Relationship between mineragraphy features of sinter ore and its gray histogram

The intelligent recognizing and processing system make a great convenience for recognition and measurement of sinter ore mineragraphy. Specially, it's crucial for a successful intelligent system to extract the minerals' features accurately and sufficiently. The paper got the relatively position of peak of the gray histogram in theory by using the calculated model of index of reflection and peak-find model of gray histogram; the parameters like mu and theta(2) of familiar minerals were gotten by statistical averaging with Gaussian gray distribution model, the information about the category and content of minerals can also be got by fitting the curve of gray histogram with Gaussian gray distribution model; the feature curves of gray histogram of two minerals were gained by synthesizing two density functions; also the feature parameters such as the number and position of peak and valley of two minerals in different ratios were obtained by differentiating to the distribution functions. The feature parameters, feature curves, and other conclusions lay the foundation for artificial intelligence system of mineragraphy recognizing and processing in sinter ore.