Research On Cutting Tool Condition Monitoring Based On Computer Vision

Tool condition monitoring based on the processed surface texture image is the method that synthesized visual and texture analysis technique to monitor tool wear condition. The visual characteristic of processed surface texture image under different tool wear conditions was investigated for achieving the goal of tool wear condition monitoring.The forming process of surface texture, the morphology and image characteristics of processed surface texture were analyzed, the factors that influencing processed surface texture were compared, the rationality of tool wear monitoring method based on processed surface texture was explained.The usual preprocessing method of image was researched, analyzed and collated based on workpiece surface image in this paper. The preprocessing method adapted to workpiece surface image was discovered. The foundation for realizing the image characteristic extraction about the tool wear condition monitoring was laid. The research and exploration of data mining algorithm on workpiece surface texture is very useful for automation and unmanned production. The surface texture image was analyzed and feature data was extracted by gray-level co-occurrence matrix method, pixel space projection method, connected region labeling method and Markov Random Field (MRF) theory. The results indicated that the surface texture image could be used as evaluation standards for tool wear; it provided effective way for tool wear monitoring.For recognizing the cutting tool state, aim at classified difficulty of feature sample of distinct cutting tool state kind existing overlap region, the paper propose the model and method of cutting tool state recognition based on fuzzy decision. Aim at feature decision edge existing certain nonlinearity, the paper propose the model and method of cutting tool state recognition based on BP neural net. Then two above recognition model are fused at decisional level, the model and method of cutting tool state recognition based on classifier fusion is proposed. The efficiency and accuracy of cutting tool state recognition is remarkably raised by the fusion of two above recognition model based on different feature.