2,235 Hits in 5.0 sec

Blind Image Deconvolution by Automatic Gradient Activation

Dong Gong, Mingkui Tan, Yanning Zhang, Anton van den Hengel, Qinfeng Shi
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We thus introduce a gradient activation method to automatically select a subset of gradients of the latent image in a cutting-plane-based optimization scheme for kernel estimation.  ...  Blind image deconvolution is an ill-posed inverse problem which is often addressed through the application of appropriate prior.  ...  Blind deconvolution, also known as blind deblurring, seeks to recover the latent sharp image x from the observed blurry image y.  ... 
doi:10.1109/cvpr.2016.202 dblp:conf/cvpr/GongTZHS16 fatcat:psojs6vfivaynkxbb3wrlvg5pe


C. Zelenka, R. Koch
2015 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
<br><br> In this paper we propose the compensation of defocus and motion blur in underwater images by using blind deconvolution techniques.  ...  These adverse conditions are typical for open sea underwater bubble images.  ...  The gradient sparsity MAP deconvolution takes 9s on the image in Figure 2 , while the non-blind deconvolution Richardson-Lucy algorithm with the blur kernel set by the blind deconvolution is much faster  ... 
doi:10.5194/isprsarchives-xl-5-w5-239-2015 fatcat:b27c4bud4be3zctyx5gstdb77e

Learning Fully Convolutional Networks for Iterative Non-blind Deconvolution [article]

Jiawei Zhang, Jinshan Pan, Wei-Sheng Lai, Rynson Lau, Ming-Hsuan Yang
2016 arXiv   pre-print
In this paper, we propose a fully convolutional networks for iterative non-blind deconvolution We decompose the non-blind deconvolution problem into image denoising and image deconvolution.  ...  We train a FCNN to remove noises in the gradient domain and use the learned gradients to guide the image deconvolution step.  ...  The denoised image gradients are treated as image priors to guide the image deconvolution in the next iteration. Denoising by FCNN.  ... 
arXiv:1611.06495v1 fatcat:5bkcuyzy6bdxlkxwgs3q56b2je

Learning Fully Convolutional Networks for Iterative Non-blind Deconvolution

Jiawei Zhang, Jinshan Pan, Wei-Sheng Lai, Rynson W. H. Lau, Ming-Hsuan Yang
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
In this paper, we propose a fully convolutional network for iterative non-blind deconvolution. We decompose the non-blind deconvolution problem into image denoising and image deconvolution.  ...  We train a FCNN to remove noise in the gradient domain and use the learned gradients to guide the image deconvolution step.  ...  This work is supported in part by the SRG grant from City University of Hong Kong (No. 7004416), the National Natural Science Foundation of China (No. 61572099 and 61320106008), the NSF Career Grant 1149783  ... 
doi:10.1109/cvpr.2017.737 dblp:conf/cvpr/ZhangPLL017 fatcat:j4itkj2sg5hhjmq53vzrlmve2e

Fast Blur Detection and Parametric Deconvolution of Retinal Fundus Images [chapter]

Bryan M. Williams, Baidaa Al-Bander, Harry Pratt, Samuel Lawman, Yitian Zhao, Yalin Zheng, Yaochun Shen
2017 Lecture Notes in Computer Science  
We present an automated approach to image deconvolution which assumes that the cause of blur belongs to a set of common types.  ...  Semi-blind deblurring is a useful technique which may be developed to restore the underlying sharp image given some assumed or known information about the cause of degradation.  ...  Acknowledgement This project is funded by the National Institute for Health Researchs i4i Programme. This paper summarises independent research funded by the National  ... 
doi:10.1007/978-3-319-67561-9_22 fatcat:5iu4bodctbexblmqhfrrtnbqgu

AIDA: an adaptive image deconvolution algorithm with application to multi-frame and three-dimensional data

Erik F. Y. Hom, Franck Marchis, Timothy K. Lee, Sebastian Haase, David A. Agard, John W. Sedat
2007 Optical Society of America. Journal A: Optics, Image Science, and Vision  
We describe an adaptive image deconvolution algorithm (AIDA) for myopic deconvolution of multi-frame and three-dimensional data acquired through astronomical and microscopic imaging.  ...  Written in Numerical Python with calls to a robust constrained conjugate gradient method, AIDA has significantly improved run times over the original MISTRAL implementation.  ...  We thank Eugene Ingerman [University of California, Davis, and Lawrence Livermore National Laboratory (LLNL)] and Stephen Lane (LLNL) for making the conjugate gradient code available to us.  ... 
doi:10.1364/josaa.24.001580 pmid:17491626 pmcid:PMC3166524 fatcat:7tm4v3mtgfhlhnnimf2nkhigbe

Understanding image priors in blind deconvolution

Filip Sroubek, Vaclav Smidl, Jan Kotera
2014 2014 IEEE International Conference on Image Processing (ICIP)  
Proper estimators together with correct image priors play a fundamental role in accurate blind deconvolution.  ...  a) blurred input (b) blind deconvolution Fig. 1.  ...  In this paper, we use VB approximation to solve blind deconvolution and automatically determine all parameters in image and blur priors including noise variance.  ... 
doi:10.1109/icip.2014.7025911 dblp:conf/icip/SroubekSK14 fatcat:rch2lgjkirfcna2hte72mxp2s4

