f-PICNN: A physics-informed convolutional neural network for partial differential equations with space-time domain.
Biao YuanHe WangAna HeitorXiaohui ChenPublished in: J. Comput. Phys. (2024)
Keyphrases
- partial differential equations
- convolutional neural network
- numerical solution
- anisotropic diffusion
- level set
- image processing
- image denoising
- image enhancement
- multiscale
- face detection
- mathematical morphology
- problems in image processing
- differential equations
- energy functional
- numerical algorithms
- numerical scheme
- frequency domain
- denoising
- fourth order
- parameter space
- finite difference method
- boundary element
- nonlinear partial differential equations
- numerical methods
- diffusion equation
- finite difference
- neural network
- difference equations
- matrix valued