelligent, and unmanned harvesting of forest fruits can be preferred for the better performance of mechanical fruit harvesting, together with the high efficiency and low-loss picking in technical focus. In this review, the research progress was summarized on the mechanical harvesting of forest fruits in recent years, in terms of fruit planting and distribution in China. The planting forms mainly included the small step, steep slope, traditional, gentle slope, high ridge, and wide row dense planting. The reasons were clarified for the bottleneck of mechanical harvesting in the development of the forest fruit industry. The mechanical harvesting of forest fruit was systematically divided into one-time combined fruit picking, and fruit picking after vibration harvesting. Among them, the one-time combined fruit picking was divided into the pneumatic vibration, mechanical vibration (i.e. shaking, and impact type), orchard picking platform, and robot picking, according to the different operation modes of the picking parts. By contrast, the fruit picking after vibration harvesting was divided into the mechanical and pneumatic types, according to the different picking power modes. Then, the research status and development trends were elaborated to combine with the development requirements and application scenarios of the forest fruit industry. Four challenges were reviewed for the mechanical harvesting of forest fruit in China, including the complex working environment, the low level of appropriate mechanization, the limited basic research, and the lacking of Research and Development (R&D) talents. Accordingly, some recommendations were proposed for the mechanized harvesting of forest fruit. National support and R&D investment were the first concern to strengthen the cooperation between agricultural machinery and agronomy, thus promoting the mechanization level of the forest fruit industry. A production-university-research technical personnel training system was highly required to conquer the state-of-the-art technologies in the facilities and equipment during harvesting. It is very necessary to promote the deep integration of agricultural machinery and agronomy under the mode of mechanical harvesting of fruits and vegetables. The main focus can be made to explore the new technical mode with the planting agronomy for the mechanical harvesting of fruits and vegetables. Various agricultural requirements can be comprehensively considered to evaluate the reasonable technical system for better mechanical fruit harvesting with high performance, efficiency, and reliability. In a word, the mechanized operation of forest fruit can be widely expected to serve as the most effective way for cost saving, low labor intensity, but high efficiency during [ABSTRACT FROM AUTHOR]

Olive oil (Olea europaea L.) is widespread due to its bioactive composition, sensorial characteristics, health benefits, and protection against certain diseases. Most of papers reported so far aimed to highlight its content as well as the quality of this edible oil in many different countries. In the present work, we targeted to review the sets of factors that could influence olive oil quality (whether it is virgin olive oils, or extra-virgin olive oils produced in different areas of origin, from several genetic backgrounds, plus the olives and olive oil processing from the fruits harvesting to olive oil bottling and storage, i.e., from farmers to consumers). In addition to all the internal and external variables, the last sub-paragraph was devoted to cooking effects. Notably, olive oil quality differed depending on many factors, thus, to obtain excellent quality, producers should be alerted to the attributes of their orchards (location; latitude and altitude, pedoclimate conditions) to adapt to the appropriate growing practices, like irrigation, fertilization, and pruning. The harvest time/ripening degree, which is related to the cultivated variety and the origin areas climate, as well as the extraction methods, bottling and storage conditions (container's type and storage time) were also considered. [ABSTRACT FROM AUTHOR]

Guava producers in Brazil have reported the occurrence of small, pink spot lesions on the surface of fruits in orchards. Symptomatic fruits were harvested from commercial guava orchards, and 18 monosporic isolates were obtained. All tested isolates were pathogenic to guava fruit and showed colony and conidial characteristics indicative of cercosporoid fungi. Phylogenetic analysis of the concatenated gene sequences showed that all isolates belonged to the genus Cercospora and clustered into four haplotypes. Conidia released 3–5 germ tubes, and penetration occurred through the stomata. All isolates produced cercosporin. Histopathological analyses showed that hyphal growth occurred intercellularly and that there was a collapse of epidermal and subepidermal cells in the injured areas. In addition, the oil cavities in the lesioned tissue had large numbers of lipid droplets compared with the nonlesioned tissue. This is the first report of Cercospora spp. as the causal agent of pink spot disease of guava fruit in Brazil. [ABSTRACT FROM AUTHOR]

