基于图像序列的运动目标检测识别的关键技术研究

doi:10.3969/j.issn.1000-1158.2020.12.05

摘要
图/表
参考文献(18)
相关文章 (15)

全文: PDF (2257 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为提高运动目标检测的识别效果,通过分析、综合比较各种运动目标检测算法的优劣性,提出了基于全局自适应帧差法和基于码本模型的背景减除法对同一运动目标进行检测。通过对运动目标检测提取运动目标的掩膜,对掩膜进行外接矩形分析,从而得到包围运动目标的矩形框;将矩形框内的图片截取出来,调整该矩形并提取图片的HOG特征,最后通过训练好的SVM进行分类。在训练过程中,针对难易情况应用自举法对训练器进行优化。实验表明,与传统HOG+SVM多尺度检测算法相比,该方法在速度和准确性上可提升20%左右,可作为运动目标检测与识别的参考方法。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	薛震
	于莲芝
	胡婵娟

关键词 ：计量学, 运动目标检测, 帧差法, 码本模型, HOG特征, 支持向量机

Abstract：To improve the recognition of moving target detection, the pros and cons of various moving target detection algorithms were analyzed, the global adaptive frame difference method and the codebook model based background subtraction method was proposed to detect moving targets simultaneously for the same target.The method used the masks of moving target and then analyzed them by rectangles, so that the pictures in the rectangle can be extracted out.Next, adjust the rectangle and extract the HOG features of the image, classify through trained SVM.In the training process, the bootstrap method is applied to optimize the trainer for difficult situations.Experiments show that compared with the traditional HOG+SVM multiscale detection algorithm, this method can increase the speed and accuracy by about 20%, and it can be used as a reference method for moving target detection and recognition.

Key words： metrology moving target detection frame difference method codebook HOG feature SVM

收稿日期: 2019-04-15 发布日期: 2020-12-08

PACS:

TB96

基金资助:国家自然科学基金(61603257)

通讯作者: 于莲芝(1966-),辽宁平县人,上海理工大学副教授,从事机器学习、模式识别等方面研究。Email:yulianzhi@163.com E-mail: 953211947@qq.com

作者简介: 薛震(1994-),黑龙江北安人,上海理工大学硕士研究生,研究方向为计算机视觉、模式识别、机器学习。Email:953211947@qq.com

引用本文:

薛震,于莲芝,胡婵娟. 基于图像序列的运动目标检测识别的关键技术研究[J]. 计量学报, 2020, 41(12): 1475-1481.
XUE Zhen,YU Lian-zhi,HU Chan-juan. Research on Key Techniques of Moving Target Detection and Recognition Based on Image Sequence. Acta Metrologica Sinica, 2020, 41(12): 1475-1481.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2020.12.05 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2020/V41/I12/1475

