Dissertations / Theses on the topic 'Image restoration'
To see the other types of publications on this topic, follow the link: Image restoration.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Image restoration.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Ungan, Cahit Ugur. "Nonlinear Image Restoration." Master's thesis, METU, 2005. http://etd.lib.metu.edu.tr/upload/2/12606796/index.pdf.
Full textAbstract:
This thesis analyzes the process of deblurring of degraded images generated by space-variant nonlinear image systems with Gaussian observation noise. The restoration of blurred images is performed by using two methodsa modified version of the Optimum Decoding Based Smoothing Algorithm and the Bootstrap Filter Algorithm which is a version of Particle Filtering methods. A computer software called MATLAB is used for performing the simulations of image estimation. The results of some simulations for various observation and image models are presented.
2
Dolne, Jean J. "Estimation theoretical image restoration." Thesis, Massachusetts Institute of Technology, 2008. http://hdl.handle.net/1721.1/47859.
Full textAbstract:
Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, 2008.Includes bibliographical references.
In this thesis, we have developed an extensive study to evaluate image restoration from a single image, colored or monochromatic. Using a mixture of Gaussian and Poisson noise process, we derived an objective function to estimate the unknown object and point spread function (psf) parameters. We have found that, without constraint enforcement, this blind deconvolution algorithm tended to converge to the trivial solution: delta function as the estimated psf and the detected image as the estimated object. We were able to avoid this solution set by enforcing a priori knowledge about the characteristics of the solution, which included the constraints on object sharpness, energy conservation, impulse response point spread function solution, and object gradient statistics. Applying theses constraints resulted in significantly improved solutions, as evaluated visually and quantitatively using the distance of the estimated to the true function. We have found that the distance of the estimated psf was correlated better with visual observation than the distance metric using the estimated object. Further research needs to be done in this area. To better pose the problem, we expressed the point spread function as a series of Gaussian basis functions, instead of the pixel basis function formalism used above. This procedure has reduced the dimensionality of the parameter space and has resulted in improved results, as expected. We determined a set of weights that yielded optimum algorithm performance.
(cont.) Additional research needs to be done to include the weight set as optimization parameters. This will free the user from having to adjust the weights manually. Of course, if certain knowledge of a weight is available, then it may be better to start with that as an initial guess and optimize from there. With the knowledge that the gradient of the object obeys long-tailed distribution, we have incorporated a constraint using the first two moments, mean and variance, of the gradient of the object in the objective function. Additional research should be done to incorporate the entire distribution in the objective and gradient functions and evaluate the performance.
by Jean J. Dolne.
S.M.
3
Pai, Hung-ta. "Multichannel blind image restoration /." Digital version accessible at:, 1999. http://wwwlib.umi.com/cr/utexas/main.
Full text
4
Reichenbach, Stephen Edward. "Small-kernel image restoration." W&M ScholarWorks, 1989. https://scholarworks.wm.edu/etd/1539623783.
Full textAbstract:
The goal of image restoration is to remove degradations that are introduced during image acquisition and display. Although image restoration is a difficult task that requires considerable computation, in many applications the processing must be performed significantly faster than is possible with traditional algorithms implemented on conventional serial architectures. as demonstrated in this dissertation, digital image restoration can be efficiently implemented by convolving an image with a small kernel. Small-kernel convolution is a local operation that requires relatively little processing and can be easily implemented in parallel. A small-kernel technique must compromise effectiveness for efficiency, but if the kernel values are well-chosen, small-kernel restoration can be very effective.;This dissertation develops a small-kernel image restoration algorithm that minimizes expected mean-square restoration error. The derivation of the mean-square-optimal small kernel parallels that of the Wiener filter, but accounts for explicit spatial constraints on the kernel. This development is thorough and rigorous, but conceptually straightforward: the mean-square-optimal kernel is conditioned only on a comprehensive end-to-end model of the imaging process and spatial constraints on the kernel. The end-to-end digital imaging system model accounts for the scene, acquisition blur, sampling, noise, and display reconstruction. The determination of kernel values is directly conditioned on the specific size and shape of the kernel. Experiments presented in this dissertation demonstrate that small-kernel image restoration requires significantly less computation than a state-of-the-art implementation of the Wiener filter yet the optimal small-kernel yields comparable restored images.;The mean-square-optimal small-kernel algorithm and most other image restoration algorithms require a characterization of the image acquisition device (i.e., an estimate of the device's point spread function or optical transfer function). This dissertation describes an original method for accurately determining this characterization. The method extends the traditional knife-edge technique to explicitly deal with fundamental sampled system considerations of aliasing and sample/scene phase. Results for both simulated and real imaging systems demonstrate the accuracy of the method.
5
Boukouvala, Erisso. "Image restoration techniques and application on astronomical images." Thesis, University of Reading, 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.414571.
Full text
6
Katsaggelos, Aggelos Konstantinos. "Constrained iterative image restoration algorithms." Diss., Georgia Institute of Technology, 1985. http://hdl.handle.net/1853/15830.
Full text
7
Huang, Yumei. "Numerical methods for image restoration." HKBU Institutional Repository, 2008. http://repository.hkbu.edu.hk/etd_ra/908.
Full text
8
Yan, Ruomei. "Adaptive representations for image restoration." Thesis, University of Sheffield, 2014. http://etheses.whiterose.ac.uk/6975/.