Summary: We present the first worldwide study on the apple (Malus × domestica) fruit microbiome that examines questions regarding the composition and the assembly of microbial communities on and in apple fruit. Results revealed that the composition and structure of the fungal and bacterial communities associated with apple fruit vary and are highly dependent on geographical location. The study also confirmed that the spatial variation in the fungal and bacterial composition of different fruit tissues exists at a global level. Fungal diversity varied significantly in fruit harvested in different geographical locations and suggests a potential link between location and the type and rate of postharvest diseases that develop in each country. The global core microbiome of apple fruit was represented by several beneficial microbial taxa and accounted for a large fraction of the fruit microbial community. The study provides foundational information about the apple fruit microbiome that can be utilized for the development of novel approaches for the management of fruit quality and safety, as well as for reducing losses due to the establishment and proliferation of postharvest pathogens. It also lays the groundwork for studying the complex microbial interactions that occur on apple fruit surfaces. [ABSTRACT FROM AUTHOR]

Snake fruit blooms every three months, but only one to two of the flowering seasons result in fruit production. Failure of fruit-set causes the fruit harvest seasonally. Therefore, the aim of this research was to increase off-season production by preventing the failure of fruit-set with various exogenous auxin spray. The research was conducted in the snake fruit orchard at Bebandem Sub-District, Karangasem Regency, Bali Province, Indonesia, from February to November 2019. The experiment was designed as a nested factorial with a randomized block design. The treatment consisted of three types of auxin with a specific concentration for each, IAA at 0, 25, 50 and 75 ppm, IBA at 0, 50, 100 and 150 ppm, and NAA at 0, 50, 100 and 150 ppm. The IAA, IBA, and NAA were sprayed on the flowers and leaves 3 times in March, April, and May. The results showed the highest percentage of fruit-set, fruit weight per tree, and weight per fruit obtained with IAA treatment were 59.80 %, 2,612.91 g, and 91.06 g, respectively. These results were not statistically different to those of the IBA and NAA. Furthermore, the IAA concentration of 25, 50, and 75 ppm produced a significantly higher percentage of fruit-set compared with the control. The increased fruit-set in the IAA, IBA and NAA treatments resulted in better yield than the control. The novelty of this result is that spraying snake fruit with IAA, IBA and NAA can prevent the failure of the fruit-set. Therefore, IAA, IBA, and NAA are recommended to be applied to increase off-season fruit production of snake fruit. [ABSTRACT FROM AUTHOR]

Apples are the most popular fruits grown in Polish orchards. In order to obtain the best quality fruit, it is necessary to improve plantation maintenance, fruit harvesting, and processing. Given that many fruits are exposed to external factors, including forces that adversely affect their structure—causing them to crack, bruise, or crush—it is necessary to provide conditions that do not adversely affect their quality. Therefore, the aim of this article was to develop a simplified model of an apple that could be tested under different loads using the finite element method. The parameters of the model were selected to reflect the actual apple as accurately as possible. To assess the apples under impact load, as well as the construction of the FEM model, concrete and wooden substrates were used, where apples were dropped from height of 10 mm and 30 mm. Due to this research, an apple model was obtained that reflects the actual object very well (high R2 coefficient). In addition, the layering and distribution of surface pressures of the real and model objects from the distribution are presented. This shows that the constructed model corresponds to the behaviour of the biological material, subjected to loads in real conditions. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Agro-Environment Science is the property of Journal of Agro-Environment Science Editorial Board and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

