Rock mineral component and pore automatic segmentation method based on clustering analysis

• Main Authors: WANG RUANYUE, ZENG YU, TANG MINGZHENG, LIU YUEJIAO, LAI FUQIANG, WANG HAITAO
• Format: Patent
• Language: Chinese; English
• Published: 16.01.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; PHYSICS

The invention relates to the technical field of image separation, and particularly discloses a rock mineral component and pore automatic segmentation method based on clustering analysis, which comprises the following steps: observing minerals in a sample picture in a laboratory, preliminarily determining types according to mineral characteristics, carrying out quantitative analysis, and calculating the percentage Ai of each mineral and pore; converting the preprocessed color image into a grayscale image; clustering the grayscale image into K clusters by using a K-means algorithm; extracting the upper and lower limit thresholds of the gray scale of each cluster, and dividing the original gray scale image into K gray scale images; calculating the percentage Bi of black pixel points in each grey-scale image in the image; and step 6, calculating an absolute error Wi between Ai and Bi until Wi is less than 10%. Intelligent and accurate segmentation of minerals and pores in the slices is realized, the automation de