The three main tools in the single image restoration theory are the maximum likelihood (ML) estimator, the maximum a posteriori probability (MAP) estimator, and the set theoretic approach using projection onto convex sets (POCS). This paper utilizes the above known tools to propose a unified methodology toward the more complicated problem of superresolution restoration. In the superresolution restoration problem, an improved resolution image is restored from several geometrically warped, blurred, noisy and downsampled measured images. The superresolution restoration problem is modeled and analyzed from the ML, the MAP, and POCS points of view, yielding a generalization of the known superresolution restoration methods. The proposed restoration approach is general but assumes explicit knowledge of the linear space- and time-variant blur, the (additive Gaussian) noise, the different measured resolutions, and the (smooth) motion characteristics. A hybrid method combining the simplicity of the ML and the incorporation of nonellipsoid constraints is presented, giving improved restoration performance, compared with the ML and the POCS approaches. The hybrid method is shown to converge to the unique optimal solution of a new definition of the optimization problem. Superresolution restoration from motionless measurements is also discussed. Simulations demonstrate the power of the proposed methodology.

/pdf/restoration-of-a-single-superresolution-image-from-several-11jwkvrm8u.pdf

Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images

This paper presents a new method based on adaptive filtering theory for superresolution restoration of continuous image sequences. The proposed methodology suggests least squares (LS) estimators which adapt in time, based on adaptive filters, least mean squares (LMS) or recursive least squares (RLS). The adaptation enables the treatment of linear space and time-variant blurring and arbitrary motion, both of them assumed known. The proposed new approach is shown to be of relatively low computational requirements. Simulations demonstrating the superresolution restoration algorithms are presented.

/pdf/superresolution-restoration-of-an-image-sequence-adaptive-1cqntb2v9m.pdf

Superresolution restoration of an image sequence: adaptive filtering approach

Recent years have seen growing interest in the problem of super-resolution restoration of video sequences. Whereas in the traditional single image restoration problem only a single input image is available for processing, the task of reconstructing super-resolution images from multiple undersampled and degraded images can take advantage of the additional spatiotemporal data available in the image sequence. In particular, camera and scene motion lead to frames in the source video sequence containing similar, but not identical information. The additional information available in these frames make possible reconstruction of visually superior frames at higher resolution than that of the original data. In this paper we review the current state of the art and identify promising directions for future research. The authors are with the Laboratory for Image and Signal Analysis (LISA), University of Notre Dame, Notre Dame, IN 46556. E-mail: rls@nd.edu .

/pdf/spatial-resolution-enhancement-of-low-resolution-image-4ihq9t5iqt.pdf

Spatial Resolution Enhancement of Low-Resolution Image Sequences A Comprehensive Review with Directions for Future Research

Stimulating rays produced by a line light source are linearly irradiated onto an area of a stimulable phosphor sheet, on which a radiation image has been stored, the stimulating rays causing the sheet to emit light in proportion to an amount of energy stored thereon during its exposure to radiation. Light emitted from the exposed linear area of the sheet is received with a line sensor comprising photoelectric conversion devices arrayed along each of a length direction of the linear area of the stimulable phosphor sheet and a direction normal to the length direction. The sheet is moved with respect to the line light source and the line sensor and in a direction different from the length direction of the linear area of the sheet. Operation processing is performed on outputs of the photoelectric conversion devices, which outputs have been obtained at respective positions of movement and correspond to an identical site on the sheet.

Radiation image read-out method and apparatus

In this paper a solution is provided to the problem of obtaining a high resolution image from several low resolution images that have been subsampled and displaced by different amounts of sub-pixel shifts. In its most general form, this problem can be broken up into three sub-problems: registration, restoration, and interpolation. Previous work has either solved all three sub-problems independently, or more recently, solved either the first two steps (registration and restoration) or the last two steps together. However, none of the existing methods solve all three sub-problems simultaneously. This paper poses the low resolution to high resolution problem as a maximum likelihood (ML) problem which is solved by the expectation-maximization (EM) algorithm. By exploiting the structure of the matrices involved, the problem ran be solved in the discrete frequency domain. The ML problem is then the estimation of the sub-pixel shifts, the noise variances of each image, the power spectra of the high resolution image, and the high resolution image itself. Experimental results are shown which demonstrate the effectiveness of this approach.

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

A locally-adaptive method for obtaining sub-pixel registration of mask and opacified digital X-ray images includes the steps of match point generation, locally-adaptive image-to-image warp generation, and log subtraction, for generating a DSA image. Specifically, in match point generation, a set of two-dimensional points in the mask image and their corresponding points in the opacified image are derived. After match point generation, locally-adaptive image-to-image warp generation is performed using the image-to-image match points; that is, a transformation function is generated that maps the matched points in the mask image to their corresponding points in the opacified image. The generated transformation is then applied to the mask image data and the logarithm of the pixel value in the transformed (i.e., warped) mask image is subtracted from the logarithm of the corresponding pixel value in the opacified image.

Methods and apparatus for digital subtraction angiography

A radiation imaging system includes an image detector assembly coupled to an image processor. The image detector assembly includes a photosensor pixel array which is adapted to be disposed in a plurality of sequential imaging positions pursuant to a predetermined imaging cycle. The image processor is adapted to store the respective image data signals from each photosensor pixel in the array from each respective imaging position in a given imaging cycle in an unfiltered interleaved data set, and then apply the unfiltered data to a deblurring filter to generate a non-aliased high resolution data set appropriate for driving a display and analysis module.

High resolution radiation imaging system

In this paper, the problem of restoration of multiple misregistered, blurred and noisy images is considered. Motivation for this problem comes from the larger problem of construction of a high resolution restored image from the observed set of low-resolution images. The global approach is to decompose the problem into a sequence of sub-problems, namely registration, restoration and interpolation. The restoration problem is addressed here. For simultaneous restoration of the observed images, a parametric stochastic model for the image correlations is derived and used to obtain a linear minimum mean square error (LMMSE) estimate of the original image at each sensor. The resulting restored low-resolution images are integrated to construct a single restored high resolution image. Simulation results are presented to illustrate the performance of the proposed restoration scheme.

Stochastic model-based approach for simultaneous restoration of multiple misregistered images

A method for fluoroscopically observing a living creature with reduced x-ray dosage is usable with a video monitor for displaying frames of image samples received during respective ones of frame scan intervals that regularly and successively occur at a display frame rate sufficiently high that brightness flicker is acceptably low to a human observer; an x-ray source that can be gated on for intervals not as long as a display frame; a fluoroscopic camera, including a raster-scanning video camera and a x-ray-to-light conversion apparatus with persistence not appreciably longer than a display frame; and frame filling apparatus for grabbing frames of image samples and interpolating between successively grabbed frames of image samples, when necessary, to generate frames of image samples at said display frame rate During ones of said successive frame scan intervals, x-ray radiation is directed from the gated x-ray source through the living creature; the x-ray shadow of at least a portion of the living creature is viewed with the fluoroscopic camera; and the frame filling apparatus grabs frames of image samples from the raster-scanning video camera to be stored in frame stores and used as bases from which to compute fill frames of image samples The frame filling apparatus continuously supplies to the video monitor the frames of image samples it generates at display frame rate

Chukka Srinivas

Papers

Methods and apparatus for digital subtraction angiography

High resolution radiation imaging system

Stochastic model-based approach for simultaneous restoration of multiple misregistered images

Fluoroscopic method with reduced x-ray dosage

Unsupervised noise removal algorithms for three-dimensional confocal fluorescence microscopy