基于度量学习的跨摄像头运动目标重定位方法研究  

Transversal cameras relocation for moving object based on metric learning

作　　者：康宇 史珂豪[3] 陈佳艺 曹洋 许镇义 KANG Yu;SHI Ke-hao;CHEN Jia-yi;CAO Yang;XU Zhen-yi(Institute of Artificial Intelligence,Hefei Comprehensive National Science Center,Hefei 230088;Anhui Engineering Research Center for Intelligent Applications and Security of Industrial Internet,Ma’anshan 243032;School of Information Science and Technology,University of Science and Technology of China,Hefei 230026,China)

机构地区：[1]合肥综合性国家科学中心人工智能研究院,安徽合肥230088 [2]安徽省工业互联网智能应用与安全工程研究中心,安徽马鞍山243023 [3]中国科学技术大学信息科学技术学院,安徽合肥230026

出　　处：《计算机工程与科学》2024年第7期1256-1268,共13页Computer Engineering & Science

基　　金：国家自然科学基金(62033012,62103124);安徽省重大科技专项(202003a07020009);宿迁学院京东学院开放基金(2022JDXM14);安徽省工业互联网智能应用与安全工程研究中心开放基金(IASII22-03)。

摘　　要：近年来,我国柴油车尾气排放污染日趋严重。为了改善大气环境,需要对排放黑烟的柴油车进行监测。然而,在城市交通道路场景下,黑烟柴油车检测经常由于车辆间相互遮挡等因素,难以通过后向视频确定黑烟柴油车身份。此外,柴油车重定位相关数据的严重不足导致数据局限性较大。针对以上问题,提出了一种跨摄像头场景下的黑烟柴油车重定位方法。该方法通过引入IBN模块构建特征提取网络,提升网络模型对柴油车图像外观变化的适应性。然后,设计基于豪斯多夫距离度量学习的损失函数对特征差异性进行度量,在优化过程中增加类间距离并降低遮挡样本的影响。最后,构建了多种场景下的柴油车重定位基准数据集,并在该数据集上对所提出的方法进行实验。实验结果表明,所提出的方法取得了83.79%的相对精度,具有较高准确率。In recent years,the pollution from diesel vehicle exhaust emissions in China has become increasingly severe.In order to improve the atmospheric environment,it is necessary to monitor diesel vehicles emitting black smoke.However,in urban traffic road scenarios,the detection of black smoke vehicles is often difficult to determine through rear-view videos due to factors such as mutual obstruction between vehicles.Additionally,the severe lack of relevant data greatly limits the effectiveness of the data.To address the above problems,this paper proposes a black smoke diesel vehicle re-identification model under the cross-camera scene.By introducing the IBN module to construct a feature extraction network,the adaptability of the network model to changes in the appearance of diesel vehicle images is enhanced.A loss function based on the Hausdorff distance metric learning is designed to measure the feature differences,increasing inter-class distance and reducing the impact of occluded samples during the optimization process.Then,benchmark datasets for diesel vehicle repositioning across multiple scenarios are constructed,and the proposed method is experimented on this dataset.The experimental results show that the proposed method achieves a relative accuracy of 83.79%,demonstrating high accuracy.

关 键 词：跨摄像头 黑烟车重定位 度量学习 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]