Separation of gangue from limestone using GLCM, LBP, LTP and Tamura
Ore sorting is a useful tool to remove gangue material from the ore, and it increases the quality of the ore. The vast developments in the area of artificial intelligence allow fast processing of full-colour digital images for the preferred investigations. Three different colour spaces were used for analyzing of colour-texture features of limestone and associated gangue. The texture features were extracted using GLCM, LBP, LTP and Tamura. These features were computed from the co-occurrence matrices, which were derived using correlation method for RGB colour. For HSV and YCbCr colour spaces, the texture features were extracted from the luminance information and the colour features from chrominance information of the colour band. The performance of SVM with cubic polynomial kernel was better with 96.8% accuracy as compared to the traditional pattern classifiers (Linear and Quadratic Discriminant analysis) and modern classifiers KNN and weighted KNN.