An online rock lithological detecting method using color and texture analysis

In the crushing and comminution process, different lithological rocks strongly affect the performance of crushing equipment. However, traditional lithological identification cannot be analyzed in a timely manner for processing control. Many rock lithological classification methods using color images have been proposed in the past, but few have successfully connected their results with online equipment monitoring. In this paper, we have developed an automatic vision system to evaluate rock lithology in real-time using color and texture features. The results show that the proposed method based on color and texture is robust and outperforms previous methods that rely on any single feature alone. This will help achieve reliable and precise control of the Semi-autogenous mill (SAG) feeding process.