Research and analysis of deep learning image enhancement algorithm based on fractional differential

Abstract Aiming at the characteristics of the image processing technology course and the practicality, the deep learning and fractional differential wavelet algorithm are introduced into the image processing technology to solve the problem that the traditional algorithm loses the texture detail information during image enhancement. Through theoretical analysis, the fractional differential wavelet algorithm can greatly improve the high frequency components of the signal, enhance the IF component of the signal, and the very low frequency of the nonlinear retained signal. According to this, the application of fractional differential to image enhancement will make the edge of the image prominent. The enhanced image with clearer texture and image smoothing area information is preserved. Then, based on the classical fractional differential definition, the fractional difference equation is derived and the differential operator of the approximate depth learning method is constructed. Experiments with image enhancement show that the image enhancement method based on fractional differential operator is better than the traditional integer differential method.