Chemical Image Recognition Based on BP Neural Networks

A new method to identify the catalyst activity based on BP neural network is proposed in this paper in order to enrich the recognition methods of chemistry catalyst activity. In this method some important image features from the process of using chemistry catalyst can be extracted depending on the gray level co-occurrence matrix (GLCM) by image filtering and image segmentation. Furthermore, a BP neural network is trained by the image features and used to identify the catalyst activity. The result of experiment shows that recognition rate of chemistry catalyst is increased and the production cost is saved accordingly for the reduction of chemistry catalyst.