Optimized Dynamic Stochastic Resonance framework for enhancement of structural details of satellite images

Image enhancement is an essential tool for increasing the contrast of an image to visualize the dark and bright areas. The enhancement algorithms are very much relevant in remote sensing applications as the satellite images are normally of poor contrast. The dynamic stochastic resonance (DSR) attains the enhancement of poor contrast and low illuminated images by utilizing the internal noise. The conventional DSR method employed for enhancing the dark images demands proper tuning of bistable element parameters and appropriate transform domain which are found to be challenging. In this paper, we propose chaotic grey wolf optimizer to attain the optimized parameters of dynamic stochastic resonance in non-sub sampled shearlet transform domain (NSST) to enhance the low contrast satellite images. In addition, we have tested the proposed method on a variety of satellite images captured by different sensors of local cities and global areas. The quality of the proposed method is compared with that of recent enhancement algorithms. The proposed method demonstrates to be the most reliable in enhancing the image structure contrast while preserving the true colors of satellite images. The source code and dataset is available in https://github.com/shyamfec/ODSRF.