欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对车标受天气变化的影响而导致无法准确定位的问题，本文提出了基于Laws模板滤波的车标定位方法，根据车牌与车标的先验知识进行车标粗定位，再根据车标周围散热网的纹理信息，并利用Laws模板滤波除去车标周围散热网的干扰，实现车标的精确定位。试验结果表明，本文方法与Sobel边缘检测方法相比，可更好的保留车标纹理信息。该方法提高了车标定位的光照鲁棒性，定位结果更准确，而且对于较暗的车标也有很好的定位效果。
英文摘要：
In this paper we come up with an algorithm of vehicle-logo location based on Laws＇ mask filte- ring in order to come over the question that vehicle-logo can＇t be localized accurately because of the light change. First, we locate the vehicle-logo coarsely according to the prior knowledge of the position relation- ship between the vehicle logo and the license plate;Then, according to the type of the attered network, we use the Laws＇s mask to filter out the background of the vehicle-logo and locate the accurate logo. The re- sult shows： compared with the Sobel edge detection , the method of this paper remains more texture infor- mation, and improves the algorithm robustness for the night, and at the same time, it has a good location effect for the dark vehicle logo.
同期刊论文项目
同项目期刊论文
期刊信息
位置：