变电站巡检机器人的数字仪表自动识别技术研究

针对变电站巡检机器人数字仪表识别算法准确度不高的问题,本文提出了一种基于样本匹配算法的数字仪表读数识别算法。该算法在使用局部最大类间方差的高阶统计和阈值对仪表板上的显示区域进行定位的基础上,最后根据数字仪表显示区域和数字位置相对固定的特征采用样本匹配算法实现对数字仪表的自动读数。实验结果表明,该方法简单、高效,可有效地应用于变电站自动检测机器人的室内仪表读数自动识别任务。

Research on Digital Instrument Automatic Recognition Technology of Substation Inspection Robot

In view of the low accuracy of digital instrumentation recognition algorithms for substation inspection robots, this paper presents a digital instrument reading recognition algorithm based on template matching. The algorithm is based on template matching pattern recognition method combined with deep learning algorithm, taking into account the characteristics of the substation indoor environment. First, use the high-order statistics and thresholds of the local maximum variance between classes were used to locate the digital region on the dashboard, and then the template matching deep learning algorithm was used to compare the digital region with the selected instructional image template to finally determine the value of the numeric region. The experimental results show that the method is simple and efficient, and it can be effectively applied to the task of automatic identification of indoor meter readings of substation automatic detection robots.