基于机器视觉的高压计量数显表自动校对系统

针对肉眼不能实时同步待检测高压计量仪表和标准高压计量仪表读数问题,提出了一种基于机器视觉的数显式高压计量仪表自动校对系统。系统通过两个相机同步采集待检测高压计量仪表和标准高压计量仪表字符图像,并对采集的图像进行二值化与膨胀处理,然后根据水平投影法分割图像,提取了改进的结构特征,再用K最近邻(K-Nearest Neighbor,KNN)算法识别字符,解决了肉眼识别效率低、精确度差的问题。经过测试,该系统实现了实时、同步识别待检高压计量仪表和标准高压计量仪表的读数,且识别率达到99.6%以上。

Automatic Calibration System of High Voltage Measuring Digital Display Instrument Based on Machine Vision

In order to solve the problem that human cannot read the readings of the high voltage device to be tested and the standard device synchronously, the automatic calibration system of high voltage measuring digital display in-strument based on machine vision is proposed in this paper. Two cameras were used to collect the character image of the testing device and the standard device. Then, the binarization and the dilation were employed to the image collect-ed. Then, the horizontal-projection algorithm was used for segmentation and the improved structural features were ex-tracted. Finally, K-Nearest Neighbor (KNN) algorithm was used for classification which solved the problem of poor accuracy, low efficiency of visual identification. The experimental results show that the system can recognize the read-ings of the high - voltage device and the standard voltage device simultaneously and the recognition rate is above 99. 6% .