Intelligent Remote Sensing image post-processing via two-level robust adaptive Neural Network computing

We address a new change detection/reconstruction fused Neural Network (NN) computing oriented approach for the conventional low resolution Remote Sensing (RS) radar and/or fractional Synthetic Aperture Radar (SAR) imagery enhancement. The collaborative considerations involve the user-controllable regularization degrees of freedom adaptive adjustment in two particular RS image formation schemes. First, we adapt the Hopfield NN computing methodology for feature enhancing image reconstruction, from the low resolution initial RS imagery. Second, the Pulse Coupled Neural Network (PCNN) framework is aggregated with the Hopfield NN method to perform the correct information detection in the resultant RS image. The addressed Modified Hopfield and Pulse Coupled Neural Network (MHPC-NN) technique processes the collaborative reconstruction/detection fused task computationally efficiently, ensuring on-line dynamic updates only for higher quality information. The reported simulations verify that the developed MHPC-NN fused technique outperforms the most recently proposed iterative enhancing radar/SAR imaging methods in the achievable resolution.