• Picking robots must operate continuously throughout the orchard for commercial and practical purposes. • Efficient visual servo control methods of locomotion and picking are necessary for full-field harvesting. • Serialized keyframes from SLAM provide high-confidence global crop information. • The multi-map fusion mechanism greatly improves locomotion stability, despite computational costs. • Solely relying on multi-view positioning does not enhance picking accuracy but affects picking efficiency. Fruit-picking robots are crucial for achieving efficient orchard harvesting. To genuinely meet the commercial production needs of farmers, the new generation of fruit-picking robots must be capable of demonstrating complete and continuous observation, movement, and picking behaviors throughout complex orchards, akin to real human employees. This poses systematic challenges, as many prior researches have focused solely on a part of the continuous operation of the entire orchard, such as fruit positioning, navigation, path planning, or grasping. These isolated basic functions are important but insufficient for fulfilling operational requirements on a macro scale and continuous situation. Developing an efficient control method for each basic module and constructing their internal coordination is vital for transitioning a harvesting robot from a functional prototype to a practical machine. In this context, this study tackles the visual servo control problem for efficient locomotion, picking, and their seamless integration. A set of vision algorithms for locomotion destination estimation, real-time self-positioning, and dynamic harvesting is proposed. Additionally, a solid coordination mechanism for continuous locomotion and picking behavior is established. Each method offers distinct advantages, such as improved accuracy, adaptability to varying conditions, and enhanced picking efficiency, enabling the robot to operate autonomously and continuously. Comprehensive field experiments validated the soundness of the methods. The primary contribution of this study lies in addressing the challenge of continuous operation in an entire orchard as a systematic problem and providing new insights into control methods for the future development of highly autonomous, practical, and user-oriented fruit harvesting systems. [ABSTRACT FROM AUTHOR]

• Delaying harvest suppressed the superficial scald development. • Antioxidants in apple peel increased with advanced harvesting. • AVG treatment enhanced superficial scald incidence and severity. • Antioxidants in the blushed side were higher compared with the shaded side. The effects of plant growth regulators (PGRs) and harvest date on antioxidant contents at harvest and superficial scald development in 'NY2' apple during storage have been studied. Aminoethoxyvinylglycine (AVG, ReTain®) and 1-methylcyclopropane (1-MCP, Harvista™) were sprayed two and one weeks, respectively, before the first of three weekly harvests. The apples were stored at 0.5 °C for 20 weeks + 7 d at 20 °C. At harvest, color and index of absorbance difference (I AD) values, and contents of total phenolics, anthocyanins, chlorophylls, carotenoids, ascorbic acid (TAA), dehydroascorbic acid (DHA) glutathione (GSH), reduced glutathione (GSSG) and total antioxidant activity of blushed and unblushed sides of the apple were measured. The antioxidant contents increased with later harvest dates. Chlorophyll and GSSG were higher in unblushed than blushed skin tissues, while all antioxidant contents were highest in blushed skin tissues. Superficial scald incidence was highly correlated with less mature fruit, lower fruit anthocyanin, higher chlorophyll content at harvest. α-Farnesene, conjugated trienols (CTols), and I AD value index were measured at 0, 1, 2, 4, 6, 10, 15, and 20 weeks at 0.5 ⁰C for fruit from harvest 1. Scald incidence and severity was higher in AVG-treated fruit than untreated or 1-MCP treated fruit and did not decrease at later harvest dates. Using harvest management and preharvest application of plant growth regulators is a promising strategy to improve the nutritional value of 'NY2′ apples, reduce physiological disorders, and maintain fruit quality. [ABSTRACT FROM AUTHOR]

