摘要
车牌定位是车牌识别的关键步骤,能为车牌识别的后续步骤提供极大的便利。为了提高不同颜色车牌定位的准确性,论文提出了一种基于最大极值稳定区域的车牌文字定位和颜色检测的车牌区域粗选方法,并与支持向量机算法相结合实现车牌精确定位方法。首先,为了不遗漏潜在的车牌区域,分别使用文字定位和颜色检测方法粗选车牌候选区域;其次对车牌候选区域做倾斜矫正处理;最后提取车牌候选区域作特征信息,采用支持向量机来精确定位车牌区域。实验结果表明该方法在复杂背景情况下,对各种颜色的车牌都有较高的定位成功率。
License plate location is the key step in vehicle license plate recognition.License plate location can provide great convenience to next steps of license plate recognition.In order to improve the efficiency of different color license plate location,this method selects candidate region based on color detection and a text localization method based on maximally stable extremal regions,and combined with support vector machine(SVM)algorithm,precise location of vehicle license plate is achieved.In order to find all the potential license plate area,this method selects candidate region of license plate based on text localization and color detection,and then tilt correction is done for the candidate regions of license plates.Finally,the license plate candidate region feature information is extracted as the feature information input to support vector machine to identify and locate the license plate region.The experiment results show that this method can eliminate the background interference.The recognition effect and rate of multi-license plate location are high.The experiment results show that this method can achieve high localization rate for most kind color of license plates under complicated background.
作者
葛艳
陈晨
GE Yan;CHEN Chen(Institute of Information Science and Technology,Qingdao University of Science and Technology,Qingdao 266061)
出处
《计算机与数字工程》
2018年第3期575-579,共5页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:61273180)
山东省高等学校科技计划项目(编号:J14LN74)资助
关键词
多车牌定位
文字定位
颜色检测
支持向量机
multi-license plate location,text localization,color detection,support vector machine
