Coal-rock CT image enhancement based on spatial domain image decomposition and adaptive enhancement factor

Coal-rock CT (computed tomography) images are widely used to analyze the inner structure of coal-rock and further provide help for the study of failure mechanism, coalbed methane development and permeability calculation. However, coal-rock CT images suffer from low-contrast and the random distributed noise, and the existing methods cannot effectively solve the conflict of contrast improvement, noise reduction and detail preservation. An adaptive enhancement method of coal-rock image based on filtering and variational decomposition is proposed. In the paper, median filtering and total variational model are used to decompose the coal-rock image into three layers: noise-layer, texture-layer and structure-layer. Guided filtering is used to obtain the residual details from the image of noise-layer, and the texture-layer is fused. An adaptive enhancement factor is designed to highlight the fused details. The local mean gamma correction algorithm is used to improve the contrast of structure-layer. Finally, the enhanced structure-layer and the enhanced texture- layer are fused to obtain the enhanced coal-rock image. The experimental results show that this method can balance the relationship between denoising, contrast improvement and detail preservation of coal-rock images.