HSV空间和形态学处理相结合的车牌定位方法

License Plate Location Method Based on Combination of HSV Space and Morphology Processing

车牌定位是实现车牌自动识别的前提。在传统的基于RGB彩色空间的形态学定位方法的基础上进行改进,提出了一种HSV彩色空间和图像形态学处理相结合的车牌定位方法。在HSV彩色空间中将车牌图像分割为H,S,V这3个单通道灰度图像,分别进行去噪和二值化处理,然后将这3幅二值图像做"与"运算得到一幅能有效去除背景干扰的二值图像,再运用形态学的闭运算和开运算进行处理,得到车牌候选区域,最后利用车牌宽高比属性进行验证以确定真实的车牌位置。实验结果表明,该方法相对传统方法实现简单,车牌定位准确、高效。

License plate location is the premise to achieve automatic license plate recognition. In this paper, a license plate location method based on the combination of HSV color space and image morphology processing is proposed to improve the traditional method based on RGB color space and morphology. When converted to HSV color space, license plate image is split into three single-channel grayscale images of H, S and V which are then denoised and binarized respectively. To effectively remove the background interference, “And” operation is applied on these three binary images to generate a new binary image. Then the morphological methods such as the “Close” operation and “Open” operation are used to detect the candidate regions of license plate. Finally , the aspect ratio property of license plate is adopted to validate them to determine the true license plate locations. Experimental results show that the proposed method is easy to implement and more accurate and efficient to locate the license plate compared with the traditional one.