• Harvest platform speed is optimized online based on fruit load and picker speeds. • Platform speed and worker-lift elevations are co-optimized: full co-robotic mode. • Manual, speed or height-only control and co-robotic mode were compared. • Full co-robotic mode was up to 25% faster in field experiments. • The optimization-based control algorithm is available as open-source. Orchard harvest-aid platforms are used in high-density orchards with SNAP (Simple, Narrow, Accessible, and Productive) tree architectures. In these orchards, trees are planted and trained into fruiting walls, and workers harvest the fruit while standing on the platform. A worker in the front – the "driver" - is responsible for controlling the platform speed and keeping it centered in the row while they pick fruit. The use of harvest-aid platforms improves the picking efficiency, safety, and ergonomics of the workers compared to ladder-based picking. However, the non-uniform fruit distribution in tree canopies results in an incoming fruit rate (demand for labor) that varies spatially and temporally and is not necessarily matched to the workers' picking positions and harvest speeds (supply of labor). This mismatch lowers the overall harvest efficiency. A previous study improved upon this mismatch by introducing an independent actuated lift for each worker and dynamically adjusting each worker's picking height based on the estimated incoming fruit distribution and the workers' harvesting rates. The horizontal moving speed of the platform was not controlled. This work presents an integrated system that optimizes the platform's travel speed and lift heights in real-time to increase the platform's harvest throughput. Simulation experiments using digitized fruit distributions investigated the algorithm's potential gain under different settings. Field experiments were performed in a commercial apple orchard in Lodi, CA, with Fuji apples on V-trellised trees using a "robotized" platform and two workers. Two modes were implemented in the experiments: the "conventional" mode, which represents current practice, where workers' heights are fixed, and the platform speed is adjusted by the front worker, and the "co-robotic" mode, where the optimizing algorithm dynamically adjusts worker's heights and platform speed. A total of 3,227 kg of apples were harvested during the experiment. The overall throughput of the "co-robotic" mode was 261.8 kg/h if apple stems were clipped and 501.1 kg/h if apple stems were not clipped. The corresponding overall throughputs of the "conventional" mode were 235.3 kg/h and 397.7 kg/h. The results showed that the "co-robotic" mode improved the harvesting throughput by 11 % (clipping) and 25 % (without clipping). The code for the models, optimization system, and simulation has been made available as open-source on https://github.com/AgRoboticsResearch/corobotic-platform [ABSTRACT FROM AUTHOR]

Currently Colombia does not have enough studies on avocado sensory quality. So, the aim of this work was to evaluate the harvest index and storage duration on avocado sensory quality. Avocado fruit were harvested from 5 farms in Antioquia. Sensory descriptors were evaluated on a scale from 0 to 5, and general quality (GQ) was evaluated from 0 to 3. Fruit with low GQ exhibit hard, wet, and fibrous textures; they taste green, astringent, spicy, sour, umami, and acid, and the odor is sour, green, and acid. Fruits have a low sensory quality at the beginning of the harvest season. As DM content increases, the sensory quality improves. Orchard location and fruit storage duration influenced its sensory descriptors and general quality. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Taiwan Agricultural Research is the property of Taiwan Agricultural Research Institute, Council of Agriculture, Executive Yuan and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The human labor required for tree crop harvesting is a major cost component in fruit production and is increasing. To address this, many existing research works have sought to demonstrate commercially viable robotic harvesting for tree crops, though successful commercial products resulting from these have been few and far between. Systems developed for specific crops such as sweet peppers or apples have shown promise, but the vast majority of cultivar types remain unaddressed, and developing a specific system for each one is inefficient. In this study, an easily modifiable development platform for robotic fruit harvesting is presented, this can be used to test specific design choices on different fruit and growing conditions. The system is evaluated in a commercial plum orchard, with no crop modifications. Both a hard and soft gripper are trialed, along with three object detector approaches and two picking motions. Some existing techniques are found to be counterproductive for plums, while soft robotics and persistent target tracking significantly improve performance. The best harvest success rate of 42%, was observed when using the soft gripper with complex motion. This is lower than expected based on prior testing with apples and indicates the difficulty in moving to new fruit types. Unique challenges specific to the plum type and growing style are examined in the context of system module design choices. [ABSTRACT FROM AUTHOR]

Fruit detection in orchards is one of the most crucial tasks for designing the visual system of an automated harvesting robot. It is the first and foremost tool employed for tasks such as sorting, grading, harvesting, disease control, and yield estimation, etc. Efficient visual systems are crucial for designing an automated robot. However, conventional fruit detection methods always a trade-off with accuracy, real-time response, and extensibility. Therefore, an improved method is proposed based on coarse-to-fine multitask cascaded convolutional networks (MTCNN) with three aspects to enable the practical application. First, the architecture of Fruit-MTCNN was improved to increase its power to discriminate between objects and their backgrounds. Then, with a few manual labels and operations, synthetic images and labels were generated to increase the diversity and the number of image samples. Further, through the online hard example mining (OHEM) strategy during training, the detector retrained hard examples. Finally, the improved detector was tested for its performance that proved superior in predicted accuracy and retaining good performances on portability with the low time cost. Based on performance, it was concluded that the detector could be applied practically in the actual orchard environment. [ABSTRACT FROM AUTHOR]

