Physics-informed neural networks for inverse problems in supersonic flows.
Ameya D. JagtapZhiping MaoNikolaus AdamsGeorge Em KarniadakisPublished in: J. Comput. Phys. (2022)
Keyphrases
- inverse problems
- neural network
- image reconstruction
- global optimization
- convex optimization
- optimization problems
- pattern recognition
- optimization methods
- regularization method
- early vision
- partial differential equations
- emission tomography
- smoothness constraint
- genetic algorithm
- cost function
- evolutionary algorithm
- image analysis
- high quality
- computationally expensive
- multi view
- simulated annealing
- optimal solution
- multiscale
- image processing