Small object detection based on dynamic adaptive channel attention feature fusion

In order to solve the feature missing and different scales features semantics gaps problem caused by convolution operations in small object detection, a small object detection method based on dynamic adaptive channel attention feature fusion is proposed in this paper. Firstly, a Tri-Neck network structure is introduced to address the semantic gaps and feature deficiency in small object detection across multiple scales. Secondly, a dynamic adaptive channel attention module is proposed to enhance weak semantic features of small objects while suppressing irrelevant information. Additionally, new activation functions and intersection-over-union loss functions are designed within the dynamic adaptive channel attention module to improve channel attention representation capability. Finally, the ResNet50 backbone network is utilized, connecting the feature pyramid network and the Tri-Neck network sequentially. Experimental results on the Pascal VOC 2007 and Pascal VOC 2012 datasets demonstrate performance improvements of 5.3% and 6.2% respectively, while on the MS COCO 2017 dataset, the proposed algorithm shows enhancements in overall performance and small object detection performance by 1.6% and 2% respectively, and on the SODA-D dataset, our proposed algorithm demonstrates superior performance compared to the suboptimal algorithm AP, resulting in a 0.9% improvement in overall accuracy.