Automatic classification of Ceramic-Tiles based on Random Forest and combined features

This paper takes plain-glazed tiles as the research object and proposes a method for color difference classification of ceramic tiles based on an image block, combined features, and Random Forest. Firstly, the image was divided into sub-images, and the features of the sub-images were calculated. Then, combine the features and use different algorithms to complete the classification. Finally, the classification effect of various algorithms was compared. The experimental results showed that the method can effectively improve the accuracy of color difference classification after the image was divided into blocks.