复杂背景下车牌定位与字符分割算法研究



本文在简要介绍国内外车牌识别技术研究现状的基础上，对车牌定位算法及字符分

割算法进行了深入的研究，主要完成以下工作：

车牌的定位算法设计阶段，本文提出了一种综合利用车牌多重特征的定位算法。该

算法充分利用了车牌的纹理，色彩及车牌长宽比等特征。经过水平梯度化，二值化，滤

波，区域连通、连通域标记筛选，色彩匹配等几个步骤，最终达到了复杂背景下汽车牌

照的定位。-Complex background license plate location and character segmentation algorithm of this paper, a brief introduction to license plate recognition technology at home and abroad on the basis of the status quo, on the license plate location and character segmentation algorithm is conducted in-depth study was completed for the following: license plates location algorithm design stage, this paper, a comprehensive utilization of multiple characteristics of license plate location algorithm. The algorithm makes full use of the license plate texture, color and license plate aspect ratio and other features. After level gradient, binarization, filtering, regional connectivity, connected component labeling filter, color matching, and several steps, ultimately achieving the complex background Locating the vehicle license.