Tenglong Yuan blue and white texture extraction method based on adaptive gamma correction and K-means clustering segmentation coupled algorithm

Abstract

Regarding the missing texture details in the texture extraction process of Tenglong Yuan blue and white, illumination is unequal. A meta blue and white texture extraction method based on adaptive gamma correction and K-means clustering segmentation coupling algorithm has been proposed. Combining the characteristics of Tenglong Yuan blue and white texture, using grayscale transformation to adjust the brightness of blue and white images, image contrast has been enhanced. Design Gaussian filter, weighted average multiscale Gaussian convolution, constructing 2D gamma convolutions for adaptive gamma correction, enhance the texture of Tenglong Yuan blue and white, the details in the blue and white texture of Tenglong Yuan are richer and more prominent. K-means clustering segmentation algorithm based on Lab space, implementing color segmentation of blue and white images, helps to segment the blue and white texture of Tenglong Yuan. Verification indicates that this method can effectively improve the contrast of blue and white texture; the accuracy rate of Tenglong Yuan blue and white texture segmentation reaches 95%. Effectively improving the accuracy of texture extraction for blue and white elements, the protection and inheritance of Yuan blue and white cultural relics are promoted.