Global–local image enhancement with contrast improvement based on weighted least squares

Abstract

Contrast improvement as well as detail preservation is important for human visualization in image enhancement. Traditional histogram equalization methods are easy to produce over-enhancement results, resulting in the loss of fine details. In order to deal with this problem, this paper proposes an image enhancement scheme (called as global–local image enhancement, GLIE) based on weighted least squares (WLS) for contrast enhancement and structural preservation, which includes the following key parts: First, we employ the WLS method to separate a low-quality image into a base image and a detail image. Then, the base image is enhanced by adaptive gamma correction based on cumulative histogram. Next, the enhanced base image jointing with the original image is used to obtain an enhancement coefficient for correcting details. Finally, the high-quality image is generated by adding the enhanced base and detail images. Experimental results in both qualitation and quantitation verify that the proposed GLIE method is effective, and even is better than the state-of-the-arts.