Color Image Enhancement with Brightness Preservation Using A Histogram Specification Approach
G. Jiang,S. C. F. Lin,C. Y. Wong,M. A. Rahman,T. R. Ren,Ngaiming Kwok,Haiyan Shi,Ying-Hao Yu,Tonghai Wu
DOI: https://doi.org/10.1016/j.ijleo.2015.08.173
IF: 3.1
2015-01-01
Optik
Abstract:Enhancement of image quality is a primary requirement before visual based information can be used in various real-world and engineering applications. With regard to a faithful reproduction of the scene information, degradations resulting in losses of contrast have to be rectified where the efficient histogram equalization is frequently employed. Moreover, it is also necessary to maintain the brightness of the corrected image in order to truly represent the scene characteristics. In this work, an approach is proposed to specify a proper histogram profile such that image intensity values are adjusted accordingly and the output brightness is maintained close to the input image. Specifically, the equalization profile is shaped by finding a balancing control threshold through carrying out an integration of rectangular and triangular sections. Experiments are conducted using a large number of natural color images and compared to other available histogram based image enhancement methods. Results have shown that the proposed approach is able to accomplish a wide set of performance goals including information content, sharpness and color quality.