摘要
在铁路运煤装车过程中为了快速、准确地识别车号,提出一种基于机器视觉的运煤车车号识别技术。将连通区域提取与投影分割法结合,实现车号的粗定位、细分割,并对图像中的断裂字符进行二次分割,构建了基于BP神经网络的分类模型进行车号识别,提升了煤炭装车的效率和精度。
In order to quickly and accurately identify the coal car number during the loading process of the coal railway, an identification method of coal car number based on machine vision was put forward. Firstly, the connected region extraction and projection segmentation method were combined to realize rough positioning and fine segmentation of the car number;secondly, the broken characters in the image were divided twice;finally, the classification model based on BP neural network was constructed to identify the car number.
作者
张兵
杨雪花
ZHANG Bing;YANG Xue-hua(Longdong Coal Mine of Datun Coal Power(Group)Co.,Ltd.,Xuzhou,Jiangsu,221613;School of Information and Control Engineering,China University of Mining and Technology,Xuzhou,Jiangsu,221116)
出处
《煤炭科技》
2020年第1期35-38,共4页
Coal Science & Technology Magazine
关键词
机器视觉
粗定位
字符分割
字符识别
machine vision
rough positioning
character segmentation
character recognition