Byoung-Ju YUN¹, Hee-Dong HONG², Jinhyoung PARK³, Hyun-Deok KIM¹, and Ho-Hyoung CHOI^4*

¹School of Electronics Engineering, IT College, Kyungpook National University, Deagu 702-701, Republic of Korea
²Department of e-Commerce, Semyung University, Jecheon, Chungbuk, 390-711, Republic of Korea
³Korea Institute of Science and Technology Information, Daejeon 305-806, Republic of Korea
⁴Department of Mobile Communication Engineering, Kyungpook National University, Deagu 702-701, Republic of Korea

(Received April 16, 2012; Revised January 30, 2013; Accepted February 7, 2013)

Over the last several decades, the tone-mapping and color correction methods have been incorporated into the display device such as LCD/LED display device or the photographic print. However, conventional methods are based on Gray-World Assumption using spatial smoothing filter to estimate the local illumination component. For this reason, most of these methods have several problems like visual perception problem, scale the contrast, gray-out, color distortion (dominate color, halo-artifact) and so on. Accordingly, this paper presents a color correction method using the dynamic cone response function based on human visual perception. The proposed method consists of tone-mapping and dynamic cone response. The tone-mapped image is obtained by using chromatic and achromatic colors. Thereafter, the resultant image is processed through the dynamic cone response function, and it is to estimate the dynamic response of human eye as well as to deal with mismatch between a real scene and the rendered image. The experimental results show that the proposed method yields better performance of color correction over the conventional methods.

Key words: HDR image, dominated color, tone-mapping, dynamic cone response, human visual perception

^*E-mail address: chhman2000@msn.com