Measurement of Surface Friction Area on Brush Based on Machine Vision

Aiming to address the challenge of accurately measuring the wear area on the surface of collector ring brushes after the running-in process, this study proposes a method for detecting the friction area based on multi-channel image difference operations and adaptive threshold segmentation. The method involves decomposing RGB images, performing difference operations, and combining adaptive global threshold segmentation to extract the target region. Furthermore, it utilizes connected component analysis, morphological operations, and area filtering to obtain the pixel count of the target region. Experimental results demonstrate that the maximum relative measurement error of this method is 4.2%, meeting the precision requirement of a ±5% error margin. Additionally, the average standard deviation of repeated measurements is 0.84 mm², indicating high measurement stability. This approach provides reliable data support for subsequent analyses of the influencing factors of the slip ring.