Cartoon and Texture Decomposition based Color Transfer for Fabric Images

ABSTRACT:

A color design process for fabric images can resort to a solution of a color transfer problem based on given color themes. Usually, the color transfer process contains an image segmentation phase and an image construction phase. In this paper, a novel color transfer method for fabric images is proposed. Comparing with classical color transfer methods, the new method has the following three main innovations. Firstly, the new method in its image segmentation phase follows an assumption that a fabric image can be decomposed into cartoon and texture components, which means 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 original image to be used to partition the fabric image. Secondly, the new color transfer method can generate more vivid color transfer results since the above texture component is used to describe yarn texture details in the image construction phase. Thirdly, the total generalized variation (TGV) regularizer is used to further improve the performance of image decomposition. Here, the TGV regularizer is good at estimating the weak lightness variation of the cartoon component with the CIELab color scheme. In addition, by using the augmented Lagrange multiplier method, we derive an efficient algorithm to search for the solutions of the proposed color transfer problem. Numerical results demonstrate that the proposed color transfer method can generate better results for fabric images.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• In existing, a variational color-theme based method is proposed. The process contains two sequential phases: the image segmentation phase (the first phase) and the new-image construction phase (the second phase).

• In the first phase, a target fabric image is partitioned into several regions by colors, where the mean colors of all different regions are combined to be the color theme of the input image. Note that the image segmentation model used is a generalized version of the model, and it can estimate a bias field of the target image from the lightness component of the image. Here, the image refers to its coordinates in the CIELab color space, and the bias field at each pixel indicates the dividing result of the lightness intensity of the pixel by the mean value of the lightness intensities in the region which the pixel belongs to.

• In the second phase, for a given color theme, a new image is constructed by changing the estimated color theme of the input image with the given color theme, and the new image is also considered as a color design proposal.

DISADVANTAGES OF EXISTING SYSTEM:

• The constructed images are still not quite satisfactory from vision. In fact, when we compare the constructed image with its corresponding original input image, some geometrical structure details of the constructed image are too smooth, while some artifacts can be found in the constructed image.

PROPOSED SYSTEM:

• In this paper, we propose a new color transfer model called as the CTD-BCT model. The corresponding color transfer results are used as color design proposals. The new color transfer model includes an image segmentation phase and an image construction phase.

• In the image segmentation phase, we design a new image segmentation model in which image decomposition and image segmentation are coupled with each other. Here, image decomposition means that an input fabric image is decomposed into its cartoon component and its texture component. The advantages of the introduction of image decomposition into image segmentation are that, firstly, the cartoon component better coincides with the local piecewise constant assumption of the image segmentation model than the original input image, and, secondly, the texture component of the image can be used in the image construction phase, which makes the color transfer results containing much more fabric texture details.

• To obtain relatively good image decomposition results, the TGV regularization term is used to measure the lightness details of the cartoon component since the TGV regularizer can protect lightness inhomogeneity in each homogeneous color region. By numerical experiments, we find that our proposed CTD-BCT model generates better color transfer results than the classical BCT model.

ADVANTAGES OF PROPOSED SYSTEM:

• In this paper the cartoon component instead of the original image is used in the image segmentation phase since the cartoon component coincides with the piecewise-constant assumption better.

• Texture component is used for extract the texture details of fabric image.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB

• Tool                      : MATLAB R2013A

REFERENCE:

Yu Han, Chen Xu, George Baciu, Min Li, Md. Robiul Islam, “Cartoon and Texture Decomposition based Color Transfer for Fabric Images”, IEEE 2017.