In Italy, more than one third of workers recruited in agriculture comes from abroad and the incidence of foreign workers is increasing. One of the areas with a consolidated presence of foreign employment in agriculture is the Autonomous Province of Trento, where 66.7% of total workers comes from other countries. The highest request of workers, expressed from June to October (corresponding with the harvest of fruits, grapevine and berries) is not satisfied by the local workforce and the recruitment of foreign workers is crucial to satisfy the farms labour needs. This situation has shaped a local regulative framework where farmers and the local institutions collaborate actively in order to guarantee the yearly satisfaction of workforce needs, minimizing the risk of labour shortage and irregular work. The paper is developed around a descriptive analysis based on the elaboration of national and provincial dataset (mainly INPS and Labour Agency) and gives a general overview of the foreign workforce components in the AP of Trento. [ABSTRACT FROM AUTHOR]

Non-destructive estimates information on the desired properties of fruit without damaging them. The objective of this work is to present an algorithm for the automatic and non-destructive estimation of four maturity stages (unripe, half-ripe, ripe, or overripe) of Fuji apples (Malus Malus pumila) using both colour and spectral data from fruit. In order to extract spectral and colour data to train a proposed system, 170 samples of Fuji apples were collected. Colour and spectral features were extracted using a CR-400 Chroma Meter colorimeter and a custom set up. The second component a ∗ of La ∗ b ∗ colour space and near infrared (NIR) spectrum data in wavelength ranges of 535–560 nm, 835–855 nm, and 950–975 nm, were used to train the proposed algorithm. A hybrid artificial neural network-simulated annealing algorithm (ANN-SA) was used for classification purposes. A total of 1000 iterations were conducted to evaluate the reliability of the classification process. Results demonstrated that after training the correction classification rate (CCR, accuracy) was, at the best state, 100% (test set) using both colour and spectral data. The CCR of the four different classifiers were 93.27%, 99.62%, 98.55%, and 99.59%, for colour features, spectral data wavelength ranges of 535–560 nm, 835–855 nm, and 950–975 nm, respectively, over the test set. These results suggest that the proposed method is capable of the non-destructive estimation of different maturity stages of Fuji apple with a remarkable accuracy, in particular within the 535–560 nm wavelength range. Image 1 • Maturation stages estimated by computer vision validated over several video frames. • Hardware used includes camera, spectrometer, and colorimeter. • Results include accuracy under four different apple maturation stage classes. • System could potentially be embeded in automatic fruit harvesting robots. • System could potentially be adapted to industry environmental conditions. [ABSTRACT FROM AUTHOR]

Fruit size at harvest is an economically important variable for high-quality table fruit production in orchards and vineyards. In addition, knowing the number and size of the fruit on the tree is essential in the framework of precise production, harvest, and postharvest management. A prerequisite for analysis of fruit in a real-world environment is the detection and segmentation from background signal. In the last five years, deep learning convolutional neural network have become the standard method for automatic fruit detection, achieving F1-scores higher than 90 %, as well as real-time processing speeds. At the same time, different methods have been developed for, mainly, fruit size and, more rarely, fruit maturity estimation from 2D images and 3D point clouds. These sizing methods are focused on a few species like grape, apple, citrus, and mango, resulting in mean absolute error values of less than 4 mm in apple fruit. This review provides an overview of the most recent methodologies developed for in-field fruit detection/counting and sizing as well as few upcoming examples of maturity estimation. Challenges, such as sensor fusion, highly varying lighting conditions, occlusions in the canopy, shortage of public fruit datasets, and opportunities for research transfer, are discussed. • Deep learning has meant a breakthrough in fruit detection/counting. • There is a need for publicly available codes and datasets for fruit detection. • Advanced fruit sizing methods based on 3D data do not require calibration targets. • Future fruit sizing methods should provide results at different degrees of occlusion. • Commercial fruit counting and sizing systems will be facilitated by lightweight CNNs. [ABSTRACT FROM AUTHOR]

