DSW-YOLO: A detection method for ground-planted strawberry fruits under different occlusion levels

• Published in: Computers and electronics in agriculture Vol. 214; p. 108304
• Main Authors: Du, Xiaoqiang, Cheng, Hongchao, Ma, Zenghong, Lu, Wenwu, Wang, Mengxiang, Meng, Zhichao, Jiang, Chengjie, Hong, Fangwei
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.11.2023
• Subjects: Attention mechanism; Deformable convolution; Picking robot; Strawberry; YOLO; YOLO; Attention mechanism; Strawberry; Deformable convolution; Picking robot
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2023.108304

•A DSW-YOLO network model is proposed. Accurate detection of strawberry fruits and their degree of occlusion in complex environments is achieved.•The DCN-ELAN module is designed to improve the feature extraction capability of the network for non-regular targets.•The SA attention mechanism is introduced to enhance the extraction of effective feature information.•The bounding box loss function is optimized to WIoU v3 to speed up the convergence of the network. The problems of stem and leaf shading, fruit overlapping, and different ripening stages pose a great challenge to picking robots, which are usual for the ground-planted strawberries in field. To achieve accurate detection of ripe strawberries and their occlusion levels, a DSW-YOLO (DCNv3-SA-WIoU-YOLO) network model is proposed based on the YOLOv7 network model. The model is designed with a DCN-ELAN module by introducing DCNv3 (Deformable Convolution v3) in the ELAN module to improve the feature extraction capability of the network for non-regular targets. Meanwhile, the SA (Shuffle Attention) mechanism is introduced at the end of the backbone network to enhance the extraction of valid information and improve detection accuracy. Optimizing the bounding box loss function to WIoU v3 (Wise-IoU v3) accelerates the convergence speed of the network. The proposed network model was tested using the ground-planted strawberry dataset, and the experimental results show that the P, R, and mAP@.5 of the improved network model are 82.8 %, 82.1 %, and 86.7 %, respectively, which are improved by 5.0 %, 1.7 %, and 2.2 %, respectively, compared with the YOLOv7 model. By comparing with other classical target detection algorithms, it is verified that the proposed model has advantages in terms of accuracy and model size. Finally, the DSW-YOLO model was deployed on Jetson Xavier NX of a strawberry picking robot, and visual recognition tests were conducted in field. The detection time of a single image is 92 ms, which satisfies the speed requirement of robot picking, and the detection results verify that the DSW-YOLO network model can quickly and accurately detect ripe strawberry fruits under different occlusion levels in a complex field environment, which provides a theoretical basis for the picking robot to achieve accurate positioning and efficient picking.