Special Issue "Machine Learning and Compressed Sensing in Image Reconstruction"

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (30 September 2019).

Special Issue Editors

Prof. Dr. Markus Haltmeier
Website
Guest Editor
Department of Mathematics, University of Innsbruck, Innsbruck, Austria
Interests: inverse problems; computed tomography; computational imaging; applied mathematics; machine learning
Prof. Dr. Linh. V. Nguyen
Website
Guest Editor
Department of Mathematics, University of Idaho, USA
Interests: inverse problems; computed tomography; computational imaging; applied mathematics; machine learning

Special Issue Information

Dear Colleagues,

The development of fast and accurate reconstruction algorithms plays a central role in modern imaging systems. Examples include x-ray tomography, ultrasound imaging, photoacoustic imaging, super-resolution imaging, and magnetic resonance imaging. Compressed sensing and machine learning are successful tools for various imaging applications. Compressed sensing techniques allow to significantly reduce the amount of data to be acquired and thereby accelerates data acquisition, reduces motion artefacts, and lowers radiation exposure. In compressed sensing, iterative algorithms based on prior information have been applied for image reconstruction. Such algorithms can be time-consuming as the forward and adjoint problems have to be computed repeatedly. Recently, a new class of algorithms based on machine learning, especially deep learning, for compressed sensing and other image reconstruction tasks appeared. With deep learning, image reconstruction can be performed efficiently using artificial neural networks, whose weights are based on training data. While still in their infancy, these techniques already show astonishing performance. This Special Issue focuses on the latest research and development of compressed sensing and machine learning for image reconstruction. Papers on the design of new reconstruction algorithms and new compressed sensing and machine learning applications are welcome. In addition, contributions on the theoretical analysis and understanding of compressed sensing and machine learning in image reconstruction are welcome.

Prof. Dr. Markus Haltmeier
Prof. Dr. Linh. V. Nguyen
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • computed tomography
  • compressed sensing
  • sampling theory
  • iterative algorithms
  • photoacoustic tomography
  • magnetic resonance imaging
  • parameter identification
  • regularization methods
  • sparse recovery
  • neural networks
  • deep learning image reconstruction
  • data consistency

