- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2019 International Conference on Optical MEMS and Nanophotonics (OMN) - Daejeon, Korea (South) (2019.7.28-2019.8.1)] 2019 International Conference on Optical MEMS and Nanophotonics (OMN) - Operation verification of tunable plasmonic color filter composed by metal-insulator-metal subwavelength grating and MEMS actuator
摘要： Signal processing in light-microscopy and cell imaging is concerned with reconstructing latent ground truth from imperfect images. This typically requires assuming prior knowledge about the latent ground truth. While this assumption regularizes the problem to an extent where it can be solved, it also biases the result toward the expected. It thus often remains unclear what prior to use for a given practical problem. We argue here that the gradient distribution of natural-scene images may provide a versatile and well-founded prior for light-microscopy images that does not impose assumptions about the geometry of the ground-truth signal, but only about its gradient spectrum. We provide motivation for this choice from different points of view, and we illustrate the resulting regularizer for use on light-microscopy images. We provide a simple parametric model for the resulting prior, leading to ef?ciently solvable variational problems. We demonstrate the use of these models and solvers in a variety of common image-processing tasks, including contrast enhancement, noise-level estimation, denoising, blind deconvolution, and dehazing. We conclude by discussing the limitations and possible interpretations of the prior.
关键词： parametric prior,gradient distribution,noise-level estimation,dehazing,naturalization,denoising,Deconvolution,variational method
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - Sozopol, Bulgaria (2019.9.6-2019.9.8)] 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - 1,55 mkm fiber laser with electronic controlled mode-locking
摘要： Signal processing in light-microscopy and cell imaging is concerned with reconstructing latent ground truth from imperfect images. This typically requires assuming prior knowledge about the latent ground truth. While this assumption regularizes the problem to an extent where it can be solved, it also biases the result toward the expected. It thus often remains unclear what prior to use for a given practical problem. We argue here that the gradient distribution of natural-scene images may provide a versatile and well-founded prior for light-microscopy images that does not impose assumptions about the geometry of the ground-truth signal, but only about its gradient spectrum. We provide motivation for this choice from different points of view, and we illustrate the resulting regularizer for use on light-microscopy images. We provide a simple parametric model for the resulting prior, leading to ef?ciently solvable variational problems. We demonstrate the use of these models and solvers in a variety of common image-processing tasks, including contrast enhancement, noise-level estimation, denoising, blind deconvolution, and dehazing. We conclude by discussing the limitations and possible interpretations of the prior.
关键词： variational method,naturalization,gradient distribution,dehazing,parametric prior,noise-level estimation,denoising,Deconvolution
更新于2025-09-19 17:13:59