Enhancing Plant Disease Detection in Agriculture Through YOLOv6 Integration with Convolutional Block Attention Module

Plant diseases pose a significant threat to global agriculture, resulting in substantial crop losses each year. To address the pressing need for rapid and accurate disease identification, this study explores the efficacy of an enhanced YOLOv6 model incorporating the Convolutional Block Attention Module (CBAM) for detecting plant diseases, utilizing an open-source dataset. Encompassing a diverse range of healthy and infected plant leaves across various disease categories and crop types, our research aims to enhance disease identification in agriculture. Our evaluation of various YOLO models revealed that the augmented YOLOv6 integrated with CBAM outperforms traditional and other YOLO variants in plant disease identification. This model achieves exceptional accuracy, highlighting the importance of attention mechanisms in boosting the diagnostic accuracy of deep learning models. Such an enhancement is pivotal for developing real-time detection tools, empowering farmers to quickly identify and tackle crop diseases, thereby potentially reducing agricultural losses. This breakthrough demonstrates the augmented YOLOv6 model with CBAM as the most effective approach among the YOLO series for plant disease detection, underscoring its significance in advancing agricultural diagnostics and management. The conducted experiments validate the advantages of integrating YOLOv6 with attention mechanisms such as CBAM in the context of agricultural disease detection. This represents a significant stride towards the development of more accurate, rapid, and reliable diagnostic tools, enhancing crop management and protection strategies.