［1］程全, 樊宇. 分块投影匹配的运动目标检测方法［J］. 红外与激光工程, 2018, 47(10): 1-4.
Cheng Q, Fan Y. Moving Target Detection Method Based on Block Projection Matching［J］. Infrared and Laser Engineering, 2018, 47(10): 1-4.
［2］Jain R. Difference and Accumulative Difference Pictures in Dynamic Scene Analysis［J］. Image and Vision Computing, 1984, 2(2): 99-108.
［3］崔杰, 胡长青, 徐海东. 基于帧差法的多波束前视声纳运动目标检测［J］. 仪器仪表学报, 2018, 39(2): 170-175.
Cui J, Hu C Q, Xu H D. Moving Target Detection for Multi-Beam Forward-Looking Sonar Based on Frame-Difference Method［J］. Chinese Journal of Scientific Instrument, 2018, 39(2): 170-175.
［4］王华, 李艾华, 崔智高,等. 复杂背景下改进视觉背景的前景检测算法［J］. 计算机应用研究, 2017, 34(4): 1261-1264.
Wang H, Li A H, Cui Z G, et al. Foreground Detection Algorithm Based on Improved Visual Background Extrac-tor in Complex Background［J］. Application Research of Computers. 2017, 34(4): 1261-1264.
［5］Goh T Y, Basah S N, Yazid H, et al. Performance anal-ysis of image thresholding: Otsu technique［J］. Measurement, 2018, 114(10): 298-307.
［6］孙亚萍. 灰度图像的直方图均衡化处理分析［J］. 河南科技, 2015, (6): 20-22.
［7］刘仲民, 何胜皎. 基于背景减除法的视频序列运动目标检测［J］. 计算机应用, 2017, 37(6): 1777-1781.
Liu Z M, He S J. Moving Object Detection Based on Background Subtraction for Video Sequence［J］. Journal of Computer Applications, 2017, 37(6): 1777-1781.
［8］程淑红, 高许. 基于多特征提取和SVM参数优化的车型识别［J］. 计量学报, 2018, 39(3): 348-352.
Cheng S H, Gao X. Vehicle Recognition Based on Muti-feature Extraction and SVM Parameter Optimization［J］. Acta Metrologica Sinica, 2018, 39(3): 348-352.
［9］Kim K, Chalidabhongse T H, Harwood D, et al. Background modeling and subtraction by codebook construction［C］//2004 International Conference on Image Processing. Singapore, 2004.
［10］苏剑臣, 李策, 杨峰. 基于边缘帧差和高斯混合模型的行人目标检测［J］. 计算机应用研究, 2018, 35(4): 1246-1249.
Su J C, Li C, Yang F. Detection Method for Pedestrian Target Based on Gaussian Mixture Model and Edge Frame Diffidence［J］. Application Research of Computers, 2018, 35(4): 1246-1249.
［11］张立国, 李晓松. 基于单目视觉的四旋翼飞行器目标跟踪算法研究［J］. 计量学报, 2018, 39(3): 342-374.
Zhang L G,Li X S.Research on Target Tracking Algorithm of the Quadrotor Aircraft Based on Monocular Vision［J］. Acta Metrologica Sinica,2018,39(3):342-374.
［12］孙小权, 邹丽英. 基于SVM的图像识别在零件分拣系统中的应用［J］. 机电工程, 2018, 35(12): 1354-1356.
Sun X Q, Zou L Y. Application of Image Recognition Based on SVM in Part Sorting System［J］. Journal of Mechanical & Electrical Engineering, 2018, 35(12): 1354-1356.
［13］金立军, 闫书佳, 刘源. 基于类haar特征与级联AdaBoost算法的防震锤识别［J］. 系统仿真学报, 2012, 11(9): 1806-1809.
Jin L J, Yan S J, Liu Y. Vibration Damper Recognition Based on Haar-Like Features and Cascade Adaboost Classifier［J］. Journal of System Simulation, 2012, 11(9): 1806-1809.
［14］王金甲,周雅倩,郝智. 基于注意力模型的多传感器人类活动识别［J］. 计量学报, 2019, 40(6): 958-969.
Wang J J, Zhou Y Q, Hao Z. Multi-sensor Human Activity Recognition Based on Attention Model［J］. Acta Metrologica Sinica, 2019, 40(6): 958-969.
［15］田仙仙, 鲍泓, 徐成. 一种改进HOG特征的行人检测算法［J］. 计算机科学, 2014, 41(9): 320-324.
Tian X X, Bao H, Xu C. Improved HOG Algorithm of Pedestrian Detection［J］. Computer Science, 2014, 41(9): 320-324.
［16］王同, 彭祺, 屠礼芬. 基于OpenCV的运动目标检测软件实现［J］. 软件导刊, 2015, 14(12): 132-133.
Wang T, Peng Q, Tu L F. Implementation of Moving Target Detection Software Based on OpenCV［J］. Software Guide, 2015, 14(12): 132-133.
［17］周建新, 吴建军, 薛军强. 适用于密集人群目标检测的多尺度检测方法［J］. 系统仿真学报, 2016, 28(10): 2603-2509.
Zhou J X, Wu J J, Xue J Q. Multi-scale Detection Method SuitableFor Target Detection in Dense Population［J］. Journal of System Simulation, 2016, 28(10): 2603-2509.
［18］倪豪, 郑慧峰, 王月兵. 基于自动种子区域生长的超声B图像缺陷分割方法［J］. 计量学报, 2018, 39(6): 878-883.
Ni H, Zheng H F, Wang Y B. Ultrasonic B Image Defect Segmentation Method Based on Automatic Seed Region Growth［J］. Acta Metrologica Sinica, 2018, 39(6): 878-883.