Full textAbstract:
In the field of image processing, building good representation models for natural images is crucial for various applications, such as image restoration, sampling, segmentation, etc. Adaptive image representation models are designed for describing the intrinsic structures of natural images. In the classical Bayesian inference, this representation is often known as the prior of the intensity distribution of the input image. Early image priors have forms such as total variation norm, Markov Random Fields (MRF), and wavelets. Recently, image priors obtained from machine learning techniques tend to be more adaptive, which aims at capturing the natural image models via learning from larger databases. In this thesis, we study adaptive representations of natural images for image restoration. The purpose of image restoration is to remove the artifacts which degrade an image. The degradation comes in many forms such as image blurs, noises, and artifacts from the codec. Take image denoising for an example. There are several classic representation methods which can generate state-of-the-art results. The first one is the assumption of image self-similarity. However, this representation has the issue that sometimes the self-similarity assumption would fail because of high noise levels or unique image contents. The second one is the wavelet based nonlocal representation, which also has a problem in that the fixed basis function is not adaptive enough for any arbitrary type of input images. The third is the sparse coding using over-complete dictionaries, which does not have the hierarchical structure that is similar to the one in human visual system and is therefore prone to denoising artifacts. My research started from image denoising. Through the thorough review and evaluation of state-of-the-art denoising methods, it was found that the representation of images is substantially important for the denoising technique. At the same time, an improvement on one of the nonlocal denoising methods was proposed, which improves the representation of images by the integration of Gaussian blur, clustering and Rotationally Invariant Block Matching. Enlightened by the successful application of sparse coding in compressive sensing, we exploited the image self-similarity by using a sparse representation based on wavelet coefficients in a nonlocal and hierarchical way, which generates competitive results compared to the state-of-the-art denoising algorithms. Meanwhile, another adaptive local filter learned by Genetic Programming (GP) was proposed for efficient image denoising. In this work, we employed GP to find the optimal representations for local image patches through training on massive datasets, which yields competitive results compared to state-of-the-art local denoising filters. After successfully dealing with the denoising part, we moved to the parameter estimation for image degradation models. For instance, image blur identification uses deep learning, which has recently been proposed as a popular image representation approach. This work has also been extended to blur estimation based on the fact that the second step of the framework has been replaced with general regression neural network. In a word, in this thesis, spatial correlations, sparse coding, genetic programming, deep learning are explored as adaptive image representation models for both image restoration and parameter estimation. We conclude this thesis by considering methods based on machine learning to be the best adaptive representations for natural images. We have shown that they can generate better results than conventional representation models for the tasks of image denoising and deblurring.
9
Sandor, Viviana. "Wavelet-based digital image restoration." W&M ScholarWorks, 1998. https://scholarworks.wm.edu/etd/1539623937.
Full textAbstract:
Digital image restoration is a fundamental image processing problem with underlying physical motivations. A digital imaging system is unable to generate a continuum of ideal pointwise measurements of the input scene. Instead, the acquired digital image is an array of measured values. Generally, algorithms can be developed to remove a significant part of the error associated with these measure image values provided a proper model of the image acquisition system is used as the basis for the algorithm development. The continuous/discrete/continuous (C/D/C) model has proven to be a better alternative compared to the relatively incomplete image acquisition models commonly used in image restoration. Because it is more comprehensive, the C/D/C model offers a basis for developing significantly better restoration filters. The C/D/C model uses Fourier domain techniques to account for system blur at the image formation level, for the potentially important effects of aliasing, for additive noise and for blur at the image reconstruction level.;This dissertation develops a wavelet-based representation for the C/D/C model, including a theoretical treatment of convolution and sampling. This wavelet-based C/D/C model representation is used to formulate the image restoration problem as a generalized least squares problem. The use of wavelets discretizes the image acquisition kernel, and in this way the image restoration problem is also discrete. The generalized least squares problem is solved using the singular value decomposition. Because image restoration is only meaningful in the presence of noise, restoration solutions must deal with the issue of noise amplification. In this dissertation the treatment of noise is addressed with a restoration parameter related to the singular values of the discrete image acquisition kernel. The restoration procedure is assessed using simulated scenes and real scenes with various degrees of smoothness, in the presence of noise. All these scenes are restoration-challenging because they have a considerable amount of spatial detail at small scale. An empirical procedure that provides a good initial guess of the restoration parameter is devised.
10
Ahtaiba, Ahmed Mohamed A. "Restoration of AFM images using digital signal and image processing." Thesis, Liverpool John Moores University, 2013. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.604322.
Full textAbstract:
All atomic force microscope (AFM) images suffer from distortions, which are principally produced by the interaction between the measured sample and the AFM tip. If the three-dimensional shape of the tip is known, the distorted image can be processed and the original surface form ' restored' typically by deconvolution approaches. This restored image gives a better representation of the real 3D surface or the measured sample than the original distorted image. In this thesis, a quantitative investigation of using morphological deconvolution has been used to restore AFM images via computer simulation using various computer simulated tips and objects. This thesis also presents the systematic quantitative study of the blind tip estimation algorithm via computer simulation using various computer simulated tips and objects. This thesis proposes a new method for estimating the impulse response of the AFM by measuring a micro-cylinder with a-priori known dimensions using contact mode AFM. The estimated impulse response is then used to restore subsequent AFM images, when measured with the same tip, under similar measurement conditions. Significantly, an approximation to what corresponds to the impulse response of the AFM can be deduced using this method. The suitability of this novel approach for restoring AFM images has been confirmed using both computer simulation and also with real experimental AFM images. This thesis suggests another new approach (impulse response technique) to estimate the impulse response of the AFM. this time from a square pillar sample that is measured using contact mode AFM. Once the impulse response is known, a deconvolution process is carried out between the estimated impulse response and typical 'distorted' raw AFM images in order to reduce the distortion effects. The experimental results and the computer simulations validate the performance of the proposed approach, in which it illustrates that the AFM image accuracy has been significantly improved. A new approach has been implemented in this research programme for the restoration of AFM images enabling a combination of cantilever and feedback signals at different scanning speeds. In this approach, the AFM topographic image is constructed using values obtained by summing the height image that is used for driving the Z-scanner and the deflection image with a weight function oc that is close to 3. The value of oc has been determined experimentally using tri al and error. This method has been tested 3t ten different scanning speeds and it consistently gives more faithful topographic images than the original AFM images.
