SCA-YOLO: a new small object detection model for UAV images

• Published in: The Visual computer Vol. 40; no. 3; pp. 1787 - 1803
• Main Authors: Zeng, Shuang, Yang, Wenzhu, Jiao, Yanyan, Geng, Lei, Chen, Xinting
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2024; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial Intelligence; Computer Graphics; Computer Science; Computer vision; Datasets; Deep learning; Design; Feature extraction; Feature maps; Image Processing and Computer Vision; Modules; Multilayers; Neural networks; Object recognition; Original Article; Semantics; Stitching; Telematics; Unmanned aerial vehicles; Small object detection; Spatial and coordinate attention enhancement; Feature fusion; UAV images
• ISSN: 0178-2789; 1432-2315
• DOI: 10.1007/s00371-023-02886-y

Object detection from UAV (unmanned aerial vehicle) images is a crucial and challenging task in the field of computer vision. The task suffers from the difficulties of small dense objects, low pixel occupation of objects, and features that are not easily extracted in images. In this paper, we proposed a multilayer feature fusion algorithm named SCA-YOLO (spatial and coordinate attention enhancement YOLO) for small object detection with hybrid attention mechanisms. It uses the single-stage detection algorithm YOLOv5 as the base framework. Firstly, a hybrid attention module with associated coordinate attention is designed to enhance the feature extraction of small objects. Secondly, to address the problem that small objects are vulnerable to being disturbed by the complex background information on UAV images, an improved SEB (simple and efficient bottleneck) module is designed to further distinguish foreground and background features. Thirdly, a multilayer feature fusion structure is built to perform channel stitching of shallow and deep feature maps, as well as to enrich the semantic information of shallow features by adding horizontal jump connections. Finally, experiments are conducted on the VisDrone2020 dataset, which involves a large number of small objects photographed by drones. In addition, we also conduct extended experiments on the DOTA dataset and PASCAL VOC dataset. Comparative experimental results indicate that the proposed method considerably improves the accuracy of small object detection on multiple benchmark datasets.