人脸检测与跟踪算法优化及Web端实现

摘要

In order to solve the problems about release,expansion and maintenance in the using process of system software which realization face detection and tracking,HTML5-based face detection and tracking is proposed.By combining HTML5 Canvas technology with computer vision algorithms library,using a face feature classifier based on Viola-Jones algorithm and a variety of programming languages such as JavaScript,it implements HTML5-based face detection and tracking function on the page.Experimental results show that the method not only can effectively reduce the cost of resources,support the implementation of face detection and tracking algorithm on Web,but also meet the real-time requirements of face detection and tracking,and ensure the detection rate of face detection,with better detection effect.%针对实现人脸检测与跟踪功能的系统软件在使用过程中产生的不易发布、扩展和维护等问题,提出一种基于HTML5实现人脸检测与跟踪的方法.通过将HTML5 Canvas技术和一种计算机视觉算法库相结合,利用基于Viola-Jones算法原理的人脸特征分类器和JavaScript等多种程序语言,在网页上实现基于HTML5的人脸检测与跟踪的功能.实验结果表明,该方法不仅能有效地减少资源的开销,支持人脸检测与跟踪算法在Web端的实现,而且还满足了入脸检测和跟踪过程中的实时性要求,保证了人脸检测的检测率,具有更优的检测效果.