摘要
近年来,由于鱼眼相机被广泛应用于智能监控领域,不少学者提出了针对鱼眼图像的行人检测算法。然而,鱼眼图像场景复杂且存在畸变,其导致的数据集样本分布和算法监督分配的不均衡问题会降低检测器性能。针对上述问题,首先提出了一种针对鱼眼图像行人检测任务的数据增强方法,该方法由模式采样增强和角度直方图增强两部分组成。其中模式采样增强专注于鱼眼图像难例样本挖掘,生成的新样本丰富了鱼眼图像中心附近的行人模式;角度直方图增强基于直方图均衡的思想,对鱼眼图像样本角度分布做平滑处理,缓解了检测器对单一场景的过拟合问题。此外,基于鱼眼图像Anchor-Free行人检测器,提出将定位质量预测与监督信息融合,将Focal Loss推广到连续域用以优化检测器定位分支的监督分配。实验结果表明,所提数据增强算法能够有效缓解鱼眼图像数据集的分布不均衡,在Anchor-Based和Anchor-Free检测网络上均展现了较好的效果;连续Focal Loss结合定位质量优化监督,在不增加Anchor-Free检测器推理计算复杂度的前提下,将整体性能提升了3.8%。
In recent years,due to the fisheye camera is widely used in the field of intelligent monitoring,many scholars propose pedestrian detection algorithm for fisheye image.However,the fisheye scene is complex and distorted,which leads to the imbalance of data set sample distribution and algorithm supervision allocation,which will reduce the performance of the detector.To solve these problems,a data enhancement method for pedestrian detection in fisheye image is proposed,which consists of pattern sampling enhancement and angle histogram enhancement.The pattern sampling enhancement focuses on the mining of difficult samples in fisheye image,and the generated new samples enrich the pedestrian patterns near the center of fisheye image.Angle histogram enhancement is based on the idea of histogram equalization,which smooths the angle distribution of fisheye image samples to alleviate the over fitting problem of single scene.In addition,based on Anchor-free fisheye image pedestrian detector,the fusion of location quality prediction and supervision information is proposed to extend Focal Loss to continuous domain to optimize the supervision allocation of detector location branches.Experimental results show that the proposed data enhancement algorithm can effectively alleviate the uneven distribution of fisheye image data set,and show good results in both Anchor-Based and Anchor-Free detectors.The continuous Focal Loss optimization localization supervision method improves the overall performance by 3.8%without increasing the reasoning complexity of the Anchor-Free detector.
作者
司绍峰
张赛强
李庆
陈本瑶
SI Shao-feng;ZHANG Sai-qiang;LI Qing;CHEN Ben-yao(University of Chinese Academy of Sciences,Beijing 100049,China;Institute of Microelectronics of the Chinese Academy of Sciences,Beijing 100029,China;Huzhou Special Equipment Inspection Center,Huzhou,Zhejiang 313000,China)
出处
《计算机科学》
CSCD
北大核心
2022年第S02期371-376,共6页
Computer Science
基金
基于电梯安全一体化管理与应用技术的研究(NQI20200125)
作者简介
司绍峰,1029865870@qq.com,born in 1996,postgraduate.His main research interests include computer vision and object tracking;通信作者:李庆,liqing@ime.ac.cn,born in 1972,Ph.D,professor,Ph.D supervisor.His main research interests include artificial intelligence and multi-source information fusion
