摘要
危化品车辆抓拍识别是加强危险品运输监管,保障社会公共安全的关键。基于图像处理和极限学习机,提出了一种在途危化品运输车辆抓拍识别系统及算法。对抓拍危化品车辆图像二值化处理,通过模糊算法和形态学膨胀操作提取车牌区域。对提取的区域滤波处理和形态学膨胀操作,提取车牌字符特征。将提取的车牌字符特征作为极限学习机的输入,危化品车辆车牌的33个字母数字作为输出,达到危化品车辆车牌识别的目的。结果表明,所提出的算法对危化品车辆所处环境具有更好的适应性,同时具有更高的抓拍识别精度。这对危化品车辆行驶管理具有一定的实用价值。
Capture and identification of dangerous chemical vehicles is the key to strengthen the supervision of dangerous goods transportation and ensuring social and public safety.Based on image processing and extreme learning machine,a capture recognition system and algorithm for in⁃transit hazardous chemical transport vehicles were proposed..The binary processing of the captured hazardous chemical vehicle image was carried out,and the license plate area was extracted through the fuzzy algorithm and morphological expansion operation.The extracted area was filtered and morphologically expanded,and the character features of the license plate were extracted.The extracted license plate characters were used as the input of the extreme learning machine,and 33 letters and numbers of the license plate of the dangerous chemical vehicle were used as the output to achieve the purpose of license plate recognition of the dangerous chemical vehicle.The results showed that the proposed algorithm had better adaptability to the environment of hazardous chemical vehicles,and had higher accuracy of capture recognition.This has certain practical value for the management of dangerous chemical vehicles.
作者
方小娟
FANG Xiaojuan(Henan Transportation Technician College,Zhumadian 463000,Henan China)
出处
《粘接》
CAS
2024年第7期113-116,共4页
Adhesion
基金
河南省职业教育教学改革研究与实践项目(项目编号:豫教〔2020〕43151号)。
关键词
危化品车辆
车牌识别
字符分割
特征提取
极限学习机
dangerous chemical vehicles
license plate recognition
character segmentation
feature extraction
extreme learning machine