11
Li, Wai-Mo 1964. "Sensor modeling and image restoration for a CCD pushbroom imager." Thesis, The University of Arizona, 1987. http://hdl.handle.net/10150/276601.
Full textAbstract:
Following the development of detector technology, remote sensing image detection is being implemented with charge-coupled devices (CCD), which have promising features. The French SPOT system is the first civilian satellite sensor employing a CCD in its detection unit. In order to obtain the system transfer function (TF), a linear system model is developed in the across- and along-track directions. The overall system TF, including pixel sampling effects, is then used in the Wiener filter function to derive an optimal restoration function. A restoration line spread function (RLSF) is obtained by the inverse Fourier transform of the Wiener filter and multiplied with a window function. Simulation and empirical tests are described comparing the RLSF to standard kernels used for image resampling for geometric correction. The RLSF results in superior edge enhancement as expected.
12
Jammal, Ghada. "Multiscale image restoration in nuclear medicine." Phd thesis, [S.l.] : [s.n.], 2001. http://elib.tu-darmstadt.de/diss/000100/GJammal.pdf.
Full text
13
May, Kaaren Lonna. "Blind image restoration via constrained optimisation." Thesis, Imperial College London, 1999. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.313788.
Full text
14
Kwan, Chun-kit, and 關進傑. "Fast iterative methods for image restoration." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2000. http://hub.hku.hk/bib/B31224520.
Full text
15
Lee, Richard. "3D non-linear image restoration algorithms." Thesis, University of East Anglia, 1996. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.338227.
Full text
16
Morris, Robin David. "Image sequence restoration using Gobbs distributions." Thesis, University of Cambridge, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.387724.
Full text
17
Morris, Octavius John. "Image restoration using composite signal models." Thesis, Imperial College London, 1986. http://hdl.handle.net/10044/1/38111.
Full text
18
Pryce, Jonathan Michael. "The statistical mechanics of image restoration." Thesis, University of Edinburgh, 1993. http://hdl.handle.net/1842/12805.
Full textAbstract:
Image restoration is concerned with the recovery of an 'improved' image from a corrupted picture, utilizing a prior model of the source and noise processes. We present a Bayesian derivation of the posterior probability distribution, which describes the relative probabilities that a certain image was the original source, given the corrupted picture. The ensemble of such restored images is modelled as a Markov random field (Ising spin system). Using a prior on the density of edges in the source, we obtain the cost function of Geman and Geman via information theoretic arguments. Using a combination of Monte Carlo simulation, the mean field approximation, and series expansion methods, we investigate the performance of the restoration scheme as a function of the parameters we have identified in the posterior distribution. We find phase transitions separating regions in which the posterior distribution is data-like, from regions in which it is prior-like, and we can explain these sudden changes of behaviour in terms of the relative free energies of metastable states. We construct a measure of the quality of the posterior distribution and use this to explore the way in which the appropriateness of the choice of prior affects the performance of the restoration. The data-like and prior-like characteristics of the posterior distribution indicate the regions of parameter space where the restoration scheme is effective and ineffective respectively. We examine the question of how best to use the posterior distribution to prescribe a single 'optimal' restored image. We make a detailed comparison of two different estimators to determine which better characterizes the posterior distribution. We propose that the TPM estimate, based on the mean of the posterior, is a more sensible choice than the MAP estimate (the mode of the posterior), both in principle and in practice, and we provide several practical and theoretical arguments in support.
19
Kwan, Chun-kit. "Fast iterative methods for image restoration /." Hong Kong : University of Hong Kong, 2000. http://sunzi.lib.hku.hk:8888/cgi-bin/hkuto%5Ftoc%5Fpdf?B22956281.
Full text
20
Wu, Hsien-Huang. "Image restoration for improved spectral unmixing." Diss., The University of Arizona, 1992. http://hdl.handle.net/10150/186114.
Full textAbstract:
Because of the resolution limitations in remote sensing, the radiance recorded by the detector at each pixel is the integrated sum of the spectral radiance of all materials within the detector instantaneous-field-of-view (IFOV). If the detector IFOV covers more than one object class, the radiance detected is not characteristic of any single class but a mixture of all classes. These mixed pixels will have spectral signatures that fall within the convex hull formed by the signatures of all the classes. Traditional classifiers are therefore usually left with many misclassified or unclassified pixels. To remedy this problem, unmixing algorithms which decompose each pixel into a combination of several classes have been successfully applied to estimate the percentage of each class inside one pixel. In this dissertation, unmixing error of the least squares unmixing algorithm that is caused by the intrinsic data variance, system PSF blurring, detection noise, and band-to-band misregistration is analyzed and evaluated. For high unmixing accuracy, image restoration is proposed to remove the PSF blurring degradation. To objectively assess the restoration performance and expedite the design of our application-oriented restoration scheme, and objective criterion based on the measurement of spectral fidelity in frequency domain is suggested. Based on this criterion, a detailed comparison between the conventional Wiener filter and sampled Wiener filter is conducted, which highlights the significance of sampling aliasing and verifies the results obtained visually by other researchers. Our study shows that contrary to restoration for visual purposes, a partial restoration scheme, instead of full restoration, should be used for a better unmixing performance. Also, the sampling aliasing, which is an artifact and should be suppressed in traditional restoration application, is actually a signal component which needs to be restored for unmixing. Under fair SNR conditions ($\ge$30dB), the proposed restoration scheme can reduce the total unmixing error up to 40% to 70% depending on the scene complexity.
21
Miller, Casey Lee. "Image restoration using trellis-search methods." Diss., The University of Arizona, 1999. http://hdl.handle.net/10150/288963.
