摘要
驾驶员预警系统对提高行车安全具有重要作用,而驾驶员人脸图像识别处理是其中的关键部分之一。该文首先介绍了驾驶员预警系统的原理结构,然后讨论了驾驶员预警系统中人脸图像识别预处理的滤波和细化两个环节。通过使用卡尔曼滤波算法对动态噪声为有色噪声的图像进行滤波,既实现了图像去噪,也实现了目标跟踪;使用脱壳算法对图像进行细化得到图像的线条描述,为驾驶员人脸图像识别的后续处理提供了可靠的保证。实践证明,论文中采用的预处理方法实时性高,效果好。
Driver Forewarning System is most important for improving vehicle traveling safety, and driver＇s face image recognition is the key part of the Driver Forewarning System. The principle framework of Driver Forewarning System is introduced firstly. Then two sections ofpre - processing,namely the filtering and making thinner are discussed. By using KARLMAN filter algorithm, the colorful noise is filtered, and the target tracking is realized at the same time. By using elimination shell algorithm to make the image thin, the line description is made. These work provides a credible guarantee for suceedent driver＇s face image recognition, and experiment shows that the method described in this paper is with real time and good effect iveness.
出处
《计算机仿真》
CSCD
2006年第6期252-254,共3页
Computer Simulation
关键词
驾驶员
预警
卡尔曼滤波
人脸识别
Driver
Forewarning
Karlman filter
Face recognization
