Litchi is often harvested by clamping and cutting the branches, which are small and can easily be damaged by the picking robot. Therefore, the detection of litchi branches is particularly significant. In this paper, an fully convolutional neural network-based semantic segmentation algorithm is proposed to semantically segment the litchi branches. First, the DeepLabV3+ semantic segmentation model is combined with the Xception depth separable convolution feature. Second, transfer learning and data enhancement are used to accelerate the convergence and improve the robustness of the model. Third, a coding and a decoding structure are adopted to reduce the number of network parameters. The decoding structure uses upsampling and the shallow features to fuse, and the same weight is assigned to ensure that the shallow feature semantics and the deep feature semantics are evenly distributed. Fourth, using atrous spatial pyramid pooling, we can better extract the semantic pixel position information without increasing the number of weight parameters. Finally, different sizes of hole convolution are used to ensure the prediction accuracy of small targets. Experiment results demonstrated that the DeepLabV3+ model using the Xception_65 feature extraction network obtained the best results, achieving a mean intersection over union (MIoU) of 0.765, which is 0.144 higher than the MIoU of 0.621 of the original DeepLabV3+ model. Meanwhile, the DeepLabV3+ model using the Xception_65 network has greater robustness, far exceeding the PSPNet_101 and ICNet in detection accuracy. The aforementioned results indicated that the proposed model produced better detection results. It can provide powerful technical support for the gripper picking robot to find fruit branches and provide a new solution for the problem of aim detection and recognition in agricultural automation.

It is a challenging problem to detect the apple in natural environment using traditional object recognition algorithms due to occlusion, fluctuating illumination and complex backgrounds. Deep learning methods for object detection make impressive progress, which can automatically extract the number, pixel position, size and other features of apples from the images. In this paper, four deep learning recognition models, Faster RCNN based on AlexNet, Faster RCNN based on ResNet101, YOLOv3 based on DarkNet53 and improved YOLOv3 were employed to carry out recognition experiments on red and green apple under three illumination and two image sharpness conditions, with the transfer learning to accelerate the training process. The results showed that improved YOLOv3 model had the best recognition effect among the four detection models. F1 value of red apple recognition was 95.0%, 94.6% and 94.1% for normal, insufficient and excessive illumination, respectively, and F1 value of green apple recognition was 94.9%, 94.0% and 91.1%. There were F1 value of 92.8% and 92.1% for red and green apple recognition in blurred images, respectively. Moreover, improved YOLOv3 algorithm still had the better performance for occlusion, spot, overlap and incomplete apples, with a recognition recall rate higher than 88.5%. It can be concluded that improved YOLOv3 algorithm can provide a more efficient way for apple detection in natural environment.

This paper proposes PestNet, a lightweight method for classifying crop pests, which improves upon MobileNet-V2 to address the high model complexity and low classification accuracy commonly found in pest classification research. Firstly, the training phase employs the AdamW optimizer and mixup data augmentation techniques to enhance the model’s convergence and generalization capabilities. Secondly, the Adaptive Spatial Group-Wise Enhanced (ASGE) attention mechanism is introduced and integrated into the inverted residual blocks of the MobileNet-V2 model, boosting the model’s ability to extract both local and global pest information. Additionally, a dual-branch feature fusion module is developed using convolutional kernels of varying sizes to enhance classification performance for pests of different scales under real-world conditions. Lastly, the model’s activation function and overall architecture are optimized to reduce complexity. Experimental results on a proprietary pest dataset show that PestNet achieves classification accuracy and an F1 score of 87.62% and 86.90%, respectively, marking improvements of 4.20 percentage points and 5.86 percentage points over the baseline model. Moreover, PestNet’s parameter count and floating-point operations are reduced by 14.10% and 37.50%, respectively, compared to the baseline model. When compared with ResNet-50, MobileNet V3-Large, and EfficientNet-B1, PestNet offers superior parameter efficiency and floating-point operation requirements, as well as improved pest classification accuracy.