The Possibility of AI-Derived Remote Sensing Simulation

Most of the cases of AI-derived remote sensing analysis are static or quasi-dynamic at most. The discussion in this letter seeks to prove the feasibility of purely deep learning (AI)-derived remote sensing simulation, and a dual purpose Hopfiled Neural Network (HNN) was proposed to this end. Our preliminary findings indicate a satisfactory recording of accuracy, with the averaged Kappa above 0.7 and FOM above 0.8.