To achieve accurate detection of tomato fruit maturity and enable automated harvesting in natural environments, this paper presents a more lightweight and efficient maturity detection algorithm, YOLO-DGS, addressing the challenges of subtle maturity differences between regular and cherry tomatoes, as well as fruit occlusion. First, to enhance feature extraction at various levels of abstraction in the input data, this paper proposes a novel segment-wise convolution module, C2f-GB. This module performs convolution in stages on the feature map, generating more feature maps with fewer parameters and computational resources, thereby improving the model’s feature extraction capability while reducing parameter count and computational cost. Next, based on the YOLO v10 algorithm, this paper removes redundant detection layers to enhance the model’s ability to capture specific features and further reduce the number of parameters. This paper then integrates a bidirectional feature pyramid network (BiFPN) into the neck network to improve feature capture across different scales, enhancing the model’s ability to handle objects of varying sizes and complexities. Finally, we introduce a novel channel attention mechanism that allows the network to dynamically adjust its focus on channels, efficiently utilizing available information. Experimental results demonstrate that the improved YOLO-DGS model achieves a 2.6% increase in F1 score, 2.1% in recall, 2% in mAP50, and 1% in mAP50-95. Additionally, inference speed is improved by 12.5%, and the number of parameters is reduced by 26.3%. Compared to current mainstream lightweight object detection models, YOLO-DGS outperforms them, offering an efficient solution for the tomato harvesting robot vision system in natural environments.

This paper proposes a lightweight fire detection algorithm, YOLO-LRP, for UAV firefighting scenarios. The algorithm is improved based on YOLOv10, where the MobileOne reparameterized backbone is adopted to reduce computational complexity. The OREPA module is integrated into the neck network to enhance training efficiency and feature fusion. The standard CBS module is replaced with GhostConv to further compress the model size, and the number of detection layers is streamlined to focus on small and medium fire targets. In addition, the SE channel attention mechanism is introduced to strengthen the extraction of key fire features under complex interference. Experimental results show that YOLO-LRP achieves 93.6% mAP50, only 2.26 M parameters, and 434.7 FPS in server-side tests. These improvements make the model more suitable for deployment on resource-constrained UAV platforms.