res++: Characterization of imaging systems, image degradation, motion characterization
Description
We developped methods that improve the resolution of images (such as light microscopy images) by locally estimating image distortion. The method also allows to estimate an imaged object's distance from a camera's focal plane.
Our method allows estimating the parameters of a spatially variant Point Spread Function (PSF) model using a Convolutional Neural Network (CNN).
Publications
- Estimating Nonplanar Flow from 2D Motion-blurred Widefield Microscopy Images via Deep Learning - Idiap Publications
- DeepFocus: a Few-shot Microscope Slide Auto-Focus using a Sample Invariant CNN-based Sharpness Function - Idiap Publications
- Spatially-Variant CNN-Based Point Spread Function Estimation for Blind Deconvolution and Depth Estimation in Optical Microscopy - Idiap Publications
- Semi-Blind Spatially-Variant Deconvolution in Optical Microscopy with Local Point Spread Function Estimation by Use of Convolutional Neural Networks - Idiap Publications
Links
-
GitHub - idiap/semiblindpsfdeconv: Code for "Semi-Blind Spatially-Variant Deconvolution in Optical Microscopy with Local Point Spread Function Estimation By Use Of Convolutional Neural Networks" ICIP 2018
-
GitHub - idiap/deepfocus: Pytorch implementation of "DeepFocus: a Few-Shot Microscope Slide Auto-Focus using a Sample Invariant CNN-based Sharpness Function"; Adrian Shajkofci 2020
-
GitHub - idiap/psfestimation: Code for the PyTorch implementation of "Spatially-Variant CNN-based Point Spread Function Estimation for Blind Deconvolution and Depth Estimation in Optical Microscopy", IEEE Transactions on Image Processing, 2020.
-
GitHub - idiap/flowestimation: PyTorch implementation of "Estimating Nonplanar Flow from 2D Motion-blurred Widefield Microscopy Images via Deep Learning", submitted to IEEE ISBI, 2021
Advantages
The method does not require instrument- or object-specific calibration but can be applied , in many cases, directly during imaging.
Applications
- Image enhancement (deconvolution, deblurring)
- Depth estimation
- Flow estimation
Technology Readiness Level
TRL 4
Contact us for more information
- Interested in using our technologies?
- Interested to know more about the licensing possibilities and conditions?