Methods and algorithms of image recognition for mineral rocks in the mining industry

This paper describes development of methods and algorithms of image recognition for mineral rocks. Algorithms of the cluster and morphological analysis to determinate colors and shapes for composition of rocks are described. This approach is actual because of existence of objects with similar color-brightness characteristics, but different shapes or objects that have similar color-brightness characteristics. Preliminary determination of group membership allows reducing the computational complexity of classification. Sorting by group produces color of the object is determined at the stage of segmentation. Few examples of segmentation algorithms in the solving of mineral rock recognition problems are described and discussed