Preserving perceived brightness of displayed image over different illumination conditions

摘要

In this paper, we propose an image correction method for preserving a brightness appearance of displayed images under several illumination conditions based on a novel human visual system (HVS) model named as a rod and cone response (RCR) model. The brightness appearance of displayed images changes according to an adaptation of human eyes caused by a surrounding illumination and characteristics of a display device, which is defined as a device model, such as a surface reflectance and a contrast ratio. The proposed method based on the RCR and the device model can predict the brightness appearance more precisely and faster than conventional HVS models such as CIECAM02. Our prototype display device with the proposed method reduces the prediction error by 81% with 12% of a computational cost compared to CIECAM02. The proposed method improves image quality on the display device under any surrounding illumination conditions.