基于自适应特征提取的数显仪表识别系统

Digital-display instrument recognition system based on adaptive feature extraction

为了适应数显仪表各种不同的室外环境,建立了自适应于不同光照条件下的仪表自动识别系统。针对数显仪表图像的点亮区域多为红色,不亮部分偏灰白的特征,设计基于RGB彩色空间滤波方法,较好地保留了点亮区域的亮度信息;并提出改进的最大类间方差法,实现了数显仪表图像字符的自适应提取。根据数显仪表数字特征信息,采用穿线识别法,实现了数显仪表的自动识别。通过大量实测实验表明,提出的自适应特征提取算法对不同角度、不同光照环境拍摄的仪表图像都有良好的适应性,应用于数显仪表自动识别系统可得到95%的识别准确率,平均运行时间为1.136 s。

In order to make the digital-display instrument adapt to various outdoor environments, an instrument＇s automatic recognition system adaptive to various illuminations is established. According to the features that the color of the bright area of the digital-display instrument image is almost red and the color of the dark area is partial to gray, the color space filtering method based on RGB is designed to reserve the luminance information of the bright area better. The improved Otsu method is proposed to realize the adaptive extraction of image characters of the digital-display instrument. The threading recognition method is used to realize the automatic recognition of the digital-display instrument according to the digital feature information of digital-display instrument. The actual measured results of massive experiments show that the adaptive feature extraction algorithm has strong adaptability to the instrument images taken in different imaging angles and different illumination environments, and can get the identification accuracy of 95% when it is applied to the automatic recognition system of digital-display instrument; its average running time is 1.136 second.