摘要
由于车牌识别系统中车牌位置精确定位难和车牌中字符识别率低等问题。本文提出了一种基于SVM与ANN神经网络的车牌识别算法。通过Soble边缘检测算法与形态学算法相结合来确定大致的车牌轮廓,结合车牌的外接矩形的面积与长宽比来筛选出符合车牌特征的候选区域,再利用SVM分类器来判断检测到的区域中是否是车牌,来最终筛选出是车牌的区域。对于筛选出的车牌利用ANN神经网络进行车牌字符的识别。经验证,该车牌识别系统能够适用于比较复杂的环境,且识别速度快,准确率相对较高。
Due to the difficulty in accurately positioning the license plate position in the license plate recognition system and the low recognition rate of the characters in the license plate. This paper proposes a license plate recognition algorithm based on SVM and ANN neural network. The Soble edge detection algorithm is combined with the morphological algorithm to determine the approximate license plate contour. The area and aspect ratio of the circumscribed rectangle of the license plate are combined to select the candidate area that meets the license plate characteristics, and the SVM classifier is used to judge the detected area. Whether it is a license plate or not, to finally screen out the area of the license plate. For the selected license plates, the ANN neural network is used to identify the license plate characters. It has been verified that the license plate recognition system can be applied to a relatively complicated environment, and the recognition speed is fast and the accuracy is relatively high.
作者
林乾毕
LIN Qian-bi(Hangzhou University of Electronic Science and Technology, Hangzhou Zhejiang 310018)
出处
《软件》
2019年第8期105-107,共3页
Software
作者简介
林乾毕(1993-),男,硕士,研究方向:应用软件开发.
