Visual detection of moving stacked objects based on efficient multi-scale grouping and improved multi-head self-attention

Abstract To address the issue of fruit stacking and obstructing target fruits during daily fruit sorting, this paper presents a novel object detection algorithm that leverages efficient multi-scale grouping and enhanced multi-head self-attention. The proposed target detection algorithm is based on RT-DETR as the baseline to simplify optimization challenges and enhance robustness. Additionally, we introduce EMA's efficient multi-scale attention mechanism to preserve channel information, optimize the multi-head self-attention, and adopt cascade grouping to reduce computational redundancy. Furthermore, we utilize an Inner-MPDIoU loss function combined with Inner-IoU and MPDIoU to enhance the accuracy of detecting moving occluded targets. Experimental results demonstrate that the optimized RT-DETR algorithm achieves an average accuracy of 96.3% in detecting moving stacked fruit models with a detection speed of up to 67FPS. This confirms the effectiveness of our algorithm in matching and recognizing blocked fruit targets, surpassing common algorithms for recognizing obstructed targets.