Contrast Enhancement Based on Intrinsic Image Decomposition

ABSTRACT:

In this paper, we propose to introduce intrinsic image decomposition priors into decomposition models for contrast enhancement. Since image decomposition is a highly ill-posed problem, we introduce constraints on both reflectance and illumination layers to yield a highly reliable solution. We regularize the reflectance layer to be piecewise constant by introducing a weighted `1 norm constraint on neighboring pixels according to the color similarity, so that the decomposed reflectance would not be affected much by the illumination information. The illumination layer is regularized by a piecewise smoothness constraint. The proposed model is effectively solved by the Split Bregman algorithm. Then, by adjusting the illumination layer, we obtain the enhancement result. To avoid potential color artifacts introduced by illumination adjusting and reduce computing complexity, the proposed decomposition model is performed on the value channel in HSV space. Experiment results demonstrate that the proposed method performs well for a wide variety of images, and achieves better or comparable subjective and objective quality compared with state-of-the-art methods.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• In recent years, some methods propose to enhance images by adjusting the decomposed illumination layer to enhance under (over) exposed images.

• Different decomposition models lead to different illumination layers with different features and eventually affect the enhanced images to some extent. Some existing models are based on variational models using the Retinex assumption that the illumination layer is smooth while the reflectance layer is piecewise continuous.

• However, these decomposition methods are all based on uniform constraints in gradient domain, which cannot adapt to image content. The estimated illumination usually contains halo artifacts and the decomposed reflectance layer is mixed with illumination information.

• The histogram-based methods enhance image contrast by modifying histogram distributions. Histogram equalization (HE) which flattens the histogram and stretches the dynamic range of the intensity levels is the most well-known method due to its simplicity and effectiveness.

DISADVANTAGES OF EXISTING SYSTEM:

• Existing method is simple but it will change the brightness of the image and may result in over enhancement.

PROPOSED SYSTEM:

The framework of the proposed contrast enhancement method having following steps,

• Given an input color image I, we first convert it into HSV representation.

• Second, the value (V) channel image is decomposed into illumination (L) and reflectance (R) layers using the proposed intrinsic decomposition model.

• Third, the L layer is adjusted by Gamma mapping function, producing an adjusted L layer, denoted by La.

• Then, the adjusted La is multiplied by the reflectance layer R to generate the enhanced V channel image Ve.

• Since the mapping function is performed globally, we adopt the contrast limited adaptive histogram equalization (CLAHE) to further enhance the local contrast of Ve. The enhanced result is denoted as Ve1.

• Finally, the enhanced HSV image is transformed to RGB space, which yields the final result Ie.

ADVANTAGES OF PROPOSED SYSTEM:

• Our decomposition results lead to the best enhanced result.
• Using CLAHE improved the image quality.

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:

Huanjing Yue, Jingyu Yang, Xiaoyan Sun, Senior Member, IEEE, Feng Wu, Fellow, IEEE, and Chunping Hou, “Contrast Enhancement Based on Intrinsic Image Decomposition”, IEEE Transactions on Image Processing, 2017.