A Method for Detecting Miners in Underground Coal Mine Videos

摘要

Detecting miners in underground coal mine videos is significant for the production safety. But, the miners are very similar to the background in underground coal mine videos, it is difficult to detect. In this paper, we proposed a method to detect miners by detecting their helmets. The helmet is round or resembles a round from different angles to observe. The operations include background subtraction, denoising, erosion, region growing, circle detecting, and tracking. The method has been tested on the real underground coal mine videos. The experimental results show that the miners can be detected accurately even though they can not be seen clearly.