一种数字仪表显示值快速识别方法  

A Rapid Recognition Approach for the Display Value of Digital Meters

作　　者：张海波[1] 段会川[1] 张曙光 刘弘[1] 

机构地区：[1]山东师范大学计算机科学系,济南250014 [2]山东省计量科学研究所,济南250014

出　　处：《计算机工程与应用》2005年第4期223-226,共4页Computer Engineering and Applications

基　　金：国家自然科学基金项目资助(编号:60374054)

摘　　要：提出了一种数字仪表显示值的快速识别方法。该方法首先由计算机自动定位分割图像中的数字区域并实现单个数字的切分,然后对每个数字图像提取了一组具有较高区分度且计算简单的典型特征,最后,基于模糊识别的最大隶属原则,构造了一种数字识别器,实现了仪表显示值的实时识别。试验表明:该方法的识别率高达99%,对7位数字的识别时间不超过20毫秒,并且具有较强的抗干扰能力,达到了仪表显示值识别的速度和准确率要求。A rapid recognition approach for the display value of digital meters is proposed.Firstly,digits areas in the meter image are located and each digit is segmented from each other automatically.Then a group of features with high classification capacity and low complexity are extracted.Finally a digit recognizer based on the maximum membership principle of fuzzy recognition is constructed to recognize the display value of digital meters.Practical experiments show that the recognition rate is higher than 99%,and the recognition time of 7 digits is less than 20ms,which tells that the method is much more applicable for the requirements of the recognition speed and accuracy of digital meters.

关 键 词：数字识别 模糊理论 BP神经网络 数字仪表 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]