自适应遗传算法在车牌定位中的应用

车牌自动识别是现代智能交通的重要组成部分，而车牌定位技术又是车牌识别系统的核心之一。由于传统的定位算法存在适应性差，鲁棒性不强的问题，提出一种基于自适应遗传算法的车牌定位方法。先用最大类间方差法(OTSU)对车牌图像进行二值化，然后利用遗传算法对全图进行车牌特征匹配搜索，结合区域特征向量构造的适应度函数，最终找到车牌区域的最佳定位参量。测试结果表明，该算法适应性强，定位效果很好。

Application of adaptive genetic algorithm in license plate location

License Plate Recognition (LPR) plays an important role in Intelligent Transportation System and the license plate locating is the critical technique in the LPR. Because the traditional locating algorithm is not adaptive or robust, this paper proposed car license plate thick locating method based on the Adaptive Genetic Algorithm (AGA). First, use OTSU method to make license plate image binarization, and then use genetic algorithm to map the entire license plate matching search feature. With a fitness function formed by the area character vector, the optimal locating parameter of the plate area was found. It is proved by experiment that this locating method is strong and the locating effect is very good.