Variational Image Deblurring Using Modified Hopfield Neural Network

Zhang, Hongying; Peng, Qicong; Wu, Yue

doi:10.1109/icccas.2006.284669

Cited by 3 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A viable alternative which assures stability and uniqueness is regularization [11]. Machine learning (ML) approaches to image deblurring have inevitably emerged, mostly in the context of improving PSF estimates [12][13][14][15][16]. Figure 4 is a scale drawing of a cross-section of a wide-field-angle Ritchey-Chrétien telescope (RCT) consisting of a field-limiting forward baffle, a baffled hyperbolic primary mirror, a baffled hyperbolic secondary mirror, and a plane readout array.…”

Section: Image Deblurringmentioning

confidence: 99%

Numerical Focusing of a Wide-Field-Angle Earth Radiation Budget Imager Using an Artificial Neural Network

et al. 2020

View full text Add to dashboard Cite

Narrow field-of-view scanning thermistor bolometer radiometers have traditionally been used to monitor the earth’s radiant energy budget from low earth orbit (LEO). Such instruments use a combination of cross-path scanning and along-path spacecraft motion to obtain a patchwork of punctual observations which are ultimately assembled into a mosaic. Monitoring has also been achieved using non-scanning instruments operating in a push-broom mode in LOE and imagers operating in geostationary orbit. The current contribution considers a fourth possibility, that of an imager operating in LEO. The system under consideration consists of a Ritchey-Chrétien telescope illuminating a plane two-dimensional microbolometer array. At large field angles, the focal length of the candidate instrument is field-angle dependent, resulting in a blurred image in the readout plane. Presented is a full-field focusing algorithm based on an artificial neural network (ANN). Absorbed power distributions on the microbolometer array produced by discretized scenes are obtained using a high-fidelity Monte Carlo ray-trace (MCRT) model of the imager. The resulting readout array/scene pairs are then used to train an ANN. We demonstrate that a properly trained ANN can be used to convert the readout power distribution into an accurate image of the corresponding discretized scene. This opens the possibility of using an ANN based on a high-fidelity imager model for numerical focusing of an actual imager.

show abstract

Section: Image Deblurringmentioning

confidence: 99%

Numerical Focusing of a Wide-Field-Angle Earth Radiation Budget Imager Using an Artificial Neural Network

et al. 2020

View full text Add to dashboard Cite

show abstract

“…There we propose a multi-step deconvolution in multiple steps using directional gradient prior mathematically. The directional gradient is seamlessly integrated in to the least square optimization algorithm given in (2) and is solved iteratively with the directional gradients also contributing at each iteration. Using that we can approximate the solution as ^2 2…”

Section: Fig1 Block Diagram Of the Proposed Deblurring Algorithmmentioning

confidence: 99%

“…In this paper we model the above given ill-posed problem as one such optimization method and solve it as 22 x argmin y-Hx +λ Cx (2) The above equation known as L 2 -regularized minimization approach where the minimization happens for variable x. Through this minimization process we try to separate the variables H and x which otherwise appear as the convolution product (*) in equation (1).…”

Section: Introductionmentioning

confidence: 99%

Regularized Deblurring using Directional Prior with Sparse Representation

Dhar¹,

Srivastava²,

Maik³

2019

IJEAT

View full text Add to dashboard Cite

Blind deconvolution defined as simultaneous estimation and removal of blur is an ill-posed problem that can be solved with well-posed priors. In this paper we focus on directional edge prior based on orientation of gradients. Then the deconvolution problem is modeled as L2-regularized optimization problem which seeks a solution through constraint optimization. The constrained optimization problem is done in frequency domain with an Augmented Lagrangian Method (ALM). The proposed algorithm is tested on various synthetic as well as real data taken from various sources and the performance comparison is carried out with other state of the art existing methods.

show abstract

“…Z. Hongying et al [5] analyzed some techniques that working with neural network, to converge the recorded blurred image to the sharp image.…”

Section: Related Workmentioning

confidence: 99%

“…Order Statistics filters are nonlinear spatial filters which are based on ordering the pixels contained in an image. Usually, sliding window technique [4,5] is employed to perform pixel-by-pixel operation in a filtering algorithm. The local statistics obtained from the neighbourhood of the centre pixel gives a lot of information about its expected value.…”

Section: -D Order-statistic Filteringmentioning

confidence: 99%

Comparative Analysis of Image Deblurring Techniques

Singh¹,

Singh²

2016

IJCA

View full text Add to dashboard Cite

Image processing is a process to digitize the data of an image and a variety of mathematical operations are applied to enhance the image which is more applicable or pleasing to a human observer or to accomplish some of the analysis and identification tasks by computer. Due to environmental disarray or improper camera setting blur may arise in an image. Noise can also degrade the quality of a captured image in conjunction with blur. Restoration is a process to remove the blur from the image and restore the original image. There are a variety of techniques and methods have been suggested to restore a despoiled image. For a particular blur, there is a particular technique to remove it. In this paper, we discussed various image restoration methods and their study of efficiency. Image restoration has a variety of applications in various fields like video surveillance, crowed movement analysis etc..

show abstract

Variational Image Deblurring Using Modified Hopfield Neural Network

Cited by 3 publications

References 8 publications

Numerical Focusing of a Wide-Field-Angle Earth Radiation Budget Imager Using an Artificial Neural Network

Numerical Focusing of a Wide-Field-Angle Earth Radiation Budget Imager Using an Artificial Neural Network

Regularized Deblurring using Directional Prior with Sparse Representation

Comparative Analysis of Image Deblurring Techniques

Contact Info

Product

Resources

About