Showing papers on "Image scaling published in 2001"

Curve evolution implementation of the Mumford-Shah functional for image segmentation, denoising, interpolation, and magnification

Abstract: We first address the problem of simultaneous image segmentation and smoothing by approaching the Mumford-Shah (1989) paradigm from a curve evolution perspective. In particular, we let a set of deformable contours define the boundaries between regions in an image where we model the data via piecewise smooth functions and employ a gradient flow to evolve these contours. Each gradient step involves solving an optimal estimation problem for the data within each region, connecting curve evolution and the Mumford-Shah functional with the theory of boundary-value stochastic processes. The resulting active contour model offers a tractable implementation of the original Mumford-Shah model (i.e., without resorting to elliptic approximations which have traditionally been favored for greater ease in implementation) to simultaneously segment and smoothly reconstruct the data within a given image in a coupled manner. Various implementations of this algorithm are introduced to increase its speed of convergence. We also outline a hierarchical implementation of this algorithm to handle important image features such as triple points and other multiple junctions. Next, by generalizing the data fidelity term of the original Mumford-Shah functional to incorporate a spatially varying penalty, we extend our method to problems in which data quality varies across the image and to images in which sets of pixel measurements are missing. This more general model leads us to a novel PDE-based approach for simultaneous image magnification, segmentation, and smoothing, thereby extending the traditional applications of the Mumford-Shah functional which only considers simultaneous segmentation and smoothing.

Image magnification using level-set reconstruction

Abstract: Image magnification is a common problem in imaging applications, requiring interpolation to "read between the pixels". Although many magnification/interpolation algorithms have been proposed in the literature, all methods must suffer to some degree the effects of imperfect reconstruction: false high-frequency content introduced by the underlying original sampling. Most often, these effects manifest themselves as jagged contours in the image. The paper presents a method for constrained smoothing of such artifacts that attempts to produce smooth reconstructions of the image's level curves while still maintaining image fidelity. This is similar to other iterative reconstruction algorithms and to Bayesian restoration techniques, but instead of assuming a smoothness prior for the underlying intensity function it assumes smoothness of the level curves. Results show that this technique can produce images whose error properties are equivalent to the initial approximation (interpolation) used, while their contour smoothness is both visually and quantitatively improved.

Optimal image scaling using pixel classification

Abstract: We introduce a new approach to optimal image scaling called resolution synthesis (RS). In RS, the pixel being interpolated is first classified in the context of a window of neighboring pixels; and then the corresponding high-resolution pixels are obtained by filtering with coefficients that depend upon the classification. RS is based on a stochastic model explicitly reflecting the fact that pixels falls into different classes such as edges of different orientation and smooth textures. We present a simple derivation to show that RS generates the minimum mean-squared error (MMSE) estimate of the high-resolution image, given the low-resolution image. The parameters that specify the stochastic model must be estimated beforehand in a training procedure that we have formulated as an instance of the well-known expectation-maximization (EM) algorithm. We demonstrate that the model parameters generated during the training may be used to obtain superior results even for input images that were not used during the training.

Edge Direction Preserving Image Zooming: A Mathematical and Numerical Analysis

Abstract: We focus in this paper on some reconstruction/restoration methods whose aim is to improve the resolution of digital images. The main point here is to study the ability of such methods to preserve one-dimensional (1D) structures. Indeed, such structures are important since they are often carried by the image "edges." First we focus on linear methods, give a general framework to design them, and show that the preservation of 1D structures pleads in favor of the cancellation of the periodization of the image spectrum. More precisely, we show that preserving 1D structures implies the linear methods to be written as a convolution of the "sinc interpolation." As a consequence, we cannot cope linearly with Gibbs effects, sharpness of the results, and the preservation of the 1D structure. Second, we study variational nonlinear methods and, in particular, the one based on total variation. We show that this latter permits us to avoid these shortcomings. We also prove the existence and consistency of an approximate solution to this variational problem. At last, this theoretical study is highlighted by experiments, both on synthetic and natural images, which show the effects of the described methods on images as well as on their spectrum.

