基于车脸识别的套牌车检测方法分析-analysis on detection method of fake cars based on car face recognition.docxVIP

优秀毕业论文 精品参考文献资料 摘要 套牌车的出现严重扰乱了道路交通安全秩序，直接侵害了合法车主的权益， 给社会带来了极大的危害。然而，套牌车的查处主要依靠交警盘查和群众举报， 效率低下，收效甚微。以此为背景，本文提出了一种基于车脸识别的套牌车检测 方法。 本文研究的套牌车检测方法主要包括四个部分：车辆图像预处理、车牌定位 与识别、车脸定位与识别和车脸特征比较。车辆图像预处理部分主要采用了直方 图均衡化、灰度变换和中值滤波等方法，目的是抑制图像获取和传输过程中引入 的噪声，提高图像的质量，以便进行后续处理。车牌定位与识别部分主要采用了 差值积分投影和数学形态学的方法进行车牌定位。至于车牌号码的识别，由于该 技术已经十分成熟，故本文仅对其进行了简单概述。车脸定位与识别部分进行定 位和识别的车脸是指散热器隔栅和车灯所在的带状区域。首先，根据车牌与车脸 的位置关系和比例关系，粗略定位出车脸所在的车脸带；然后，使用灰度图像梯 度投影的方法从车脸带中精确定位出车脸；最后，分别使用边缘投影和几何不变 矩方法识别出隔栅和车灯的特征。车脸特征比较部分将识别出的某车车脸特征与 车辆信息库中该车车牌号对应的车脸特征进行比较。如果二者相同，则将该车认 定为真牌车，否则认定为套牌车。 实验表明，该方法切实可行，按照该方法搭建的套牌车检测仿真系统性能良 好。 关键词：套牌车 车牌定位 车脸定位 车脸识别 差分投影   Abstract The appearance of fake plate vehicles seriously disturbs traffic safety, and violates the interests of the legitimate vehicle owners, it causes serious harm to the society. But the investigation of fake plate vehicles mainly relies on the traffic police verifications and public tip-offs. It is inefficient and produces very little effect. Based on above analysis, a method of detecting fake plate vehicles based on vehicle face recognition is proposed in this paper. The fake plate vehicles detection method includes four parts: vehicle imagine preprocessing, license plate location and recognition, vehicle face location and recognition, comparison on facial features of vehicle. In the first part, such methods as histogram equalization, gray level transformation and median filter is used to suppress the noise introduced in the image acquisition and transmission. And it is possible to enhance image for the subsequent processing. In the second part, the license plate is addressed through the implementation of projection method based on mathematical morphology and variance projection. And the license plate recognition is only overviewed for its mature application in this paper. In the third part, the vehicle face consists of radiator grille and headlights. First, the vehicle face is roughly located according to the position of the license plate. Then, the vehicle face