The dry matter content of New Zealand kiwifruit is currently measured using destructive testing of a 90-fruit sample. Dry matter content varies from 14 to 20% with a standard deviation of 1.05% in the test measurement. This work investigates the use of multispectral data from satellite images of kiwifruit orchards to predict the dry matter of both green and gold kiwifruit. A novel method is developed that reduces the four-dimensional satellite data to three-dimensional unit color vectors and these show a strong linear relationship with measured dry matter. Regression on the data from a set of 20 'training' orchards yielded predictions of dry matter with a standard deviation of 0.76%. When the resulting model was applied to nine test orchards the standard deviation of predicted dry matter was 0.73%. The prediction of dry matter was accurate even when applied to images taken when the fruit was too immature for the standard dry matter measurement test. Therefore, satellite image data may provide a more accurate and non-destructive alternative to the standard 90-fruit test method. It can also provide a way to visualize the variation in dry matter content of the fruit in an orchard; showing regions that would benefit from remedial action and defining areas where fruit harvesting is optimal. Several vegetation indices, derived from satellite image data, are reported in the literature. The strength of the linear correlation between dry matter and twelve common vegetation indices was tested but was found to be much weaker than the correlation developed in this research. [ABSTRACT FROM AUTHOR]

BACKGROUND: In Europe mulberry cultivation is attracting growing interest for the valuable nutraceutical properties of its fruit; however, mulberry orchards specialized in fruit production only are not very common, therefore an adequate harvesting technique has not been developed. OBJECTIVE: The aim of this work was to evaluate an innovative harvesting system in terms of efficiency and labour required, and to monitor fruit losses. METHODS: Harvesting operations were recorded during inspections to the field during the fruit ripening season; chemical assessment on fruit quality and mechanical tests on the detachment force of fruits at different ripening stages were carried out. RESULTS: We described the structure and functions of the innovative system, its harvesting capacity (kg h–1) in comparison to traditional methods, the related losses, and the quality of collected fruits. CONCLUSIONS: The innovative harvesting system requires a minimum of four workers, but it results in a two to three-fold increase of the hourly productivity per person. Furthermore, it is constructionally and functionally very simple and it does not require important initial or maintenance investments. Overall the system can save labour costs and contribute to spreading of moriculture for fruit production. [ABSTRACT FROM AUTHOR]

Two plum cultivars 'Record' and 'Empress' and one prune cultivar 'Common Prune' were planted in spring 2014, spaced at 4.5 × 1.5 × 2.0 m, to be trained to the "Y"- and "V"-trellising systems for mechanical harvesting of dessert fruits with a canopy-contact harvester. The applied trellising systems were compared with the standard central leader system at the same spacing. Trellised trees showed a tendency to grow less well than leader-trained trees, but during the four years of training, they created a higher canopy volume than the control trees because of their spreading form. Trellised trees were able to give yields comparable to those of standard trees. Light distribution within the tree canopy was acceptable in all the training systems. After 4 years of training, the trees were suitable for mechanical harvesting. The effectiveness of mechanical harvesting varied from 85% to 90%. The quality of the fruits harvested with a canopy-contact harvester was comparable to that of manually picked fruits. The consumption of quality of fruits after mechanical harvesting rated on a 5-grade scale was 0.5 grade lower than that of manually harvested fruits. These fruits were acceptable in the local fruit market. Mechanical harvesting was 10–30 times faster compared to manual picking. The cost of the trellising system calculated per 1 ha was 2.0 times higher than that of the standard system. [ABSTRACT FROM AUTHOR]

