基于Matlab数字图像处理的车牌定位方法  

Location of Vehicle License Plate Based on Matlab Digital Image Processing

作　　者：王传丽[1] 宇仁德[1] 张齐[1] 

机构地区：[1]山东理工大学交通与车辆工程学院,山东淄博255049

出　　处：《交通标准化》2011年第13期133-136,共4页Communications Standardization

摘　　要：通过彩色图像灰度化、最佳阈值法对图像进行二值化,改进的中值滤波策略对图像去噪,小波融合技术等对图像进行预处理,从而提高图像质量。利用改进的边缘检测的方法对预处理后的图像进行定位,并运用数学形态学的知识筛选车牌区域。结果证明,该定位方法取得了较好的效果,可为车牌图像的准确识别奠定基础。Trough binarying the image by color image gray processing method and best threshold method, denoising the image by median filter strategy, and preprocessing the image by wavelet fusion technology, the image quality was improved. The use of modified edge detection method could locate the image after preprocessing, and the regional of number plate was finally screened out by means of the mathematical morphology. It was proved that this location method had achieved good results and can lay a foundation for accurately identifying the license plate.

关 键 词：图像预处理 小波分析 车牌定位 

分 类 号：U495[交通运输工程—交通运输规划与管理]