煤气储备气罐的高度的图像处理系统  

Image processing system for the height of the coal gas storage bottle

作　　者：邹治军[1] 

机构地区：[1]合肥工业大学电气与自动化工程学院,安徽合肥230009

出　　处：《仪器仪表用户》2004年第6期60-62,共3页Instrumentation

摘　　要：本文采用数字图像传感器获取煤气储存罐体上的10个数字手写体的图像信号,输入基于TMS320C6201DSP的图像处理系统中进行图像处理和字符识别,识别结果送至控制室上位机。在图像处理中,应用图像全局二值化算法、字符"有效行"特征的提取和BP神经网络,对手写体字符的识别取得了良好的效果。This article uses digital imaging sensor to obtain the image signals of the handwriting numerals representing the height of coal gas storage bottle which will be sent to system based on TMS320C6201DSP and then be binaried by global threshold. By the extracting method of ' effective-rows' feature of handwriting numerals and recognition method of BP neural networks, image processing and numerals recognition will be done and good results are acquired which will be sent to the computers of the control room.

关 键 词：气罐 图像处理系统 数字图像传感器 BP神经网络模型 煤气管网 

分 类 号：TU996.71[建筑科学—供热、供燃气、通风及空调工程]