ABSTRACT
This paper presents a new method for image fusion with sub-band architecture for fusing the multiple-exposure images of natural scenes. In captured standard dynamic range (SDR) images with different exposure settings, fine details are extracted and preserved using the proposed method by transforming the pixel-based enhancement in the image using Dynamic Refined Adaptive Weight (DRAW) map. The sequence of the images is decomposed into different sub-bands with a dual tree discrete wavelet transform. We compute the DRAW map according to the appearance of the image based on features such as exposure, contrast and saturation. The coefficients of the sub-bands are blended into a high-quality image with the fine detail preserved. The experimental results are demonstrated, and the proposed results create visually pleasing images with the very sharp details preserved.
Disclosure statement
No potential conflict of interest was reported by the authors.