The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, we propose a novel structured compressive sensing algorithm based on non-parametric Bayesian framework for the reconstruction of sparse entries with a continuous structure. A paired spike-and-slab prior is first employed to impose signal sparsity. A logistic Gaussian kernel model, which involves the logistic model and location-dependent Gaussian kernel, is then proposed to encourage...
Super-resolving a natural image is an ill-posed problem. The classical approach is based on the registration and subsequent interpolation of a given set of low-resolution images. However, achieving satisfactory results typically requires the combination of a large number of them. Such an approach would be impractical over heterogeneous rate-constrained wireless networks due to the associated communication...
A new approach to perform the acquisition and the reconstruction of spatially super-resolved hyperspectral images is presented. The proposed hyperspectral sensing strategy is based on acquiring several low-resolved grayscale images following a specific acquisition scheme which takes profit from different spectral dependent blurring kernels. The proposed model describes how output grayscale pixels...
Many statistical learning tasks deal with data which are presented in high-dimensional spaces, and the 'curse of dimensionality' phenomenon is often an obstacle to the use of many methods for solving these tasks. To avoid this phenomenon, various dimensionality reduction algorithms are used as the first key step in solving these tasks. The algorithms transform original high-dimensional data into lower...
Object tracking is an important task within the field of computer vision. Tracking accuracy depends mainly on finding good discriminative features to estimate the target location. In this paper, we introduce online feature learning in tracking and propose to learn good features to track generic objects using online convolutional neural networks (OCNN). OCNN has two feature mapping layers that are...
Frequency domain Normalized Convolution (NC) process is widely performed on images to retrieve and extract valuable information in noisy and distorted environment. Genetic Normalized Convolution (GNC) is carried out for features extraction in an image or features reconstructions in a distorted image. In this paper a hybrid approach is adopted where robust algorithm of convolution based on Normalized...
In this paper, we present a new method to reconstruct a high-resolution (HR) face image from a low-resolution (LR) observation. Inspired by position-patch based face hallucination approach, we design position-based dictionaries to code image patches, and recovery HR patch using the coding coefficients as reconstruction weights. In order to capture nonlinear similarity of face features, we implicitly...
Color image reconstruction provides a measure of the feature representation capability of the moment functions. In this work, we present the quaternion Fourier-Legendre moments in polar pixels, which are computationally faster and have a high-precision compared with other methods. In addition, to improve the performance of the array of polar pixels, we use an inherent property of the Legendre polynomials...
In this paper, an image-based and spatially variant resolution model (i.e. PSF model) was derived from reconstructions of a series of point sources, and used to recover the resolution loss in PET imaging. Since this PSF model was not only system dependent, but also reconstruction algorithm dependent, two reconstruction projectors were compared: one was the single-ray-tracing (1-ray) and another was...
A variety of approaches have been proposed to reduce the variance in reconstructed PET images. In this work, we assess the effect of different combinations of variance reduction techniques on the quality of reconstructed images. These methods include MLEM with early termination, MLEM with post-smoothing, MAPEM and MLEM with inclusion of a convolution matrix prior to the system matrix. Different combinations...
The detector response due to inter-crystal scatter, penetration and non-colinearity in PET can be modeled using blurring point spread functions (PSFs) in sinogram or image space. Incorporating PSFs into image reconstruction is expected to provide better image quality because of accurate system modeling. It has been widely observed that PSF-based reconstruction produces better spatial resolution with...
The implementation and improvement of the spatial resolution of the reconstructions of a motion correction method for unconstrained and unanesthetized rats is presented for a small bore (16 cm diameter) Siemens Inveon microPET scanner. The motion scans are corrected using the LOR rebinning technique. After correction, the images still suffer from loss of spatial resolution compared to motion-free...
Though a variety of face recognition techniques have been proposed in the literature, only a few of them considered open set recognition problems, which involves the rejection of unregistered subjects in addition to identifying persons registered in the database. Transductive confidence machine (TCM) is a novel strategy for classification associated with valid confidence, with recognition reliability...
Multi-frame super-resolution brings out much potential to reconstruct real high-resolution video sequences. This potential is achieved based on its capacity to combine missing information from different input low-resolution frames. Although there have been many studies in recent decades, super-resolution problems for real-world video processing still have many challenges. This is dues to two problems...
This paper suggests a thin cloud removing approach of remote sensing image based on robust kernel regression. Due to the influence of atmosphere condition, cloud cover is one of the most disturbance factors in remote sensing image. So cloud removal is a very important step for improving the quality of the image before making analysis. Because thin cloud is the low frequency component in remote sensing...
As a rule, the performance of almost digital image processing (DIP) algorithms and these applications directly depends on the spatial resolution of observed input images. Unfortunately, from the current image sensor technology, it is hard to take sufficient high spatial resolution images from commercial devices therefore the fantastic research attempts and, consequently, simple digital image resolution...
Compressed sensing (CS) is a promising approach to accelerate dynamic magnetic resonance imaging (MRI). Most existing CS methods employ linear sparsifying transforms. The recent developments in non-linear or kernel-based sparse representations have been shown to outperform the linear transforms. In this paper, we present an iterative non-linear CS dynamic MRI reconstruction framework that uses the...
Motion blur is usually generated when people captured a picture in the daily life. This kind of blur is often non-liner motion and may cause the blurred contents seriously in this image. Hence, how to remove the blurred image into a clear image becomes a very important scheme. In this paper, the primary aim is to propose an efficient blurred image restoration method based on fast blur-kernel estimation,...
Automatic reconstruction of 3D models is attracting increasing attention in the multimedia community. Scene recovery from video sequences requires a selection of representative video frames. Most prior work adopted content-based techniques to automate key frame extraction. However, these methods take no frame geo-information into consideration and are still compute-intensive. Here we propose a new...
In this paper, a super resolution (SR) and an enhancement algorithm are put forward. The SR method is proposed by the assumption of local self example model with nonlocal constraints, and the detail enhancement method is raised by analyzing 2-order holomorphic complete differential kernel of a single image. For the inappropriate frequency component in the SR process, we design a nonlocal constraint...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.