Full textAbstract:
Methods for the restoration of images corrupted by blur and noise are presented. During transmission through an optical or electrical channel, images become corrupted by blur and noise as a result of channel limitations (i.e. optical aberrations or a bandlimit). If images are treated as a matrix whose elements (pixels) assume a finite number of values then there is a large but finite set of possible images that can be transmitted. By treating this finite set as a 'signal' set, digital communications methods may be used to estimate the uncorrupted image given a blurred and noisy version. Specifically, row-by-row estimation, decision-feedback and vector-quantization are used to extend the 1D sequence estimation ability of the a-posteriori probability (APP) and Viterbi algorithm (VA) to the estimation of 2D images. Simulations show the 2D VA and APP algorithms return near-optimal estimates of binary images as well as improved estimates of greyscale images when compared with the conventional Wiener filter (WF) estimates. Unlike the WF, the VA and APP algorithms are shown to be capable of super-resolution and adaptable for use with signal-dependent Poisson noise corruption. Restorations of experimental data gathered from an optical imaging system are presented to support simulation results.
22
Hazra, Rajeeb. "Constrained least-squares digital image restoration." W&M ScholarWorks, 1995. https://scholarworks.wm.edu/etd/1539623865.
Full textAbstract:
The design of a digital image restoration filter must address four concerns: the completeness of the underlying imaging system model, the validity of the restoration metric used to derive the filter, the computational efficiency of the algorithm for computing the filter values and the ability to apply the filter in the spatial domain. Consistent with these four concerns, this dissertation presents a constrained least-squares (CLS) restoration filter for digital image restoration. The CLS restoration filter is based on a comprehensive, continuous-input/discrete- processing/continuous-output (c/d/c) imaging system model that accounts for acquisition blur, spatial sampling, additive noise and imperfect image reconstruction. The c/d/c model-based CLS restoration filter can be applied rigorously and is easier to compute than the corresponding c/d/c model-based Wiener restoration filter. The CLS restoration filter can be efficiently implemented in the spatial domain as a small convolution kernel. Simulated restorations are used to illustrate the CLS filter's performance for a range of imaging conditions. Restoration studies based, in part, on an actual Forward Looking Infrared (FLIR) imaging system, show that the CLS restoration filter can be used for effective range reduction. The CLS restoration filter is also successfully tested on blurred and noisy radiometric images of the earth's outgoing radiation field from a satellite-borne scanning radiometer used by the National Aeronautics and Space Administration (NASA) for atmospheric research.
23
Yau, Chin-ko, and 游展高. "Super-resolution image restoration from multiple decimated, blurred and noisy images." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2004. http://hub.hku.hk/bib/B30292529.
Full text
24
Chana, Deeph S. "Image restoration exploiting statistical models of the image capture process." Thesis, King's College London (University of London), 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.246886.
Full text
25
Tran, Dai viet. "Patch-based Bayesian approaches for image restoration." Thesis, Sorbonne Paris Cité, 2018. http://www.theses.fr/2018USPCD049.
Full textAbstract:
Les travaux présentés dans cette thèse concernent les approches bayésiennes par patchs des problèmes d’amélioration de la qualité d’images. Notre contribution réside en le choix du dictionnaire construit grâce à un ensemble d’images de haute qualité et en la définition et l’utilisation d’un modèle à priori pour la distribution des patchs dans l’espace du dictionnaire. Nous avons montré qu’un choix attentif du dictionnaire représentant les informations locales des images permettait une amélioration de la qualité des images dégradées. Plus précisément, d’un dictionnaire construit de façon exhaustive sur les images de haute qualité nous avons sélectionné, pour chaque patch de l’image dégradée, un sous dictionnaire fait de ses voisins les plus proches. La similarité entre les patchs a été mesurée grâce à l’utilisation de la distance du cantonnier (Earth Mover’s Distance) entre les distributions des intensités de ces patchs. L’algorithme de super résolution présenté a conduit à de meilleurs résultats que les algorithmes les plus connus. Pour les problèmes de débruitage d’images nous nous sommes intéressés à la distribution à priori des patchs dans l’espace du dictionnaire afin de l’utiliser comme pré requis pour régulariser le problème d’optimisation donné par le Maximum à Posteriori. Dans le cas d’un dictionnaire de petite dimension, nous avons proposé une distribution constante par morceaux. Pour les dictionnaires de grande dimension, la distribution à priori a été recherchée comme un mélange de gaussiennes (GMM). Nous avons finalement justifié le nombre de gaussiennes utiles pour une bonne reconstruction apportant ainsi un nouvel éclairage sur l’utilisation des GMMIn this thesis, we investigate the patch-based image denoising and super-resolution under the Bayesian Maximum A Posteriori framework, with the help of a set of high quality images which are known as standard images. Our contributions are to address the construction of the dictionary, which is used to represent image patches, and the prior distribution in dictionary space. We have demonstrated that the careful selection of dictionary to represent the local information of image can improve the image reconstruction. By establishing an exhaustive dictionary from the standard images, our main attribute is to locally select a sub-dictionary of matched patches to recover each patch in the degraded image. Beside the conventional Euclidean measure, we propose an effective similarity metric based on the Earth Mover's Distance (EMD) for image patch-selection by considering each patch as a distribution of image intensities. Our EMD-based super-resolution algorithm has outperformed comparing to some state-of-the-art super-resolution methods.To enhance the quality of image denoising, we exploit the distribution of patches in the dictionary space as a an image prior to regularize the optimization problem. We develop a computationally efficient procedure, based on piece-wise constant function estimation, for low dimension dictionaries and then proposed a Gaussian Mixture Model (GMM) for higher complexity dictionary spaces. Finally, we justify the practical number of Gaussian components required for recovering patches. Our researches on multiple datasets with combination of different dictionaries and GMM models have complemented the lack of evidence of using GMM in the literature
26
Veldhuizen, Todd Lawrence. "Grid filters for local nonlinear image restoration /." Waterloo, Ont. : University of Waterloo [Dept. of Systems Design Engineering], 1998. http://etd.uwaterloo.ca/etd/tveldhui1998.pdf.
