Visualization of High Dynamic Range Image with Retinex Algorithm

Nowadays, high dynamic range (HDR) radiance maps are becoming increasingly common and important in computer graphics. Meanwhile, HDR images of real scenes are very easy to construct. But the limitations of display devices are more and more exposed. This paper addresses the problem of high dynamic range scenes visualization and proposes the Retinex algorithm to enhance the rendering of high dynamic range images. The image is repeatedly processed with a ratio-product-reset-average operation. Applications of the algorithm to various high dynamic range color images yield promising results compared to the logarithmic transformation rendering algorithm. Moreover, variance and average gradient are proposed to evaluate the performance of this algorithm.