Speckle2Void: Deep Self-Supervised SAR Despeckling With Blind-Spot Convolutional Neural Networks.
Andrea Bordone MoliniDiego ValsesiaGiulia FracastoroEnrico MagliPublished in: IEEE Trans. Geosci. Remote. Sens. (2022)
Keyphrases
- sar images
- convolutional neural networks
- synthetic aperture radar
- speckle noise
- convolutional network
- sar imagery
- sar data
- automatic target recognition
- ultrasound images
- wavelet domain
- sea ice
- multiplicative noise
- edge detector
- deep learning
- multiscale
- image reconstruction
- low contrast
- speckle reduction
- ultrasound imaging
- gray level