Full textAbstract:
Thesis (M.A.Sc.)-University of Waterloo, 1998.Includes bibliographical references (leaves 109-115). Issued also in PDF format and available via the World Wide Web. Requires Internet connectivity, World Wide Web browser, and Adobe Acrobat Reader.
27
Veldhuizen, Todd. "Grid Filters for Local Nonlinear Image Restoration." Thesis, University of Waterloo, 1998. http://hdl.handle.net/10012/943.
Full textAbstract:
A new approach to local nonlinear image restoration is described, based on approximating functions using a regular grid of points in a many-dimensional space. Symmetry reductions and compression of the sparse grid make it feasible to work with twelve-dimensional grids as large as 2212. Unlike polynomials and neural networks whose filtering complexity per pixel is linear in the number of filter co-efficients, grid filters have O(1) complexity per pixel. Grid filters require only a single presentation of the training samples, are numerically stable, leave unusual image features unchanged, and are a superset of order statistic filters. Results are presented for additive noise, blurring, and superresolution.
28
Langari, Bahareh. "Multi-scale edge-guided image gap restoration." Thesis, Brunel University, 2016. http://bura.brunel.ac.uk/handle/2438/13406.
Full textAbstract:
The focus of this research work is the estimation of gaps (missing blocks) in digital images. To progress the research two main issues were identified as (1) the appropriate domains for image gap restoration and (2) the methodologies for gap interpolation. Multi-scale transforms provide an appropriate framework for gap restoration. The main advantages are transformations into a set of frequency and scales and the ability to progressively reduce the size of the gap to one sample wide at the transform apex. Two types of multi-scale transform were considered for comparative evaluation; 2-dimensional (2D) discrete cosines (DCT) pyramid and 2D discrete wavelets (DWT). For image gap estimation, a family of conventional weighted interpolators and directional edge-guided interpolators are developed and evaluated. Two types of edges were considered; ‘local’ edges or textures and ‘global’ edges such as the boundaries between objects or within/across patterns in the image. For local edge, or texture, modelling a number of methods were explored which aim to reconstruct a set of gradients across the restored gap as those computed from the known neighbourhood. These differential gradients are estimated along the geometrical vertical, horizontal and cross directions for each pixel of the gap. The edge-guided interpolators aim to operate on distinct regions confined within edge lines. For global edge-guided interpolation, two main methods explored are Sobel and Canny detectors. The latter provides improved edge detection. The combination and integration of different multi-scale domains, local edge interpolators, global edge-guided interpolators and iterative estimation of edges provided a variety of configurations that were comparatively explored and evaluated. For evaluation a set of images commonly used in the literature work were employed together with simulated regular and random image gaps at a variety of loss rate. The performance measures used are the peak signal to noise ratio (PSNR) and structure similarity index (SSIM). The results obtained are better than the state of the art reported in the literature.
29
Palmer, Alexander S. "Adaptive image restoration algorithms using intelligent techniques." Thesis, University of East Anglia, 2003. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.405233.
Full text
30
Talebi-Rafsanjan, Siamak. "Image restoration techniques for bursty erasure channels." Thesis, King's College London (University of London), 2003. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.406409.
Full text
31
Oliveira, V. A. "Maximum entropy image restoration in nuclear medicine." Thesis, University of Southampton, 1989. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.235282.
Full text
32
Vitoria, Carrera Patricia. "On data-driven models for image restoration." Doctoral thesis, Universitat Pompeu Fabra, 2021. http://hdl.handle.net/10803/672662.
Full textAbstract:
Restoration of a high-quality image from a degraded recording is an important problem in
early vision processing. In this thesis, we tackle three image restoration problems: image
inpainting, colorization, and motion blur kernel estimation for deblurring.
In the first part, we present a timeline evolution of the inpainting research using deep
learning approaches. We analyze the different approaches that have been developed for
image inpainting and test them in the context of art restoration. Additionally, we propose an
automatic semantic inpainting method able to reconstruct corrupted information of an image
by semantically interpreting the image itself. Moreover, we address the problem of automatic
detection of the regions in the image where the information is corrupted by particular lens
artifacts, namely, spot flares, and finally their reconstruction via inpainting.
In the second part, we propose an automatic colorization approach based on adversarial
learning coupled with semantic information able to infer one colorization solution for a given
image. Qualitative and quantitative results show the capacity of the proposed method to
colorize images in a realistic way achieving state-of-the-art results.
Lastly, in the third part, we propose a general, non-parametric model for dense non-uniform
motion blur estimation. Given a blurry image, a set of adaptive basis kernels, as well as the
mixing coefficients at the pixel level, are estimated, producing a per-pixel map of motion blur.
This rich but efficient forward model of the degradation process allows the utilization of
existing tools for solving inverse problems.La restauració d'una imatge d'alta qualitat a partir d'una versió degradada és un problema important en el processament de la visió a etapes primerenques. En aquesta tesi, abordem tres problemes de restauració d'imatges: inpanting, colorització i estimació dels nuclis de borrositat per moviment per a la reducció de la borrositat. En la primera part, presentem una evolució temporal dels mètodes d'inpainting basats en aprenentatge profund. Analitzem els diferents enfocs que s'han desenvolupat i els testegem en el context de la restauració d'art. També, proposem un mètode automàtic per inpainting capaç de reconstruir la informació corrupta d'una imatge interpretant semànticament la pròpia imatge. A més, abordem el problema de la detecció automàtica de les regions en la imatge on la informació està corrompuda per artefactes d'un tipus particular anomenat "flare spot", i finalment els reconstruïm mitjançant un algoritme d'inpainting. En la segona part, proposem un algoritme de colorització automàtica basat en aprenentatge adversari juntament amb la incorporació d'informació semàntica. El mètode proposat és capaç d'estimar una de les moltes possibles solucions. Els resultats qualitatius i quantitatius mostren la capacitat del mètode proposat per coloritzar imatges de manera realista aconseguint resultats competitius i de l'estat de l'art. Finalment, en la tercera part, proposem un model general no paramètric per a l'estimació densa dels nuclis de borrositat de moviment no uniformes per a la reducció de la borrositat. Donada una imatge borrosa, s'estimen un conjunt de nuclis de base adaptatius a la imatge donada, així com els coeficients de la barreja a nivell de píxel, produïnt un mapa per píxel de desenfocament pel moviment. Aquest model complet i eficient del procés de degradació permet l'utilització d'eines existents per a resoldre problemes inversos.