A new image deconvolution method with fractional regularisation

Bryan M Williams, Jianping Zhang, Ke Chen
2016 Journal of Algorithms & Computational Technology  
In this paper, we propose a new regularisation technique for image reconstruction in the blind and non-blind deconvolution problems where the precise cause of blur may or may not be known.  ...  Image deconvolution is an important pre-processing step in image analysis which may be combined with denoising, also an important image restoration technique, and prepares the image to facilitate diagnosis  ...  In the section 'A blind deconvolution model with fractional regularisation', we extend this idea to the case of blind deconvolution.  ... 
doi:10.1177/1748301816660439 fatcat:pogzl2s4pnhnjejpd4ed5cm6pu

Spatially adaptive total variation deblurring with split Bregman technique

Mahdi Dodangeh, Isabel N. Figueiredo, Gil Gonçalves
2018 IET Image Processing  
In this paper we describe a modified non-blind and blind deconvolution model by introducing a regularization parameter that incorporates the spatial image information.  ...  Indeed, we have used a weighted total variation (TV) term, where the weight is a spatially adaptive parameter based on the image gradient. The proposed models are solved by the split Bregman method.  ...  (2) Semi-blind deconvolution, in which the kernel belongs to a known class of parametric functions. (3) Non-blind deconvolution where only the true image is unknown.  ... 
doi:10.1049/iet-ipr.2017.0302 fatcat:e3qqq5eu5zd7xhkq4ynhh47oiu

Adaptive blind signal processing-neural network approaches

S. Amari, A. Cichocki
1998 Proceedings of the IEEE  
Keywords-Blind deconvolution and equalization, blind separation of signals, independent component analysis (ICA), natural gradient learning, neural networks, self-adaptive learning rates, unsupervised  ...  Learning algorithms and underlying basic mathematical ideas are presented for the problem of adaptive blind signal processing, especially instantaneous blind separation and multichannel blind deconvolution  ...  It is effective to analyze the differential in terms of , since the natural Riemannian gradient [6] - [8] is automatically implemented by it and the equivariant properties investigated in [22] automatically  ... 
doi:10.1109/5.720251 fatcat:jg337aeuxnd3rec634qd3qjfde

Three-dimensional blind deconvolution of SPECT images

M. Mignotte, J. Meunier
2000 IEEE Transactions on Biomedical Engineering  
In order to improve the resolution of these 3-D images and then to facilitate their interpretation, we propose herein to extend a recent image blind deconvolution technique (called the nonnegativity support  ...  Index Terms-Image restoration, Markov random field (MRF) model, single photon emission computed tomography (SPECT) imagery, three-dimensional (3-D) blind deconvolution, unsupervised segmentation.  ...  Janicki of the Centre Hospitalier de l'Université de Montréal (CHUM), Montréal, Canada, for providing the SPECT images.  ... 
doi:10.1109/10.821781 pmid:10721635 fatcat:22gj7zsdojbrdot7howg2rsz64

Assessing the Sharpness of Satellite Images: Study of the PlanetScope Constellation [article]

Jérémy Anger, Carlo de Franchis, Gabriele Facciolo
2019 arXiv   pre-print
In this work, we quantify the sharpness of images from the PlanetScope constellation by estimating the blur kernel from each image.  ...  The method is fully blind and automatic, and since it does not require the knowledge of any satellite specifications it can be ported to other constellations.  ...  Estimating the blur kernel from a single image is an active field of research, especially for natural images since it is a necessary step of most blind deblurring methods [8, 9, 10] .  ... 
arXiv:1904.09159v1 fatcat:blvrve6hb5htpiwaopaivlersa

A Machine Learning Approach for Non-blind Image Deconvolution

Christian J. Schuler, Harold Christopher Burger, Stefan Harmeling, Bernhard Scholkopf
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
Abstract Image deconvolution is the ill-posed problem of recovering a sharp image, given a blurry one generated by a convolution. In this work, we deal with space-invariant nonblind deconvolution.  ...  This step amplifies and colors the noise, and corrupts the image information.  ...  However, in this case the deblurring quality is currently more limited by errors in the blur estimation than in the non-blind deconvolution step.  ... 
doi:10.1109/cvpr.2013.142 dblp:conf/cvpr/SchulerBHS13 fatcat:olxzg4c7cbdv7o6pyuxlv4wfey

Extreme Zoom Surveillance: System Design and Image Restoration

Yi Yao, Besma R. Abidi, Mongi A. Abidi
2007 Journal of Multimedia  
major Based on "Digital imaging with extreme zoom: system design and image restoration", by Yi Yao, Besma Abidi, and Mongi Abidi which appeared in the  ...  Considering both the speed of convergence and robustness to image degradations induced by high system magnifications and long observation distances, we introduce an auto-focusing algorithm based on sequential  ...  We also implemented unsharp masking, L-R image deconvolution, and maximum likelihood (ML) blind deconvolution [30] .  ... 
doi:10.4304/jmm.2.1.20-31 fatcat:mh2yccveljgjbozjz5whrlnahm

Single Image Deblurring Using Motion Density Functions [chapter]

Ankit Gupta, Neel Joshi, C. Lawrence Zitnick, Michael Cohen, Brian Curless
2010 Lecture Notes in Computer Science  
We use existing spatially invariant deconvolution methods in a local and robust way to compute initial estimates of the latent image.  ...  We show that 6D camera motion is well approximated by 3 degrees of motion (in-plane translation and rotation) and analyze the scope of this approximation.  ...  We would also like to thank Qi Shan for useful discussions regarding blind/non-blind image deblurring methods.  ... 
doi:10.1007/978-3-642-15549-9_13 fatcat:lbwdv4wnbjet5plcx75ts53mzy
« Previous Showing results 1 — 15 out of 2,235 results