作　　者： (杨亭亭）; (曾洁）; (刘宾坤）; (曾奕哲）; (张育华）;

机构地区： 大连交通大学,大连116028

出　　处： 《单片机与嵌入式系统应用》 2017年第9期59-63,共5页

摘　　要： 为提高车辆识别码(VIN)录入的效率,设计并开发基于Android平台的车辆VIN识别系统。首先调用Android手机摄像头完成对VIN的定位及其图像获取;其次对获取的VIN原始图像进行灰度化、顶帽变换、图像增强和二值化;然后基于Google公司提供的开源Tesseract-OCR识别库[1]对预处理后的VIN图像进行字符特征提取,与训练生成的VIN字符库做匹配识别;最后为确保获得正确结果,对识别输出的VIN根据其编码规则进行校验和人工校正[2]。经过各种测试证明,各项功能实现效果很好,能有效地识别出多数车辆识别码[3]。 In order to improve the efficiency of Vehicle Identification Number（VlN） entry,a VIN recognition system based on Android platform is proposed.First, the Android camera completes the position and image acquisition of the VIN.Secondly, the acquired original image of vehicle VIN is processed by graying, top-hat transformation, image enhancement and binarization. Then, character features are extracted from the preprocessed image of the vehicle VIN based on the open-source Tesseraet-OCR identification library provided by Google, and are matched the vehicle VIN character library generated by the training.Finally, in order to acquire the correct result, the rec ognition outputs need to be verified automatically according to the VIN coding rules or corrected manually.The test results prove that the various functions achieve good results,and can effectively identify the majority of vehicle identification code.

关 键 词： 识别系统 图像预处理