La restauración de una imagen de alta calidad a partir de una versión degradada es un problema importante en el procesamiento de la visión en etapas tempranas. En esta tesis, abordamos tres problemas de restauración de imágenes: inpanting , colorización y estimación de los núcleos de borrosidad por movimiento para la reducción de la borrosidad. En la primera parte, presentamos una evolución temporal de los métodos de inpainting basados en aprendizaje profundo. Analizamos los diferentes enfoques que se han desarrollado y los probamos en el contexto de la restauración de arte. Además, proponemos un método automático para inpainting capaz de reconstruir la información corrupta de una imagen interpretando semánticamente la propia imagen. Además, abordamos el problema de la detección automática de las regiones en la imagen donde la información está corrompida por artefactos de lentes particulares llamados "flare spot" y finalmente se reconstruyen mediante un algoritmo de inpainting. En la segunda parte, proponemos un algoritmo de colorización automática basado en aprendizaje adversario junto con la incorporación de información semántica. El algoritmo es capaz de estimar una de las múltiples posibles soluciones. Los resultados cualitativos y cuantitativos muestran la capacidad del método propuesto para colorear imágenes de manera realista logrando resultados competitivos con el estado del arte. Por último, en la tercera parte, proponemos un modelo general no paramétrico para la estimación densa de los núcleo de movimiento no uniformes para la reducción de la borrosidad. Dada una imagen borrosa, se estiman un conjunto de núcleos de base adaptativos a la imagen dada, así como los coeficientes de mezcla a nivel de píxel, produciendo un mapa por píxel de desenfoque de movimiento. Este modelo completo y eficiente del proceso de degradación permite la utilización de herramientas existentes para resolver problemas inversos.
33
Liao, Haiyong. "Computational methods for bioinformatics and image restoration." HKBU Institutional Repository, 2010. http://repository.hkbu.edu.hk/etd_ra/1103.
Full text
34
Fong, Wai Lam. "Numerical methods for classification and image restoration." HKBU Institutional Repository, 2013. http://repository.hkbu.edu.hk/etd_ra/1488.
Full text
35
Goda, Matthew. "Wavelet domain image restoration and super-resolution." Diss., The University of Arizona, 2002. http://hdl.handle.net/10150/289808.
Full textAbstract:
Multi-resolution techniques, and especially the wavelet transform provide unique benefits in image representation and processing not otherwise possible. While wavelet applications in image compression and denoising have become extremely prevalent, their use in image restoration and super-resolution has not been exploited to the same degree. One issue is the extension 1-D wavelet transforms into 2-D via separable transforms versus the non-separability of typical circular aperture imaging systems. This mismatch leads to performance degradations. Image restoration, the inverse problem to image formation, is the first major focus of this research. A new multi-resolution transform is presented to improve performance. The transform is called a Radially Symmetric Discrete Wavelet-like Transform (RS-DWT) and is designed based on the non-separable blurring of the typical incoherent circular aperture imaging system. The results using this transform show marked improvement compared to other restoration algorithms both in Mean Square Error and visual appearance. Extensions to the general algorithm that further improve results are discussed. The ability to super-resolve imagery using wavelet-domain techniques is the second major focus of this research. Super-resolution, the ability to reconstruct object information lost in the imaging process, has been an active research area for many years. Multiple experiments are presented which demonstrate the possibilities and problems associated with super-resolution in the wavelet-domain. Finally, super-resolution in the wavelet domain using Non-Linear Interpolative Vector Quantization is studied and the results of the algorithm are presented and discussed.
36
Achddou, Raphaël. "Synthetic learning for neural image restoration methods." Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAT006.
