The texture characteristics of both stone and lump coal are found to be most significant as far as separation is concerned and a separation algorithm based on Mahalnobis' distance discrimination function is developed.
With a variety of image processing and pattern recognition methods, the image characteristics of both stone and lump coal have been studied in this paper. It is found that the texture characteristics are most significant as far as separation is concerned. A separation algorithm based on Mahalnobis' distance discrimination function is also developed. The corre- sponding method has been tested in the lab with large samples of lump coal and stone on a moving belt. The results are quite encouraging.