Abstract
Identifying and statistically analyzing soybean pod types are crucial for seed evaluation and yield estimation. Traditional visual assessment by breeding personnel is time-consuming, labor-intensive, and prone to subjective bias, especially with large datasets. Automatic assessment methods usually struggle with the highly confusing pod types with two and three seeds, affecting the model’s identification accuracy. To address these issues, we propose the FEI-YOLO model, an improved YOLOv5s object detection model, to enhance the distinction between pod types and improve model efficiency. FasterNet and the original C3 module are integrated to reduce parameters and computational load, enhancing both detection accuracy and speed. To strengthen the feature extraction and representation for specific targets, the Efficient Multi-Scale Attention (EMA) module is incorporated into the C3 module of the backbone network, improving the identification of similar pod types. Inner-IoU is combined with CIoU as the loss function to further enhance detection accuracy and generalization. Experiments comparing FEI-YOLO with the baseline YOLOv5s show that FEI-YOLO achieves an mAP@0.5 of 98.6%, a 1.5% improvement. Meanwhile, the number of parameters is reduced by 13.2%, and FLOPs decreased by 10.8%, in demonstrating the model's effectiveness and efficiency, enabling rapid and accurate identification of soybean pod types from images.