Full textAbstract:
La photographie occupe aujourd'hui une place prépondérante dans nos vies. De plus, les attentes en terme de qualité des images augmentent tandis que la taille des appareils imageurs diminuent. Dans ce contexte, l'amélioration des algorithmes de traitement d'image est primordial.Dans ce manuscrit, on s'intéresse particulièrement aux tâches de restauration des images. Le but est de produire une image propre à partir d'une ou plusieurs observations bruitées de la même scène. Pour ces problèmes, les méthodes d'apprentissage profond ont connu un essor spectaculaire dans la dernière décennie, surpassant l'état de l'art pour la grande majorité des tests traditionnels.Bien que ces méthodes produisent des résultats impressionnants, elles présentent un certain nombre d'inconvénients. Tout d'abord, elles sont difficilement interprétables de part leur fonctionnement “boite noire”. De plus, elles généralisent assez mal à des modalités d'acquisition ou de distorsion absentes de la base de donnée d'apprentissage. Enfin, elles nécessitent des bases de données volumineuses, qui sont parfois difficile à acquérir.On se propose d'attaquer ces différents problèmes en remplaçant l'acquisition des données par un algorithme simple de génération de d'image, basé sur le modèle feuilles mortes. Bien que ce modèle soit très simple, les images générées ont des propriétés statistiques proches de celles des images naturelles et de nombreuses propriétés d'invariances (échelle, translation, rotation, contraste…). Entraîner un réseau de restauration avec ce genre d'image nous permet d'identifier les propriétés importantes des images pour la réussite des réseaux de restauration. De plus, cette méthode permet de s'affranchir de l'acquisition des données, qui peut s'avérer fastidieuse.Après avoir présenté ce modèle, on montre dans un premier temps que la méthode proposée permet d'obtenir des performances de restauration très proches des méthodes traditionnelles pour des tâches relativement simples. Après quelques adaptations du modèle, l'apprentissage synthétique permet aussi de s'attaquer à des problèmes concrets difficiles, comme le débruitage d'images RAW. On propose ensuite une étude statistique de distribution des couleurs des images naturelles, permettant d'élaborer un modèle parametrique réaliste d'échantillonnage des couleurs pour notre algorithme de génération. Enfin, on présente une nouvelle fonction de perte perceptuelle basée sur les protocoles d'évaluation des cameras, faisant intervenir les images feuilles mortes. Les entrainement réalisés avec cette fonction montre qu'on peut conjointement optimiser l'évaluation des appareils, tout en conservant des performances identiques sur les images naturellesPhotography has become an important part of our lives. In addition, expectations in terms of image quality are increasing while the size of imaging devices is decreasing. In this context, the improvement of image processing algorithms is essential.In this manuscript, we are particularly interested in image restoration tasks. The goal is to produce a clean image from one or more noisy observations of the same scene. For these problems, deep learning methods have grown dramatically in the last decade, outperforming the state of the art for the vast majority of traditional tests.While these methods produce impressive results, they have a number of drawbacks. First of all, they are difficult to interpret because of their "black box" operation. Moreover, they generalize rather poorly to acquisition or distortion modalities absent from the training database. Finally, they require large databases, which are sometimes difficult to acquire.We propose to attack these different problems by replacing the data acquisition by a simple image generation algorithm, based on the dead leaves model. Although this model is very simple, the generated images have statistical properties close to those of natural images and many invariance properties (scale, translation, rotation, contrast...). Training a restoration network with this kind of image allows us to identify the important properties of the images for the success of the restoration networks. Moreover, this method allows us to get rid of the data acquisition, which can be tedious.After presenting this model, we show that the proposed method allows to obtain restoration performances very close to traditional methods for relatively simple tasks. After some adaptations of the model, synthetic learning also allows us to tackle difficult concrete problems, such as RAW image denoising. We then propose a statistical study of the color distribution of natural images, allowing to elaborate a realistic parametric model of color sampling for our generation algorithm. Finally, we present a new perceptual loss function based on camera evaluation protocols, using the dead leaf images. The training performed with this function shows that we can jointly optimize the evaluation of the cameras, while keeping identical performances on natural images
37
Li, Ming De 1937. "Maximum likelihood restoration of binary objects." Thesis, The University of Arizona, 1987. http://hdl.handle.net/10150/276574.
Full textAbstract:
A new approach, based on maximum likelihood, is developed for binary object image restoration. This considers the image formation process as a stochastic process, with noise as a random variable. The likelihood function is constructed for the cases of Gaussian and Poisson noise. An uphill climb method is used to find the object, defined by its "grain" positions, through maximizing the likelihood function for grain positions. In addition, some a priori information regarding object size and contour of shapes is used. This is summarized as a "neighbouring point" rule. Some examples of computer generated images with different signal-to-noise ratios are used to show the ability of the algorithm. These cases include both Gaussian and Poisson noise. For noiseless and low noise Gaussian cases, a modified uphill climb method is used. The results show that binary objects are fairly well restored for all noise cases considered.
38
Pan, Ruimin Reeves Stanley J. "Efficient image restoration algorithms for near-circulant systems." Auburn, Ala., 2007. http://repo.lib.auburn.edu/Send%2011-10-07/PAN_RUIMIN_55.pdf.
Full text
39
Beattie, Robert Scott. "Side scan sonar image formation, restoration and modelling." Thesis, Robert Gordon University, 1996. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.318551.
Full text
40
Sonogashira, Motoharu. "Variational Bayesian Image Restoration with Transformation Parameter Estimation." Kyoto University, 2018. http://hdl.handle.net/2433/232409.
Full text
41
Reddy, Anandigari Bharath Kumar, and Siba Prasad Tudu. "Image Restoration Techniques." Thesis, 2013. http://ethesis.nitrkl.ac.in/5078/1/109EI0347.pdf.
Full textAbstract:
Image restoration is the process of restoring degraded images which cannot be taken again or the process of obtaining the image again is costlier. We can restore the images by prior knowledge of the noise or the disturbance that causes the degradation in the image. Image restoration is done in two domains: spatial domain and frequency domain. In spatial domain the filtering action for restoring the images is done by directly operating on the pixels of the digital image. In frequency domain the filtering action is done by mapping the spatial domain into the frequency domain by taking fourier transform of the image function. By mapping the image into frequency domain an image can provide an insight for filtering operations. After the filtering, the image is remapped into spatial domain by inverse fourier transform to obtain the restored image. Different noise models were studied. Different filtering techniques in both spatial and frequency domains, were studied and improved algorithms were written and simulated using matlab. Restoration efficiency was checked by taking peak signal to noise ratio(psnr) and mean square error(mse) into considerations.
42
Tsai, Jeng-Shiun, and 蔡政勳. "Image Restoration for Noncausal Image Model." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/49247182298506450873.
Full textAbstract:
碩士國立中山大學
電機工程學系研究所
92
Image generating system is usually considered as a noncausal system. The Kalman filter and the Wiener filter are two important linear filters for signal estimation. They are developed for the causal signal and noncausal signal respectively. However, the Kalman filter can also be applied to the noncausal system by rewriting the signal generating equation. In this thesis, we study the performance of the Wiener filter and the Kalman filter applied to image restoration. Our experiments have demonstrated that the rank of list for error performance is: the full order Winner filter, the Kalman filter, the reduced Kalman filter, the three-order Wiener filter. This performance is consisted with the amount of data used in the linear estimation. On the other hand the list for computation performance is as following: the reduced Kalman filter, the three-order Wiener filter, the Kalman filter, the full order Wiener filter. The efficiency of the reduced Kalman filter can be understood by the computation saving of huge updating procedures. It should be noted that the efficiency of applying the regular Kalman filter in this thesis is achieved by fully employed the special form of system matrix involved. In addition to the above noncausal image model, a causal image model can also be built if the central pixel is assumed to be affected only by the left and the upper pixels. The second model is not natural but is obviously advantageous in computation efficiency compared to the first model. However, the first model is much better than the second model error performance. Therefore, it is suggested that the natural image should be modeled as a noncausal model.
