Reconstruction of a high-resolution image by simultaneous registration, restoration, and interpolation of low-resolution images

Tom, Brian C.; Katsaggelos, Aggelos K.

doi:10.1109/icip.1995.537535

Cited by 123 publications

(69 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Registration and restoration estimated simultaneously by using maximum likelihood (ML) and the expectation maximization (EM). Later, the same authors included interpolation into the framework and estimated all of the unknowns using EM in [32]. In [33] a Joint MAP Restoration based method is introduced that simultaneously estimates HR image restoration and transmission parameters using an optimization process.…”

Section: Joint Map Restorationmentioning

confidence: 99%

A Survey on Super-Resolution Methods for Image Reconstruction

Karimi¹,

Kangarloo²,

Javadi³

2014

IJCA

View full text Add to dashboard Cite

Today, in many applications of Machine vision, image SuperResolution is preferred. Super-Resolution is estimation of a high-resolution image from an image or several low resolution images. Popular techniques in the field of enhancing images can be used to remove noise or blurring. In this paper, an overview of super resolution methods has been presented. Types of resolution methods have been used so far can be divided into three groups as frequency-domain methods, spatial domain methods and techniques can be classified as the wavelet domain. Super-resolution methods in different domains have different characteristics and comparison between these methods is usually done using a special index in one domain. In this paper, we will introduce these indexes and review best techniques used in all three domains.

show abstract

Section: Joint Map Restorationmentioning

confidence: 99%

A Survey on Super-Resolution Methods for Image Reconstruction

Karimi¹,

Kangarloo²,

Javadi³

2014

IJCA

View full text Add to dashboard Cite

show abstract

“…[6] More significantly with the comparison of the usual pixel based spatial data adaptive method, the method based on proposed region can do better i.e., in the super resolution process it helps in avoiding the effect of noise and maintains the robustness with changes in the intensity of noise. [10] First of all we have to take an original image and after that apply down sampling algorithm to design so that we can extract spatial information. After that we add noise to the sampled data such as Gaussian noise and then we can filter the data using clustering algorithm to calculate K means clustering of the image which is taken as a sample .…”

Section: Proposed Methodsmentioning

confidence: 99%

Image De-blurring and Supper-Resolution by Adaptive Sparse Domain Selection and Regularization

Banafar¹,

Kourav²

2016

IJCA

View full text Add to dashboard Cite

The method of total variation is used as a significant and competent image prior model in the regularization based area of image processing. However, as the model showing total variation supports a piecewise steady explanation, this process is classify below high intensity noise in the level areas of the picture is often poor, and a small number of pseudo edges are produced. Under this work we build up a spatially adaptive total variation model. Initially, we extract the spatial data based on each and every pixel, then two filtering process are combined to control the collision of pseudo edges. It also includes, the spatial information weight is build and classified with k-means clustering, and the cluster controls the center value of regularization strength in every region. The tentative results, of both simulated as well as genuine datasets, exhibit that the projected methodology can effectively diminish the pseudo edges formed by the total variation regularization in the flat areas, and maintain the limited smoothness of the HR images. The proposed region based spatial information adaptive variation model can effectively reduce the cause of noise on the spatial data extraction and maintain strength with changes in the noise intensity in the SR process as compare to traditional pixel based spatial information adaptive methodology.

show abstract

“…It is natural that the nadir image should be assigned the largest weighted value, and the weighted values of the other angle images should be decreased with the increase of the relative angle between the nadir image and the k-th angle image. Therefore, following the relationship between the spatial resolutions of the two images, we build the following weighting function: (9) where W k represents the weight for the k-th angle image, and θ k is the relative angle between the k-th angle image and the nadir image.…”

Section: Adaptive Weighting Methodsmentioning

confidence: 99%

“…The idea of SRR was first proposed in 1984 by Tsai and Huang [5] to improve the spatial resolution of Landsat TM images, using multiple under-sampled images with sub-pixel displacements in the frequency domain. Since then, the super-resolution reconstruction technique has developed greatly, and there have been various classical reconstruction frameworks proposed, such as the maximum a posteriori (MAP) [6], projection onto convex sets (POCS) [7], non-uniform interpolation [8], maximum likelihood [9,10], the iterative back-projection approach (IBP) [11], mixed maximum a posteriori/projection onto convex sets (MAP/POCS) [12], and so on. Generally speaking, the SRR methods in the frequency domain have a fast processing speed, but it is usually difficult to integrate the prior knowledge of the reconstruction image.…”

Section: Introductionmentioning

confidence: 99%

Super-Resolution Reconstruction for Multi-Angle Remote Sensing Images Considering Resolution Differences

et al. 2014

View full text Add to dashboard Cite

Multi-angle remote sensing images are acquired over the same imaging scene from different angles, and share similar but not identical information. It is therefore possible to enhance the spatial resolution of the multi-angle remote sensing images by the super-resolution reconstruction technique. However, different sensor shooting angles lead to different resolutions for each angle image, which affects the effectiveness of the super-resolution reconstruction of the multi-angle images. In view of this, we propose utilizing adaptive weighted super-resolution reconstruction to alleviate the limitations of the different resolutions. This paper employs two adaptive weighting themes. The first approach uses the angle between the imaging angle of the current image and that of the nadir image. The second is closely related to the residual error of each low-resolution angle image. The experimental results confirm the feasibility of the proposed method and demonstrate the effectiveness of the proposed adaptive weighted super-resolution approach.

show abstract

Reconstruction of a high-resolution image by simultaneous registration, restoration, and interpolation of low-resolution images

Cited by 123 publications

References 7 publications

A Survey on Super-Resolution Methods for Image Reconstruction

A Survey on Super-Resolution Methods for Image Reconstruction

Image De-blurring and Supper-Resolution by Adaptive Sparse Domain Selection and Regularization

Super-Resolution Reconstruction for Multi-Angle Remote Sensing Images Considering Resolution Differences

Contact Info

Product

Resources

About