Blind inverse gamma correction with maximized differential entropy

Unwanted nonlinear gamma distortion frequently occurs in a great diversity of images during the procedures image acquisition, processing, and/or display. And often varies with capture setup change and luminance variation. Blind inverse correction, which automatically determines proper restoration value from given image, is paramount importance to attenuate distortion. For blind an adaptive transformation method (AGT-ME) proposed directly maximized differential entropy model. corresponding optimization has mathematical concise closed-form solution, resulting efficient implementation accurate AGT-ME. Considering human eye non-linear perception sensitivity, modified version AGT-ME-VISUAL also achieve better visual performance. Tested on variable datasets, AGT-ME could obtain estimation large range (0.1 3.0), outperforming state-of-the-art methods. Besides, were applied three typical applications, including automatic adjustment, natural/medical contrast enhancement, fringe projection profilometry restoration. Furthermore, AGT-ME/ general can be seamlessly extended masked multi-channel (color or spectrum) multi-frame video, free arbitrary tuning parameter. Python code (https://github.com/yongleex/AGT-ME) provided for interested users.