43
HUANG, JIAN-HUA, and 黃建華. "Image restoration using nonstationary image models." Thesis, 1986. http://ndltd.ncl.edu.tw/handle/74758214209148412852.
Full text
44
Chiang, Chi-Yuan, and 江啟遠. "Old Stereoscopic Image Restoration." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/60300921052010919264.
Full textAbstract:
碩士國立臺灣大學
資訊網路與多媒體研究所
99
Image restoration has been a major subject of computer vision. Repairing damaged photos requires several computer vision techniques such as denoising, detection, relighting, color-transfer, and inpainting. In this thesis, we proposed an automatic system to repair noise and light inconsistency problems. Our approach combined modern image restoration techniques with additional information gathered from stereoscopic photographs to enhance the effect of image restoration and optimize stereoscopic experience of old stereoscopic photographs.
45
Anwar, Saeed. "Data-Driven Image Restoration." Phd thesis, 2018. http://hdl.handle.net/1885/148622.
Full textAbstract:
Every day many images are taken by digital cameras, and people
are demanding visually accurate and pleasing result. Noise and
blur degrade images captured by modern cameras, and high-level
vision tasks (such as segmentation, recognition, and tracking)
require high-quality images. Therefore, image restoration
specifically, image
deblurring and image denoising is a critical preprocessing step.
A fundamental problem in image deblurring is to recover reliably
distinct spatial frequencies that have been suppressed by the
blur kernel. Existing image deblurring techniques often rely on
generic image priors that only help recover part of the frequency
spectrum, such as the frequencies near the high-end. To this end,
we pose the following specific questions: (i) Does class-specific
information offer an advantage over existing generic priors for
image quality restoration? (ii) If a class-specific prior exists,
how should it be encoded into a deblurring framework to recover
attenuated image frequencies? Throughout this work, we devise a
class-specific prior based on the band-pass filter responses and
incorporate it into a deblurring strategy. Specifically, we show
that the subspace of band-pass filtered images and their
intensity distributions serve as useful priors for recovering
image frequencies.
Next, we present a novel image denoising algorithm that uses
external, category specific image database. In contrast to
existing noisy image restoration algorithms, our method selects
clean image “support patches” similar to the noisy patch from
an external database. We employ a content adaptive distribution
model for each patch where we derive the parameters of the
distribution from the support patches. Our objective function
composed of a Gaussian fidelity term that imposes category
specific information, and a low-rank term that encourages the
similarity between the noisy and the support patches in a robust
manner.
Finally, we propose to learn a fully-convolutional network model
that consists of a Chain of Identity Mapping Modules (CIMM) for
image denoising. The CIMM structure possesses two distinctive
features that are important for the noise removal task. Firstly,
each residual unit employs identity mappings as the skip
connections and receives pre-activated input to preserve the
gradient magnitude propagated in both the forward and backward
directions. Secondly, by utilizing dilated kernels for the
convolution layers in the residual branch, each neuron in the
last convolution layer of each module can observe the full
receptive field of the first layer.
46
Lee, Cheng-Han, and 李承翰. "Super-Resolution Image Restoration from Image Sequence." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/49332385492055610977.
Full textAbstract:
碩士國立臺灣大學
資訊工程學研究所
102
The field of super-resolution has a wide area of applications. In order to display relatively low-quality content on high-resolution displays, the need for super resolution algorithms has become an urgent market priority. A method of super-resolution based on project-onto-convex-sets (POCS) is proposed in this thesis. In the super-resolution process, a set of low-quality images is given, and a single improved-resolution image is desired. We adopt frequency-domain method to estimate motion and enhance the result of high-resolution image by logarithmic image processing model.
47
Kondermann, Daniel Chen Yunqiang. "Multiple image restoration and enhancement /." 2006. http://www.gbv.de/dms/ilmenau/abs/512074879konde.txt.
Full text
48
Huang, Chun-Jen, and 黃俊仁. "Restoration of corrupted Ultrasound Image." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/66143617100906703302.
Full textAbstract:
碩士國立海洋大學
電機工程學系
92
This thesis presents a method to enhance ultrasonic images that are commonly plagued with a special type of acoustic noise called speckles. To reduce the noise effect, filters such as the adaptive weighted median filter, adaptive speckle suppression filter, and two-dimensional Weighted Savitzky-Golay filter (2-D SGF) have been studied. We have found that 2-D SGF has better image restoration quality. Based on the least squares fitting of a polynomial function to image intensities, the 2-D SGF can preserve edges while performing noise reduction. However, the 2-D SGF requires a relatively large fixed mask, which inevitably incurs unnecessary computation cost. In this thesis, a variable-mask approach is presented to improve the computation performance of the 2-D SGF. Advantage of using variable mask size is that computation time for filtering can be greatly saved at the expense of relatively little computations needed for detecting bumpy regions and estimating the corresponding mask size. In addition, the useful properties from characterizing the 2-D SGF are utilized to implement a SGF using higher-order neural networks.
49
"Deep Learning for Image Restoration." 2016. http://repository.lib.cuhk.edu.hk/en/item/cuhk-1292595.
Full text
50
Lu, Zheng Jie, and 盧政傑. "Image restoration using neural networks." Thesis, 1994. http://ndltd.ncl.edu.tw/handle/43931165300008047823.
Full text