High-precision automatic extraction method of cultural relics disease based on texture

Abstract

The invention discloses a texture-based high-precision automatic extraction method of cultural relics diseases, comprising: step 1, acquiring an orthophoto image of the cultural relic; The pixel segmentation method performs SLIC clustering on the orthophoto image data of cultural relics, and divides the diseased area of ​​cultural relics into several pixel sets with similar sizes and the same attributes; The class center fuses the diseased areas of the cultural relics together, and calculates the diseased situation statistically. The invention also discloses a texture-based high-precision automatic extraction system for cultural relic diseases, including an image segmentation module, a clustering module and a disease statistics module. The invention greatly reduces the interference of human participation in the process of cultural relics disease investigation, provides basic data for the restoration, improvement and protection scheme formulation and research of cultural relics, and has practical significance and scientific research value for the digital protection of cultural relics.

Description

technical field [0001] The invention relates to a texture-based high-precision automatic extraction method for cultural relic disease. Background technique [0002] The research on the extraction of cultural relics disease is the basis of cultural relics protection and cultural relics research, and the digitization of cultural relics in the cultural relics protection project is the premise of the extraction of cultural relics diseases. At present, for the extraction of cultural relics and diseases, the manual extraction method based on the digital information of cultural relics is still generally used. The problems are: (1) If there is no on-site lighting environment adjustment parameter and no color card for subsequent color correction when acquiring image data, the disease cannot be guaranteed. The correctness of texture information; (2) In the process of disease investigation, the disease image data obtained in many cases is not orthophoto image, and the reliability of th...

AI Technical Summary

Problems solved by technology

At present, for the extraction of cultural relics and diseases, the manual extraction method based on the digital information of cultural relics is still generally used. The problems are: (1) If there is no on-site lighting environment adjustment parameter and no color card for subsequent color correction when acquiring image data, the disease cannot be guaranteed. The correctness of texture information; (2) In the process of disease investigation, the disease image data obtained in many cases is not orthophoto image, and the reliability of the investigation results cannot be guaranteed if this image is used for disease investigation; (3) Most of the disease extraction is The application software manually extracts the diseased area, the workload is heavy and the efficiency is low

The main problems are: (1) The multi-threshold segmentation method may not be able to obtain the threshold on images with flat histograms or peaks; (2) The cluster-based segmentation method needs to determine the number of clusters first, and the clustering results The image subject to the initial clustering center is larger; (3) The region-based segmentation method is prone to over-segmentation and combined with region merging to obtain better results. Generally, this method is used in medical image analysis and other fields

Images