YOLO SSPD: a small target cotton boll detection model during the boll-spitting period based on space-to-depth convolution

作者：Mengli Zhang, Wei Chen, Pan Gao, Yong‐Quan Li, Fei Tan, Yuan Zhang, Shiwei Ruan, Peng Xing, Li Guo · 发表于：Frontiers in Plant Science · 年份：2024 · DOI：10.3389/fpls.2024.1409194 · 被引用次数：16 · 研究领域：Smart Agriculture and AI、Remote Sensing in Agriculture、Plant Surface Properties and Treatments

Introduction: Cotton yield estimation is crucial in the agricultural process, where the accuracy of boll detection during the flocculation period significantly influences yield estimations in cotton fields. Unmanned Aerial Vehicles (UAVs) are frequently employed for plant detection and counting due to their cost-effectiveness and adaptability. Methods: Addressing the challenges of small target cotton bolls and low resolution of UAVs, this paper introduces a method based on the YOLO v8 framework for transfer learning, named YOLO small-scale pyramid depth-aware detection (SSPD). The method combines space-to-depth and non-strided convolution (SPD-Conv) and a small target detector head, and also integrates a simple, parameter-free attentional mechanism (SimAM) that significantly improves target boll detection accuracy. Results: ) of 0.86, with a root mean square error (RMSE) of 12.38 and a relative root mean square error (RRMSE) of 11.19% for boll counts. Discussion: The findings indicate that YOLO SSPD can significantly improve the accuracy of cotton boll detection on UAV imagery, thereby supporting the cotton production process. This method offers a robust solution for high-precision cotton monitoring, enhancing the reliability of cotton yield estimates.