计算机工程与应用2024，Vol.60Issue(14)：250-256,7.DOI:10.3778/j.issn.1002-8331.2305-0442
适用于鱼眼图像的改进YOLOv7目标检测算法
Improved YOLOv7 Object Detection Algorithm for Fisheye Images
摘要
Abstract
Images taken by fisheye cameras are characterized by wide field of view,geometric distortion and large scale variance,which bring great challenges to object detectors based on general convolutional networks.Existing object detec-tion algorithms can be further improved with respect to network structure design,feature learning to be applicable to the distorted object detection task on fisheye images.To mitigate the effect of radial distortion on fisheye images,a multi-head attention module with multi-branch stacking structure is used in the YOLOv7 backbone to capture global contextual information.Meanwhile,a simple and efficient layer aggregation structure combining deformable convolutions is used on the Neck side of YOLOv7 to achieve effective multi-scale feature fusion.Experiments are conducted on the public com-prehensive fisheye image dataset VOC_360,and the results show that the improved YOLOv7 fisheye image object detector effectively achieves detection accuracy of 84.3％and 70.4％for mAP50 and mAP50:95,respectively,which is 3.1 percentage points and 6.4 percentage points higher than the baseline model YOLOv7,respectively.关键词
目标检测/鱼眼图像/多头注意力/可变形卷积/YOLO算法Key words
object detection/fisheye image/multi-head attention/deformable convolution/YOLO algorithm分类
信息技术与安全科学引用本文复制引用
吴兆东,徐成,刘宏哲,付莹,蹇木伟..适用于鱼眼图像的改进YOLOv7目标检测算法[J].计算机工程与应用,2024,60(14):250-256,7.基金项目
国家自然科学基金(62171042,62102033,62006020) （62171042,62102033,62006020）
北京市重点科技项目(KZ202211417048) （KZ202211417048）
北京市属高等学校高水平科研创新团队建设支持计划项目(BPHR20220121) （BPHR20220121）
北京市自然科学基金(4232026) （4232026）
协同创新中心(CYXC2203). （CYXC2203）
