A Study on DWT Filter to Work on the Geometric Components of Fabric Images using Colour Transfer

Abstract

A completely unique color transfer method for cloth images is proposed. Firstly, the new method in its image segmentation phase follows an assumption that a cloth image are often decomposed into cartoon and texture components, which suggests the new color transfer method in its image segmentation phase incorporates an image decomposition process. The advantage of the innovation is that the cartoon component is more suitable than the primary image to be wont to partition the material image. Secondly, the new color transfer method can generate more vivid color transfer results since the above texture component is employed to elucidate yarn texture details within the image construction phase. Thirdly, the entire generalized variation regularize is used to further improve the performance of image decomposition. Additionally, by using the augmented Lagrange multiplier method, we derive an efficient algorithm to seem for the solutions of the proposed color transfer problem. Better results for cloth images are often generated by color transfer method.