Computer vision-based limestone rock-type classification using probabilistic neural network

摘要: Abstract Proper quality planning of limestone raw materials is an essential job maintaining desired feed in cement plant. Rock-type identification integrated part for mine. In this paper, a computer vision-based rock-type classification algorithm proposed fast and reliable without human intervention. A laboratory scale model was developed using probabilistic neural network (PNN) where color histogram features are used as input. The image histogram-based that include weighted mean, skewness kurtosis extracted all three space red, green, blue. total nine input the PNN model. smoothing parameter selected judicially to develop optimal or close optimum PPN validated test data set results reveal can perform satisfactorily classifying rock-types. Overall error mis-classification below 6%. When compared with other algorithms, it observed method performs substantially better than algorithms.