摘要
视频运动目标检测与跟踪算法是计算机视觉领域的一个核心课题,也是智能视频监控系统的关键底层技术。它融合了图像处理、人工智能等领域的研究成果,已经广泛应用于安保监控、智能武器、视频会议、视频检索等各个领域。因此,检测与跟踪算法研究具有极其重要的理论意义和实用价值。
运动目标检测与跟踪涉及到计算机图像处理、视频图像处理、模式识别、以及人工智能等诸多领域,广泛地应用于军事、工业、生活等各个方面。研究内容分为三个方面:图像的预处理、运动目标的检测和运动目标的跟踪。在图像的预处理方面,采用均值滤波,抑制噪声;并应用形态学的方法进行滤波和去除小黑点,改善图像质量。
在运动目标的检测方面在运动目标检测方面,本文对常用的三种方法进行了分析,包括帧间差分法和背景差分法,并指出其优缺点及主要的适用范围;重点研究了帧间差分法,帧间差分法比较简单,对环境的适应能力强,但是检测到的运动目标不精确。
在运动目标的跟踪方面,也做了初步的研究。采用最小外界矩形框来定位目标,借助最近领域法预测目标位置,缩小目标的搜索范围,提高目标的跟踪速度。
关键词:运动目标检测,运动目标跟踪,颜色直方图,视频监视系统
Video moving target detection and tracking algorithm is a core area of computer vision issues, but also the key to intelligent video surveillance system underlying technology. It combines image processing, artificial intelligence research, has been widely used in security monitoring, smart weapons, video conferencing, video retrieval and other fields. Therefore, the detection and tracking algorithm is extremely important theoretical and practical value.
Moving target detection and tracking related to computer image processing, video, image processing, pattern recognition and artificial intelligence fields, widely used in military, industrial, and other aspects of life. Research is divided into three areas: image preprocessing, moving target detection and tracking of moving targets. In image preprocessing, the use of filtering, noise suppression; and apply the method of morphological filtering and removal of black specks and improve image quality.
In moving target detection in motion detection, the paper three commonly used methods of analysis, including inter-frame difference and background difference method, and pointed out their advantages and disadvantages and the main scope of application; focus on the frame difference method, frame difference method is relatively simple, adaptable to the environment, but the detected moving target imprecise.
Tracking the moving target, but also made a preliminary study. Rectangular box with the outside world to locate the minimum target, with the most recent method to predict the target location areas and narrow the search target to improve the tracking speed.
EYWORDS:Moving target detection,Moving target tracking,Color histogram,Video surveillance systems