Pathogen infection has greatly reduced crop production. As the symptoms of diseases usually appear when the plants are infected severely, rapid identification approaches are required to monitor plant diseases at early the infection stage and optimize control strategies. Hyperspectral imaging, as a fast and nondestructive sensing technology, has achieved remarkable results in plant disease identification. Various models have been developed for disease identification in different plants such as arable crops, vegetables, fruit trees, etc. In these models, important algorithms, such as the vegetation index and machine learning classification and methods have played significant roles in the detection and early warning of disease. In this paper, the principle of hyperspectral imaging technology and common spectral characteristics of plant disease symptoms are discussed. We reviewed the impact mechanism of pathogen infection on the photo response and spectrum features of the plants, the data processing tools and algorithms of the hyperspectral information of pathogen-infected plants, and the application prospect of hyperspectral imaging technology for the identification of plant diseases.

Recent progress and future prospects for mechanized harvesting of fruit crops with shaking systems

Highlights A novel real-time image semantic segmentation network for orchards, termed AFC-ResNet18, was designed and tested. The AFC-ResNet18 model outperformed the SwiftNet network in terms of segmentation depth. The AFC-ResNet18 model achieved the highest accuracy in the architecture performance testing. The AFC-ResNet18 model won first place in the orchard scene test with 72.5% accuracy. Abstract. Semantic segmentation is a fundamental prerequisite for the real-time understanding of scenes. This understanding is essential for developing automated devices that can enhance productivity. Orchards, being labor-intensive and time-consuming workplaces, urgently require automated equipment to boost efficiency. Therefore, the objective of this article is to develop a real-time image semantic segmentation network tailored for orchard environments. This development aims to offer significant new insights into the design of automated maintenance and harvesting equipment. Based on ResNet, the 2015 classification champion network, a novel real-time image semantic segmentation network termed AFC-ResNet18, which used an attentional feature complementary module (AFC) to fuse RGB and depth image information, was designed and systematically tested. Interestingly, in the segmentation ability tests, the AFC-ResNet18 model outperformed the SwiftNet network in terms of segmentation depth. Surprisingly, in the architecture performance testing, the AFC-ResNet18 model achieved the highest accuracy. Noteworthily, in the orchard scene test, the AFC-ResNet18 model won first place with 72.5% accuracy. Predictably, these findings may accelerate the development of novel automated equipment to maintain the orchard worldwide, especially AFC-ResNet18 based robots. Keywords: Attentional feature complementary module, Orchard, Real-time, Robots, Semantic segmentation.

To investigate the variation patterns of spectral characteristics of main branches in fruit trees due to uneven distribution of growth, pruning, and fruit quality, as well as to determine the optimal fruit vi-bration harvesting, a vibration differential model of tree branches was constructed using the concen-trated mass method. A three-dimensional model of the fruit tree was built using finite element software. The spectral characteristics of each branch were obtained using the linear sweep frequency method. The method of changing the mass of each branch was employed to simulate the variation in branch mass. The frequency of occurrence, mean acceleration, and coefficient of variation of the inherent frequencies were used as evaluation indicators. The results revealed that when the vibration frequency of the fruit tree was at the peak frequency in the tree trunk spectrum, the frequency of occurrence and mean accel-eration were optimal. And the coefficient of variation of branch acceleration was less than 1, indicating a moderate dispersion. Additionally, reducing the branch mass resulted in an increase in the peak fre-quency of the tree trunk spectrum. Therefore, the growth, pruning, and uneven distribution of the branches in fruit trees can alter the spectral characteristics. When harvesting fruit by vibration, selecting the peak frequency in the trunk vibration spectrum as the excitation force frequency yields the best fruit detachment.

The multi-scale model is key to the evaluation of shear or compression mechanical properties of jujube branches at the microlevel. In this study, multi-scale models of the microstructure of jujube branches, namely, the natural model and the artificially simplified model, were built using computed tomography and Python. The equivalent compression and shear properties of the microstructure were predicted by the finite element method. The prediction results of the artificially simplified model were compared with those of the natural model, and the prediction deviation of the artificially simplified model was analyzed. The results showed that the transverse shear modulus obtained by the artificially simplified model was slightly higher than the predicted value of the natural model, with an average deviation of 1.84%. The transverse shear strain energy obtained by the artificially simplified model was lower than the predicted result of the natural model, with a maximum deviation of 5.59%. The cell distribution direction had an important effect on the accuracy of prediction for the transverse elastic modulus, and the deviation was 16.53%, indicating that the further optimization of the cell distribution direction could enhance the prediction accuracy and the effectiveness of the model. The results can be used to evaluate the unit compression or shear mechanical properties of plant stalks with hierarchical structuring and to analyze the quantitative relationship between the microstructure and equivalent mechanical properties.

Accurate determination of soil and contact parameters is crucial for tillage machinery design; however, the interactions among soil, tools, and roots in citrus orchards covered with green manure remain insufficiently defined. This study, therefore, combined physical experiments with DEM simulations to characterize these interactions. Using significance analysis and response surface methodology (RSM), the effects of major factors on angle of repose (AoR) and initial slip angle (ISA) at varying soil depths were evaluated, enabling precise calibration of both external (soil–machinery) and internal (particle–particle) parameters. Subsequently, a GA-BP optimization model was constructed to enhance calibration accuracy, yielding optimal values for the soil-to-soil rolling friction coefficient (γ = 0.125–0.136), soil-to-65Mn static friction coefficient (μ′ = 0.431 − 0.540), and soil surface energy (JKR = 0.952 − 1.091 J·m⁻²). Shear tests using the bonding V2 model were conducted to calibrate the Bonding parameters of green manure stems and roots, while pull-out tests and simulations were used to validate the root–soil parameters. Direct shear tests confirmed the model’s reliability, with errors in internal friction angle and cohesion below 10%. These findings may contribute to improving DEM simulation accuracy for soil improvement under green manure coverage and support the optimization of soil tillage in citrus orchards.