Digital Image Processing And Pattern Recognition Technology In The Research Of Sinter Analysis

The ingredient analysis of sinter microscope image is a challenging research area. To accomplish automatic recognition task on it, this paper proposed an improved statistical hypothesis method for mesh texture ingredient recognition; and with Sobel edge detection, Skeleton Extraction, Hough Transformation another method is proposed on the recognition for needle like or striation texture ingredient. The experimental results show that the methods proposed in this paper are efficient.To automate analyze the sinter property quickly and correctly, an automated software, which include three major functions: auto-sampling processing on micrograph, segmentation and recognition of sinter components, as well as analyses of sinter components, has be designed and realized in the research project on sinter.The auto-sampling processing on micrograph is composed of auto-exposure processing, auto-focusing processing, and auto-collect processing.Sinter is a method for making objects from powder; by heating the material (below its melting point) until its particles adhere to each other. There are totally 6 classes of sinter component. The goal must be achieved is to identify the 6 major category and the further target is identify all the 9 component classes in minor category.The user interface of the software is friendly, well organized, is an example for similar industrial automation software. Experimentations show that the software runs well and steady with high efficiency, and the results are very correctly. The arithmetic used in the automatic processing could offer much help to other project of automatic processing of micrograph and recognition of microstructure, while have a high practicality value in the industrial field of energy sources, refinement metallurgy, refinement chemical engineering.