Published Papers (9 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Open AccessArticle
Phase Extraction from Single Interferogram Including Closed-Fringe Using Deep Learning
Appl. Sci. 2019, 9(17), 3529; https://doi.org/10.3390/app9173529 - 28 Aug 2019
Cited by 2
Abstract
In an optical measurement system using an interferometer, a phase extracting technique from interferogram is the key issue. When the object is varying in time, the Fourier-transform method is commonly used since this method can extract a phase image from a single interferogram. [...] Read more.
In an optical measurement system using an interferometer, a phase extracting technique from interferogram is the key issue. When the object is varying in time, the Fourier-transform method is commonly used since this method can extract a phase image from a single interferogram. However, there is a limitation, that an interferogram including closed-fringes cannot be applied. The closed-fringes appear when intervals of the background fringes are long. In some experimental setups, which need to change the alignments of optical components such as a 3-D optical tomographic system, the interval of the fringes cannot be controlled. To extract the phase from the interferogram including the closed-fringes we propose the use of deep learning. A large amount of the pairs of the interferograms and phase-shift images are prepared, and the trained network, the input for which is an interferogram and the output a corresponding phase-shift image, is obtained using supervised learning. From comparisons of the extracted phase, we can demonstrate that the accuracy of the trained network is superior to that of the Fourier-transform method. Furthermore, the trained network can be applicable to the interferogram including the closed-fringes, which is impossible with the Fourier transform method. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Adaptive Bat Algorithm Optimization Strategy for Observation Matrix
Appl. Sci. 2019, 9(15), 3008; https://doi.org/10.3390/app9153008 - 26 Jul 2019
Cited by 1
Abstract
Bat algorithm, as an optimization strategy of the observation matrix, has been widely used. Observation matrix has a direct impact on the reconstructed signal accuracy as a projection transformation matrix, and it has been widely used in various algorithms. However, for the traditional [...] Read more.
Bat algorithm, as an optimization strategy of the observation matrix, has been widely used. Observation matrix has a direct impact on the reconstructed signal accuracy as a projection transformation matrix, and it has been widely used in various algorithms. However, for the traditional experimental process, randomly generated observation matrices often result in a larger reconstruction error and unstable reconstruction results. Therefore, it is a challenge to retain more feature information of the original signal and reduce reconstruction error. To obtain a more accurate reconstruction signal and less memory space, it is important to select an effective compression and reconstruction strategy. To solve this problem, an adaptive bat algorithm is proposed to optimize the observation matrix in this paper. For the adaptive bat algorithm, we design a dynamic adjustment strategy of the optimal radius to improve its global convergence ability. The results of our simulation experiments verify that, compared with other algorithms, it can effectively reduce the reconstruction error and has stronger robustness. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
On-the-Fly Machine Learning for Improving Image Resolution in Tomography
Appl. Sci. 2019, 9(12), 2445; https://doi.org/10.3390/app9122445 - 14 Jun 2019
Cited by 2
Abstract
In tomography, the resolution of the reconstructed 3D volume is inherently limited by the pixel resolution of the detector and optical phenomena. Machine learning has demonstrated powerful capabilities for super-resolution in several imaging applications. Such methods typically rely on the availability of high-quality [...] Read more.
In tomography, the resolution of the reconstructed 3D volume is inherently limited by the pixel resolution of the detector and optical phenomena. Machine learning has demonstrated powerful capabilities for super-resolution in several imaging applications. Such methods typically rely on the availability of high-quality training data for a series of similar objects. In many applications of tomography, existing machine learning methods cannot be used because scanning such a series of similar objects is either impossible or infeasible. In this paper, we propose a novel technique for improving the resolution of tomographic volumes that is based on the assumption that the local structure is similar throughout the object. Therefore, our approach does not require a training set of similar objects. The technique combines a specially designed scanning procedure with a machine learning method for super-resolution imaging. We demonstrate the effectiveness of our approach using both simulated and experimental data. The results show that the proposed method is able to significantly improve resolution of tomographic reconstructions. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Artificial Bee Colony Programming Descriptor for Multi-Class Texture Classification
Appl. Sci. 2019, 9(9), 1930; https://doi.org/10.3390/app9091930 - 10 May 2019
Cited by 3
Abstract
Texture classification is one of the machine learning methods that attempts to classify textures by evaluating samples. Extracting related features from the samples is necessary to successfully classify textures. It is a very difficult task to extract successful models in the texture classification [...] Read more.
Texture classification is one of the machine learning methods that attempts to classify textures by evaluating samples. Extracting related features from the samples is necessary to successfully classify textures. It is a very difficult task to extract successful models in the texture classification problem. The Artificial Bee Colony (ABC) algorithm is one of the most popular evolutionary algorithms inspired by the search behavior of honey bees. Artificial Bee Colony Programming (ABCP) is a recently introduced high-level automatic programming method for a Symbolic Regression (SR) problem based on the ABC algorithm. ABCP has applied in several fields to solve different problems up to date. In this paper, the Artificial Bee Colony Programming Descriptor (ABCP-Descriptor) is proposed to classify multi-class textures. The models of the descriptor are obtained with windows sliding on the textures. Each sample in the texture dataset is defined instance. For the classification of each texture, only two random selected instances are used in the training phase. The performance of the descriptor is compared standard Local Binary Pattern (LBP) and Genetic Programming-Descriptor (GP-descriptor) in two commonly used texture datasets. When the results are evaluated, the proposed method is found to be a useful method in image processing and has good performance compared to LBP and GP-descriptor. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Image Recovery with Data Missing in the Presence of Salt-and-Pepper Noise
Appl. Sci. 2019, 9(7), 1426; https://doi.org/10.3390/app9071426 - 04 Apr 2019
Abstract
In this paper, an image recovery problem under the case of salt-and-pepper noise and data missing that degrade image quality is addressed if they are not effectively handled, where the salt-and-pepper noise as the impulsive noise is remodeled as a sparse signal due [...] Read more.
