摘要
针对战场感知及侦破现场中传统人工主观经验检验与识别模式误差较大的问题,提出了一种基于人工智能的足迹识别与特征提取方法。采用三维形貌重构系统进行足迹图像采集,并将数字图像处理算法与传统足迹检验法结合,提取足迹的区域关系特征和形状长度特征,进而采用支持向量机的模式识别方法对提取的特征进行立体足迹身份鉴别对比实验。实验结果表明,所提方法准确率超过人工鉴别准确率,达到99.1%,可应用于战场感知及侦破现场足迹准确检测与识别,也可推广应用于人体身份鉴别的相关领域。
For the problem that the traditional way of recognition based on experiences has a big deviation between recognition patterns in battleground awareness and crime scene investigation,a footprint recognition and feature extraction method based on artificial intelligence is proposed.It uses a three-dimensional(3D)reconstruction system to collect footprint images,then combines digital image processing algorithms with traditional footprint inspection methods to extract regional relationship features and shape-length features of collected footprints.Further,it carries out a 3D footprint identification comparison experiment on the extracted features by using the pattern recognition method based on support vector machine.The results indicate that the accuracy rate of the proposed method reaches 99.1%,which exceeds the accuracy rate of traditional way of recognition.This method can be applied not only in battlefield awareness and accurate detection and identification of footprints on crime scene,but also in human body identification in related fields.
作者
高毅
穆治亚
张群兴
仲元昌
GAO Yi;MU Zhiya;ZHANG Qunxing;ZHONG Yuanchang(College of Criminal Science and Technology,Criminal Investigation Police University of China,Shenyang 110035,China;Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;Key Laboratory of Sichuan Higher Education Criminal Inspection,Luzhou 646000,China;The 7th Military Representative Office of Equipment Development Department in Chongqing Region,Chongqing 400060,China;School of Electrical Engineering,Chongqing University,Chongqing 400044,China;State Key Laboratory of Power Transmission Equipment&System Security and New Technology,Chongqing 400044,China)
出处
《电讯技术》
北大核心
2020年第7期739-745,共7页
Telecommunication Engineering
基金
辽宁省自然科学基金引导计划项目(20180550153)
中央高校基本科研业务费(2019CDCGTX302,2020CDCGTX055)
证据科学教育部重点实验室2019年开放基金项目(2019KFKH02)
刑事检验四川高校重点实验室开放课题(2018YB03)。
关键词
足迹识别
人工智能
模式识别
三维形貌重构
数字图像处理
footprint recognition
artificial intelligence
pattern recognition
3D reconstruction
digital image processing
