Abstract:In order to realize the automation of beneficiation, industrial cameras to collect the image of ore particles on the conveyor belt, Matlab sequentially performs bilateral filtering and noise reduction processing on the acquired images, optimal threshold segmentation of genetic algorithm, watershed segmentation, ore particle pixel calibration and pixel area was used to investigate the particle size distribution of the ore. The results show that compared with the data of artificial screening, the results of image processing were shown to be within the error tolerance. It is concluded that it is feasible to use the image processing method to detect the ore particle size, which can provide data guidance for the crushing of ore and promote the automation and intelligence of mining production.