In this paper, an image recovery problem under the case of salt-and-pepper noise and data missing that degrade image quality is addressed if they are not effectively handled, where the salt-and-pepper noise as the impulsive noise is remodeled as a sparse signal due to its impulsiveness and the data missing pattern, denoted by a sparse vector, contains only zeros and ones to formulate the data missing. In particular, the salt-and-pepper noise and data missing are reformatted by their sparsity, respectively. The wavelet and framelet domains are explored to sparsely represent the image in order to accurately reconstruct the clean image. From the reformulations conducted and to recover the image, under one optimization framework, a joint estimation is developed to perform the image recovery, the salt-and-pepper noise suppression, and the missing patter estimation. To solve the optimization problem, two efficient solvers are developed to obtain the joint estimation solution, and they are based on the alternating direction method of multipliers (ADMM) and accelerated proximal gradient (APG). Finally, numerical studies verify that the joint estimation algorithm outperforms the state-of-the-art approaches in terms of both objective and subjective evaluation standards. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Tensor Robust Principal Component Analysis via Non-Convex Low Rank Approximation
Appl. Sci. 2019, 9(7), 1411; https://doi.org/10.3390/app9071411 - 03 Apr 2019
Cited by 2
Abstract
Tensor Robust Principal Component Analysis (TRPCA) plays a critical role in handling high multi-dimensional data sets, aiming to recover the low-rank and sparse components both accurately and efficiently. In this paper, different from current approach, we developed a new t-Gamma tensor quasi-norm as [...] Read more.
Tensor Robust Principal Component Analysis (TRPCA) plays a critical role in handling high multi-dimensional data sets, aiming to recover the low-rank and sparse components both accurately and efficiently. In this paper, different from current approach, we developed a new t-Gamma tensor quasi-norm as a non-convex regularization to approximate the low-rank component. Compared to various convex regularization, this new configuration not only can better capture the tensor rank but also provides a simplified approach. An optimization process is conducted via tensor singular decomposition and an efficient augmented Lagrange multiplier algorithm is established. Extensive experimental results demonstrate that our new approach outperforms current state-of-the-art algorithms in terms of accuracy and efficiency. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Fast Cylinder Shape Matching Using Random Sample Consensus in Large Scale Point Cloud
Appl. Sci. 2019, 9(5), 974; https://doi.org/10.3390/app9050974 - 07 Mar 2019
Cited by 10
Abstract
In this paper, an algorithm is proposed that can perform cylinder type matching faster than the existing method in point clouds that represent space. The existing matching method uses Hough transform and completes the matching through preprocessing such as noise filtering, normal estimation, [...] Read more.
In this paper, an algorithm is proposed that can perform cylinder type matching faster than the existing method in point clouds that represent space. The existing matching method uses Hough transform and completes the matching through preprocessing such as noise filtering, normal estimation, and segmentation. The proposed method completes the matching through the methodology of random sample consensus (RANSAC) and principal component analysis (PCA). Cylindrical pipe estimation is based on two mathematical models that compute the parameters and combine the results to predict spheres and lines. RANSAC fitting computes the center and radius of the sphere, which can be the radius of the cylinder axis and finds straight and curved areas through PCA. This allows fast matching without normal estimation and segmentation. Linear and curved regions are distinguished by a discriminant using eigenvalues. The linear region is the sum of the vectors of linear candidates, and the curved region is matched by a Catmull–Rom spline. The proposed method is expected to improve the work efficiency of the reverse design by matching linear and curved cylinder estimation without vertical/horizontal constraint and segmentation. It is also more than 10 times faster while maintaining the accuracy of the conventional method. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Product Innovation Design Based on Deep Learning and Kansei Engineering
Appl. Sci. 2018, 8(12), 2397; https://doi.org/10.3390/app8122397 - 26 Nov 2018
Cited by 4
Abstract
Creative product design is becoming critical to the success of many enterprises. However, the conventional product innovation process is hindered by two major challenges: the difficulty to capture users’ preferences and the lack of intuitive approaches to visually inspire the designer, which is [...] Read more.
Creative product design is becoming critical to the success of many enterprises. However, the conventional product innovation process is hindered by two major challenges: the difficulty to capture users’ preferences and the lack of intuitive approaches to visually inspire the designer, which is especially true in fashion design and form design of many other types of products. In this paper, we propose to combine Kansei engineering and the deep learning for product innovation (KENPI) framework, which can transfer color, pattern, etc. of a style image in real time to a product’s shape automatically. To capture user preferences, we combine Kansei engineering with back-propagation neural networks to establish a mapping model between product properties and styles. To address the inspiration issue in product innovation, the convolutional neural network-based neural style transfer is adopted to reconstruct and merge color and pattern features of the style image, which are then migrated to the target product. The generated new product image can not only preserve the shape of the target product but also have the features of the style image. The Kansei analysis shows that the semantics of the new product have been enhanced on the basis of the target product, which means that the new product design can better meet the needs of users. Finally, implementation of this proposed method is demonstrated in detail through a case study of female coat design. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Open AccessArticle
Infrared Image Super-Resolution Reconstruction Based on Quaternion Fractional Order Total Variation with Lp Quasinorm
Appl. Sci. 2018, 8(10), 1864; https://doi.org/10.3390/app8101864 - 10 Oct 2018
Cited by 10
Abstract
Owing to the limitations of the imaging principle as well as the properties of imaging systems, infrared images often have some drawbacks, including low resolution, a lack of detail, and indistinct edges. Therefore, it is essential to improve infrared image quality. Considering the [...] Read more.
Owing to the limitations of the imaging principle as well as the properties of imaging systems, infrared images often have some drawbacks, including low resolution, a lack of detail, and indistinct edges. Therefore, it is essential to improve infrared image quality. Considering the information of neighbors, a description of sparse edges, and by avoiding staircase artifacts, a new super-resolution reconstruction (SRR) method is proposed for infrared images, which is based on fractional order total variation (FTV) with quaternion total variation and the L p quasinorm. Our proposed method improves the sparsity exploitation of FTV, and efficiently preserves image structures. Furthermore, we adopt the plug-and-play alternating direction method of multipliers (ADMM) and the fast Fourier transform (FFT) theory for the proposed method to improve the efficiency and robustness of our algorithm; in addition, an accelerated step is adopted. Our experimental results show that the proposed method leads to excellent performances in terms of an objective evaluation and the subjective visual effect. Full article
(This article belongs to the Special Issue Machine Learning and Compressed Sensing in Image Reconstruction)
Show Figures

Figure 1

Back to TopTop