Neural Adaptive Fractional Order Differential based Algorithm for Medical Image Enhancement

In this paper, we propose an adaptive fractional differential calculus based technique for image enhancement. The adaptive fractional order used in the fractional differential mask is computed through a neural network based scheme. The training of the neural network is achieved by using adaptive fractional orders calculated by means of AFDA (Adaptive Fractional Differential Approach) algorithm for different medical images. After training, the neural network calculates the appropriate adaptive fractional order that will be substituted in the mask to enhance the image. We perform some experiments on medical images then compare the enhancing performance with that of the AFDA algorithm, demonstrating that the proposed method leads to a better quality of enhanced images, giving rise to clearer edges and richer texture with less computational complexity.