一种基于区域熵值的车牌定位方法  

A District-Entropy-Based Method for Locating Vehicle License Plates

作　　者：张国敏[1] 殷建平[1] 祝恩[1] 

机构地区：[1]国防科技大学计算机学院,湖南长沙410073

出　　处：《计算机工程与科学》2004年第5期31-33,共3页Computer Engineering & Science

基　　金：国家自然科学基金重点资助项目 ( 60 3 73 0 2 3 )

摘　　要：因搜索空间太大 ,现有的车牌号码识别系统所采用的图象分割方法一般说来效率较低。为此 ,本文提出了一种基于区域熵值的车牌定位方法 ,可快速判定车牌的粗略位置与颜色 ,即得到一个属于车牌区域的参考点和车牌的背景色 ,从而使得随后的图象分割仅在该参考点的周边区域进行 ,大大缩小了搜索空间 。Generally speaking, the image segmentation method adopted in vehicle license plate recognition systems at present is not efficient due to its large search space. This paper presents a district-entropy-based method for locating vehicle license plates to solve the problem. A point belonging to the district of a vehicle license plate, which represents its location approximately, and the background color of the vehicle license plate, can be decided rapidly by our method. Consequently, the image segmentation can be carried out only on the region around the point, which reduces the search space and increases the efficiency greatly.

关 键 词：车牌号码识别系统 区域熵值 图象分割方法 车牌定位 定位算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]