利用HSV色彩空间进行车牌定位的研究
首发时间:2016-02-18
摘要:为了解决复杂环境中车牌的定位问题,提出了一个结合车牌颜色特征、灰度特征、几何形状特征的通用车牌定位方法,首先根据中国车牌的颜色特征,提出了一种基于HSV色彩空间的颜色定位车牌的方法,解决了车牌识别准确率不高的基本难题;然后采用改进的Sobel算子对车牌图像进行边缘检测得到车牌底色所在的矩形区域;最后将可能的车牌区域统一尺寸,并进行验证,最后分割出车牌区域。采用自然复杂环境中拍摄的图像对新方法进行验证的实验表明,与其他方法相比,新方法可在一定限度内自适应车牌的类型、大小、数量和方向,并对汽车在图像中的位置以及图像背景的限制较少,是一种适用性较强的方法。
For information in English, please click here
Research of License Plate Location Based on HSV Space
Abstract:In order to solve the problem of license plate location in complex environment ,a novel approach to LP extraction was presented ,which utilized LP's color information ,gray feature and geometry characteristics. Firstly, according to the color feature of Chinese license plate, a new method based on HSV color space is proposed, which solves the problem of the low accuracy of license plate location. Then using the improved Sobel edge detection methods to get the plate background where the rectangular area; will probably end up the license plate region of uniform size and verified. Finally, the segmentation of license plate region. Experiments using a new method of image captured in the natural complex environment show that Compared with other methods, the new method can be applied to a certain extent, the type, size, number and direction of the license plate.
Keywords: license Plate extraction color features Sobel algorithm edge detection
基金:
论文图表:
引用
No.4677899113395214****
同行评议
共计0人参与
勘误表
利用HSV色彩空间进行车牌定位的研究
评论
全部评论