摘要
图像或视频中的车牌识别(LPR)算法包括车牌区域的提取、车牌字符的分割和车牌字符的识别3个部分。由于车牌格式的多样性和采集过程中室外光照条件的不均匀性,增加了车牌定位识别的难度。因此,大多数识别方法只能在有限的条件下实施,如固定的照明、限制一定的车速、指定的路线和固定的背景等。车牌识别(LPR)算法由两个模块组成,一个模块用于定位牌照,一个模块用于牌照中字符的识别。对静止图像或视频序列中的车辆,利用边缘检测和数学形态学算法进行车牌定位和识别;通过算法优化,测试并调整处理时间的实验,验证了定位计算能力和识别效率;为研究人员提供了一个公共图像数据库。
The algorithm of license plate recognition(LPR)in image or video includes three parts:the extraction of license plate area,the segmentation of license plate characters and the recognition of license plate characters.Due to the diversity of license plate formats and the non-uniformity of outdoor lighting conditions in the acquisition process,the difficulty of license plate location recognition is increased.Therefore,most identification methods can only work under limited conditions,such as fixed lighting,limited speed,designated route and fixed background,etc..The license plate recognition(LPR)algorithm is composed of two modules,one is for locating license plates;and the other is for recognizing characters in license plates.The purpose of this paper is to use edge detection and mathematical morphology algorithm to locate and recognize vehicle license plates in still images or video sequences.An experiment to adjust the processing time is made by optimizing the algorithm,and verifies the positioning calculation ability and recognition efficiency.Finally,a common image database is provided for the researchers.
作者
司均飞
SI Jun-fei(Anhui Institute of International Business,Hefei 230001,China)
出处
《长春工程学院学报（自然科学版）》
2020年第3期108-113,共6页
Journal of Changchun Institute of Technology：Natural Sciences Edition
关键词
图像处理
车牌识别(LPR)
数学形态学
边缘检测
image processing
license plate recognition(LPR)
mathematical morphology
edge detection
