Substation Equipment Abnormal Voice Analysis Method Based on Image Processing

摘要

An audio signal analysis method using digital image processing technology is proposed, which is used to detect the abnormal signal of substation equipment. First of all, the sound signal acquisition with pickups that robot carrying. Secondly, the waveform of the sound signal is transformed into the image, and the image processing technique is used to process the image, and then the horizontal projection of the waveform image are analyzed. Finally, edge detection and Hough line detection are performed on the horizontal projection, the ratio of area which the line segment is calculated. The ratio of the area compare with the set threshold value, which greater than the threshold value is abnormal signal. The experimental results show that the algorithm can achieve the abnormal analysis of the sound signal, and the algorithm is simple and accurate.