Parameter estimation in super-resolution image reconstruction problems

Abad, J.; Vega, Miguel; Molina, Rafael; Katsaggelos, Aggelos K.

doi:10.1109/icassp.2003.1199573

Cited by 8 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition to the above mentioned methods in the frequency domain, some other SR algorithms of this domain have borrowed the methods that have been usually used in the spatial domain; among them are: [119], [211], [321], [370], [589] which have used a Maximum Likelihood (ML) method (Section 5.1.5), [144], [178], [201] which have used a regularized ML method, [197], [221], [267], [491], [511], [567] which have used a MAP method (Section 5.1.6), and [141], [175] which have implemented a Projection Onto Convex Set (POCS) method (Section 5.1.4). These will all be explained in the next section.…”

Section: Wavelet Transformmentioning

confidence: 99%

Super-resolution: a comprehensive survey

Nasrollahi

Moeslund

2014

Machine Vision and Applications

600

292

View full text Add to dashboard Cite

Super-resolution, the process of obtaining one or more high-resolution images from one or more low-resolution observations, has been a very attractive research topic over the last two decades. It has found practical applications in many real world problems in different fields, from satellite and aerial imaging to medical image processing, to facial image analysis, text image analysis, sign and number plates reading, and biometrics recognition, to name a few. This has resulted in many research papers, each developing a new superresolution algorithm for a specific purpose. The current comprehensive survey provides an overview of most of these published works by grouping them in a broad taxonomy. For each of the groups in the taxonomy, the basic concepts of the algorithms are first explained and then the paths through which each of these groups have evolved are given in detail, by mentioning the contributions of different authors to the basic concepts of each group. Furthermore, common issues in super-resolution algorithms, such as imaging models and registration algorithms, optimization of the cost functions employed, dealing with color information, improvement factors, assessment of super-resolution algorithms, and the most commonly employed databases are discussed.

show abstract

Section: Wavelet Transformmentioning

confidence: 99%

Super-resolution: a comprehensive survey

Nasrollahi

Moeslund

2014

Machine Vision and Applications

600

292

View full text Add to dashboard Cite

show abstract

“…Therefore, these two transforms have sometimes been combined together into the Fourier-Wavelet Regularized Deconvolution [19]. In addition to the above mentioned methods in the frequency domain, some other SR algorithms of this domain have borrowed the methods that have been usually used in the spatial domain; among them are: [20], [21], [22] which have used a Maximum Likelihood (ML) method , [23], [24] which have used a regularized ML method, [25], [26] which have used a MAP method , and [27], [28] which have implemented a Projection Onto Convex Set (POCS) method.…”

Section: Wavelet Transformmentioning

confidence: 99%

Single and Multi frame Image super-resolution and its Performance Analysis: A Comprehensive Survey

Chopade¹,

Patil²

2015

IJCA

View full text Add to dashboard Cite

Super-resolution is the process of combining one or more low-resolution images to obtain a high-resolution image. It has been a very interesting topic for the research over the last few years. It is used for practical applications in more realistic problems faced in different areas, which range over satellite and aerial imaging of biomedical image processing as well as in daily routine life like different biometrics in offices and industries. In this survey article, more focus is given on basic algorithms and their classification based methodology used to implement it. Due to its vast scope of applications researchers are developing a novel suerresolution algorithm for a specific intention based on single and multiframe image resolution. The proposed comprehensive survey gives an overview of most of published works based on its performance analysis. In this survey, the basic concepts of the algorithms are explained and then their performance analyses through which each of these methods have developed are mentioned in detail. Furthermore, different issues in superresolution algorithms for single and multiframesuch as models and registration algorithms, optimization of the deployment of methods, improvement in quality of image .

show abstract

“…In the research of image restoration, the most traditional method is only using a single input image to reconstruct highresolution image [1, 2, 3,4]. Recently, the image restoration has been enhanced from a sequence of low-resolution images [5,6] because spatio-temporal information of the sequence provides priori knowledge to correct the restoration.…”

Section: Introductionmentioning

confidence: 99%

“…However, in this paper, the parameters estimation (model identification) is automatically and integrally achieved by using a statistic method -maximum likelihood (ML) estimator. On the other hand, the estimation of the hyperparameters based on bayesian analysis for an image restoration model has been proposed in [4]. These hyperparameters are considered as the coefficients to weight the model terms.…”

Section: Introductionmentioning

confidence: 99%

State-space model identification and Kalman filtering for image sequence restoration

Nishiyama

2008

2008 15th IEEE International Conference on Image Processing

View full text Add to dashboard Cite

A novel image restoration method is proposed to resolve a problem that the traditional restoration method performs poorly when the kind of image degradation model from highto low-resolution is unconfirmed. In this paper, the proposed method includes a conceptual frame of state space model (SSM) in order to achieve a general model for accurately estimating the high-resolution image sequence from its incomplete low-resolution observation sequence. Here the parameters of SSM are calculated by a statistic approach -maximum likelihood (ML) estimator. By using the most effective filter of SSM -Kalman filter to estimate, we find that the estimated image sequence is closer to the actual one than the bi-linear interpolation, so that the proposed method can be used to improve the restoration results.Index Terms-image sequence restoration, state space model, maximum likelihood estimator, Kalman filter.

show abstract

Parameter estimation in super-resolution image reconstruction problems

Cited by 8 publications

References 12 publications

Super-resolution: a comprehensive survey

Super-resolution: a comprehensive survey

Single and Multi frame Image super-resolution and its Performance Analysis: A Comprehensive Survey

State-space model identification and Kalman filtering for image sequence restoration

Contact Info

Product

Resources

About