Realtime 2D deconvolution system and method
View Patent ↗A real time 2D deconvolution system and method for processing a time sequence, or video sequence, of microscopy images. A system is provided that includes a first algorithm for processing a first image in the set of 2D images, wherein the first algorithm calculates an estimated point spread function (PSF) by analyzing data in the first image; and a second algorithm for processing a set of subsequent images in the set of 2D images, wherein each subsequent image is processed based on the estimated PSF, under the assumption of slowly-varying image change. A PSF refinement system may optionally improve the PSF estimate, between each image, as computation time permits.
1. A system for processing a set of two dimensional (2D) images, comprising:
a computer adapted for processing data including the 2D images, wherein said computer executes
a first algorithm for processing a first image in the set of 2D images, wherein the computer executes the first algorithm to calculate an estimated point spread function (PSF) by analyzing data in the first image to perform a blind deconvolution; and a second algorithm for processing a set of subsequent images in the set of 2D images, wherein the computer processes each subsequent image based on said estimated PSF.
2. The system of claim 1 , wherein the first algorithm comprises: an iterative process for calculating the estimated PSF; and a termination algorithm for terminating the iterative process.
3. The system of claim 1 , wherein the set of 2D images are collected with an optics device, and wherein the first algorithm does not use information about the optics device to calculate the estimated PSF.
4. The system of claim 1 , wherein the first algorithm utilizes an expected maximization algorithm.
5. The system of claim 1 , wherein the second algorithm adaptively refines the estimated PSF after processing each of the subsequent images.
6. The system of claim 1 , wherein the second algorithm utilizes a Wiener filter.
7. The system of claim 6 , wherein the Weiner filter includes a constant k that can be adjusted to alter an amount of noise filtering.
8. The system of claim 1 , further comprising a scene change detection system for detecting changes in the set of 2D images.
9. The system of claim 8 , further comprising a frame averaging system for averaging images together to improve filtering.
10. The system of claim 8 , wherein the scene change detection system performs a terminate operation or an entire recalculation of the estimated PSF estimate if a substantial scene change is detected.
11. A physically tangible computer readable medium containing a representation of a program for processing a set of two dimensional (2D) images, comprising: program code configured for processing a first image in the set of 2D images using a first algorithm, wherein the first algorithm calculates an estimated point spread function (PSF) by analyzing data in the first image to perform a blind deconvolution; and program code configured for processing each of a plurality of subsequent images in the set of 2D images with a second algorithm, wherein each subsequent image is processed based on the estimated PSF.
12. The computer-readable medium of claim 11 , wherein the first algorithm comprises an iterative process for calculating the estimated PSF.
13. The computer-readable medium of claim 11 , wherein the set of 2D images are collected with an optics device, and wherein the first algorithm does not use information about the optics device to calculate the estimated PSF.
14. The computer-readable medium of claim 11 , wherein the first algorithm utilizes an expected maximization algorithm and the second algorithm utilizes a Wiener filter.
15. The computer-readable medium of claim 11 , wherein the second algorithm adaptively refines the estimated PSF after processing each of the subsequent images.
16. The computer-readable medium of claim 11 , wherein the Weiner filter includes a constant k that can be adjusted to alter an amount of filtering.
17. The computer-readable medium of claim 11 , further comprising a frame averaging system for averaging images together to improve filtering.
18. The computer-readable medium of claim 11 , further comprising a scene change detection system for detecting scene changes in the set of 2D images.