• Lenticel breakdown (LB) occurring after storage was higher in apples from low humidity orchard. • Preharvest calcium application enhanced LB only in apples from specific field conditions. • Thinner peel may contribute to the higher LB. • Open lenticels at harvest is not the sole cause for LB. • Bitter pit was not reduced by calcium application. It has been demonstrated that postharvest calcium application enhanced lenticels breakdown (LB), occurring after storage. In this study, we examined whether preharvest in-orchard calcium foliar application can also enhances lenticels breakdown. Calcium foliar application was examined on fruit harvested separately from northern and southern sides of the trees in two orchards. Four in-orchard treatments were applied: No treatment (Control), CaCl 2 , Ca(NO 3) 2 and KCl. The orchards were located in regions with different climatic conditions. LB damage was higher in the low humidity orchard in comparison to that in the higher humidity orchard. Moreover, LB was enhanced by preharvest calcium application only in apples from the northern side of the trees, located in the lower relative humidity orchard. Preharvest calcium application reduced the peel thickness of apples from both tree sides in the high relative humidity orchard, and only in the northern side of the low humidity orchard. Hence, in-orchard calcium foliar application can enhance LB only under specific field climate conditions. Reduced peel thickness by calcium in the northern side of the low relative humidity orchard, might have led to a higher post storage LB. Nevertheless, the status of open lenticels at harvest can only partially explain the postharvest LB. In addition, the force needed to puncture the peel does not explain LB results. In conclusion, this study supports the finding that calcium increases LB, but only under certain environmental conditions. [ABSTRACT FROM AUTHOR]

• First methodology based on computer vision for olive yield estimation. • NN fed with descriptors specially designed to visually characterise fruit yield. • Yield is modelled from the amount of visible fruit, its dispersion and aggregation. • A yield overestimation of 2.64% was registered, which is valuable for the industry. • The results validate computer vision as tool for olive-fruit yield estimation. Accurate and early olive-fruit yield estimation is a greatly desired objective in oliviculture as a tool to increase profitability and sustainability of the exploitations. This paper presents the design and testing of a novel methodology that, unlike previous proposals using models fed with indirect variables, exploits computer vision techniques to estimate yield from visual features of existing visible fruit, as traditionally done by field experts. The design comprised a neural network fed with 16 descriptors, 8 calculated per canopy face, including the number of visible fruits (1 descriptor), the area of exposed fruit (1 descriptor), and other analytical descriptors aimed at mathematically modelling fruit dispersion (2 descriptors) and aggregation (4 descriptors). The methodology was trained on a set of 37 sample points (74 images), and externally validated on a set of 10 (20 images), manually taken in a super-intensive olive orchard of the Picual Olea europaea L. variety located in Elvas (Portugal), two months prior to fruit harvesting. The total actual yield manually measured corresponding to the external validation set was 173.23 k g , providing the methodology an estimated yield of 177.80 k g , what implied a root-mean-square-error of 0.9914 kg per sample point, and an overestimation of 2.64 %. [ABSTRACT FROM AUTHOR]

This article provides an overview of preharvest factors that affect peach fruit quality attributes. Improvement of peach fruit quality is impossible postharvest. Hence, optimum peach quality at harvest and during postharvest and subsequently, consumer satisfaction, is achievable through understanding the influence of preharvest environmental and orchard factors. ‘Quality’ definition for producers, packers, shippers and consumers is discussed, with a description of the most important peach quality attributes and the physiological mechanisms that affect them. The effect of cultivar, rootstock, harvest time, crop load management, light manipulation, fruit position in the canopy, irrigation, fertilization and the impact of the growing environment on peach and nectarine harvest quality are reviewed. The development of new technologies that help clarify the biological and horticultural bases of preharvest factors on peach fruit quality can help increase peach consumption. [ABSTRACT FROM AUTHOR]

Since 2013, the Univeristy of Agronomic Sciences and Veterinary Medicine of Bucharest yearly organizes at the Istriţa Nursery and Didactic-Experimental Farm the "Sweet Cherry Fest" in the first decade of June. The event is part of the "Agriculture for Life, Life for Agriculture" International Conference programme and consist in a large sweet cherry exhibition and tasting of fresh sweet cherry fruits gathered from own trial orchards or from comercial sweet cherry plantations within the country. Depending on the environmental conditions and weather conditions of each year, the samples of cherries for tasting session varies a lot but comprise as many as possible varieties, most of them from early to medium harvest time. In 2017, the tasting panel evaluated 24 samples coming from 17 varieties. These were as follows: 'Early red', 'Burlat', 'Celeste', 'Kordia', 'Firm red', 'Grace Star', 'New Star', 'Early Bigi', 'Giant red', 'Summit', 'Ferrovia', 'Ulster', 'Van', 'Rubin', 'Katalin', 'Skeena' and 'Lapins'. Best score was recorded by 'Early red' followed by 'New Star' and 'Giant red'. The fruit size and appearance of the 'Early red' and 'Giant red' varieties were rewarded by tasting panel with high appreciation scores. The taste was also remarked by the consumers for 'Early red' and also the aroma for 'New Star'. [ABSTRACT FROM AUTHOR]

