Analysis and extensions of the Frankle-McCann Retinex algorithm

‹ Retinex Algorithm – Process of estimating the per-pixel gains necessary to achieve dynamic range compression in the original image ‹ Extension – Suppress artifacts – Reduce the number of iterations Abstract 4 / 29 ‹ High Dynamic Range – More accurate rendition of the original image – Realistically converted to a low resolution ‹ Retinex theory – Illuminant estimation, gamut mapping and HDR imaging – ratio-product-reset-average operation – From a maximum brightness default state to a final image that has dark region in the input image