To enhance the efficiency of mechanical automatic picking of cherry tomatoes in a precision agriculture environment, this study proposes an improved target detection algorithm based on YOLOv5n. The improvement steps are as follows: First, the K-means++ clustering algorithm is utilized to update the scale and aspect ratio of the anchor box, adapting it to the shape characteristics of cherry tomatoes. Secondly, the coordinate attention (CA) mechanism is introduced to expand the receptive field range and reduce interference from branches, dead leaves, and other backgrounds in the recognition of cherry tomato maturity. Next, the traditional loss function is replaced by the bounding box regression loss with dynamic focusing mechanism (WIoU) loss function. The outlier degree and dynamic nonmonotonic focusing mechanism are introduced to address the boundary box regression balance problem between high-quality and low-quality data. This research employs a self-built cherry tomato dataset to train the target detection algorithms before and after the improvements. Comparative experiments are conducted with YOLO series algorithms. The experimental results indicate that the improved model has achieved a 1.4% increase in both precision and recall compared to the previous model. It achieves an average accuracy mAP of 95.2%, an average detection time of 5.3 ms, and a weight file size of only 4.4 MB. These results demonstrate that the model fulfills the requirements for real-time detection and lightweight applications. It is highly suitable for deployment in embedded systems and mobile devices. The improved model presented in this paper enables real-time target recognition and maturity detection for cherry tomatoes. It provides rapid and accurate target recognition guidance for achieving mechanical automatic picking of cherry tomatoes.

The automated harvesting of clustered fruits relies on fast and accurate visual perception. However, the obscured stem diameters via leaf occlusion lack any discernible texture patterns. Nevertheless, our human visual system can often judge the position of harvesting points. Inspired by this, the aim of this paper is to address this issue by leveraging the morphology and the distribution of fruit contour gradient directions. Firstly, this paper proposes the calculation of fruit normal vectors using edge computation and gradient direction distribution. The research results demonstrate a significant mathematical relationship between the contour edge gradient and its inclination angle, but the experiments show that the standard error projected onto the Y-axis is smaller, which is evidently more conducive to distinguishing the gradient distribution. Secondly, for the front view of occluded lychee clusters, a fully convolutional, feature prototype-based one-stage instance segmentation network is proposed, named the lychee picking point prediction network (LP³Net). This network can achieve high accuracy and real-time instance segmentation, as well as for occluded and overlapping fruits. Finally, the experimental results show that the LP³Net based on this study, along with lychee phenotypic features, achieves an average location accuracy reaching 82%, significantly improving the precision of harvesting point localization for lychee clusters.

Obtaining consistent multi-view images of litchis is crucial for various litchi-related studies, such as data augmentation and 3D reconstruction. This paper proposes a two-stage model that integrates the Mask2Former semantic segmentation network with the Wonder3D multi-view generation network. This integration aims to accurately segment and extract litchis from complex backgrounds and generate consistent multi-view images of previously unseen litchis. In the first stage, the Mask2Former model is utilized to predict litchi masks, enabling the extraction of litchis from complex backgrounds. To further enhance the accuracy of litchi branch extraction, we propose a novel method that combines the predicted masks with morphological operations and the HSV color space. This approach ensures accurate extraction of litchi branches even when the semantic segmentation model’s prediction accuracy is not high. In the second stage, the segmented and extracted litchi images are passed as input into the Wonder3D network to generate multi-view of the litchis. After comparing different semantic segmentation and multi-view synthesis networks, the Mask2Former and Wonder3D networks demonstrated the best performance. The Mask2Former network achieved a mean Intersection over Union (mIoU) of 79.79% and a mean pixel accuracy (mPA) of 85.82%. The Wonder3D network achieved a peak signal-to-noise ratio (PSNR) of 18.89 dB, a structural similarity index (SSIM) of 0.8199, and a learned perceptual image patch similarity (LPIPS) of 0.114. Combining the Mask2Former model with the Wonder3D network resulted in an increase in PSNR and SSIM scores by 0.21 dB and 0.0121, respectively, and a decrease in LPIPS by 0.064 compared to using the Wonder3D model alone. Therefore, the proposed two-stage model effectively achieves automatic extraction and multi-view generation of litchis with high accuracy.