Comparison of the Rock Core Image Segmentation Algorithm

Rock core is an important basic geological data in oil field exploration and development. The scanned rock core image plays an important role in sediment environment inference, reservoir property and oily content analysis. Edge detection of rock core image is the primary content of rock core image digital research. This paper introduces several different image semantic segmentation algorithms, compares the actual segmentation effects of several algorithms, and draws the conclusion that K-means has the best segmentation effect among the four image segmentation algorithms, with the highest sense of detail level and the strongest definition.