Robot harvesting has emerged as an urgent need for apple industry due to a sharp decline in agricultural labor. In the field of harvesting robots, the use of multiple robotic arms to improve operational efficiency and promote industrial applications has gained significant attention. Despite the significant progress made in multi-arm harvesting robots in recent years, their widespread application in orchard production is hindered by insufficient efficiency of operation and the accuracy of fruit positioning. This paper focuses on the precise perception and multi-arm collaborative control issues of harvesting robots, and proposes a multi-arm apple harvesting robot system. Firstly, the paper introduces the hardware and software integration method and kinematic configuration of the robot, and presents its workspace division and asynchronous sequential operation mode. Secondly, the paper proposes a stereo vision fruit recognition and localization algorithm based on multi-task deep learning to enhance the accuracy of apple fruit positioning and a method of combining multiple perspectives to acquire a global fruit map is introduced. Finally, the paper presents a multi-arm task planning method based on the Markov game framework to optimize the target harvesting order of each arm and improve the collaboration efficiency. The effectiveness of the robot and its perception and control methods are verified through multiple field experiments in orchards. The field trials showed that the proposed vision system reduces the median locating error of the robot system by up to 44.43%; the proposed task planning algorithm can reduce the average cycle time by 33.3% compared to the heuristic-based algorithm, and time taken for optimizing task planning ranged from 1.14 s to 1.21 s; and the robot's harvest success rate varied from 71.28% to 80.45%, and the average cycle time ranged from 5.8 s to 6.7 s. • A visual recognition and localization algorithm by multiple stereo cameras. • A point-cloud-processing pipeline by a frustum-based method. • A task planning scheme based on multi-agent reinforcement learning framework. • A highly integrated four-arm apple harvesting robotic system. [ABSTRACT FROM AUTHOR]

• The response to storage atmosphere of 'Monalisa' apple was studied. • Several CO 2 partial pressures on 'Monalisa' apples were evaluated. • 'Monalisa' apples are very susceptible to high CO 2 injury. • 1-MCP has no effect on the susceptibility to high CO 2 injury. • Late harvested fruit are more prone to develop physiological disorders. This study aimed to determine the effects of harvest maturity, 1-MCP treatment and storage conditions with high CO 2 partial pressures on 'Monalisa' apples physicochemical quality and susceptibility to physiological disorders and decay during long-term storage, plus 7 d of shelf life at 22 °C. The study was composed by two experiments. In Experiment 1, fruit were harvested in one growing season (2011) at the same maturity stage and were treated or not treated with 1-MCP (1 µL L-1). In Experiment 2, fruit were harvested in two growing seasons (2019 and 2020), at two maturity stages. In both experiments, all fruit were stored under CA with four CO 2 partial pressure (0.5, 1.5, 3.0 and 4.5 kPa) and regular air (RA, standard of comparison) for 6 or 7 months at 0.8 °C, plus 7 d shelf life at 22 °C. CA was very effective on delaying fruit ripening, senescent disorders and decay incidences, regardless of the CO 2 partial pressure. However, under CA, 'Monalisa' apples were very susceptible to CO 2 injury, expressed as dark flesh browning and cavities that were exacerbated with increasing CO 2 partial pressures. Therefore, 'Monalisa' apples should be stored under CA with CO 2 no higher than 0.5 kPa. The response of 'Monalisa' apples to high CO 2 is more pronounced in late harvested fruit, which were also more prone to develop senescent flesh browning, cracking and rough skin. 1-MCP application had no effect on 'Monalisa' apple susceptibility to CO 2 damages, while it reduced fruit softening and acidity loss in both RA and CA storages. [ABSTRACT FROM AUTHOR]