Image interpolation using wavelet based hidden Markov trees

Abstract: Hidden Markov trees in the wavelet domain are capable of accurately modeling the statistical behavior of real world signals by exploiting relationships between coefficients in different scales. The model is used to interpolate images by predicting coefficients at finer scales. Various optimizations and post-processing steps are also investigated to determine their effect on the performance of the interpolation. The interpolation algorithm was found to produce noticeably sharper images with PSNR values which outperform many other interpolation techniques on a variety of images.

A new edge-directed image expansion scheme

Abstract: A new interpolation-based scheme for image expansion is introduced. In the proposed method, the fidelity and sharpness of edges in the expanded image is emphasized. This is achieved by applying edge-directed interpolation and edge sharpening operations. Compared with other methods, our method provides natural and sharp expanded images with light computational cost.

An image interpolator with image improvement for LCD controller

Abstract: An image interpolator with image improvement is proposed and implemented for a FPD (flat panel display) in this paper. We propose an adaptive pseudomedian filter for the interpolation of images and a real-time contrast controller for image improvement. The proposed interpolation algorithm is based on a pseudomedian filter and uses transposed sub-windows according to the correlation of pixels in the image. The proposed contrast controller uses improved histogram stretching methods and does not require field and frame memory for processing the pixels. The operation of the proposed method has been verified and designed using computer simulation and VHDL. The proposed method achieves better performance than the others from the point of view of the edge and local characteristics.

One-pass super-resolution images

Abstract: A super-resolution image is generated from a pixel image by first performing an initial image interpolation, creating an interpolated low resolution image. The interpolated low resolution image is then partitioned into overlapping low resolution patches. The low resolution patches are then processed in a raster scan order. For each low-resolution patch, a scaled mid band patch is generated. A search vector is constructed from pixels in the scaled mid band input patch, and pixels in an overlap region of adjacent previously predicted high band patches. A nearest index vector to the search vector is located in a training database, the nearest index vector has an associated high band output patch. The high band output patch is then combined with the interpolated low resolution patch to predict pixel values for the corresponding high resolution patch of the super-resolution image.

Two-dimensional non-linear interpolation system based on edge information and two-dimensional mixing interpolation system using the same

Abstract: A 2-dimensional non-linear interpolation system and method based on edge information includes an edge detector, an edge direction modifier, a near-edge coefficient generator, a filter coefficient generator and a non-linear interpolation unit. The edge detector detects edge information among pixels from a video signal applied through an input terminal. The edge direction modifier converts the edge information detected by the edge detector on the basis of a center point among peripheral pixels of an interpolation position and outputs modified edge information. The near-edge coefficient generator converts the coordinates of the interpolation position based on the modified edge information to generate a converted interpolation position, generates edge patterns corresponding to the converted interpolation position, and generates a plurality of 2-dimensional interpolation coefficients in response to predetermined one-dimensional non-linear interpolation filter coefficients. The filter coefficient generator generates the one-dimensional non-linear interpolation filter coefficients in response to the coordinates of the converted interpolation position, the edge patterns and predetermined one-dimensional filter coefficients. The non-linear interpolation unit multiplies data values associated with the peripheral pixels by the plurality of 2-dimensional non-linear interpolation coefficients to perform non-linear interpolation. Accordingly, even when a video image is magnified using non-linear interpolation, the resolution of a text or graphic image can be maintained without distortion of edges and aliasing.

Wavelet domain image interpolation via statistical estimation

Abstract: We propose a new wavelet domain image interpolation scheme based on statistical signal estimation. A linear composite MMSE estimator is constructed to synthesize the detailed wavelet coefficients as well as to minimize the mean squared error for high-resolution signal recovery. Based on a discrete time edge model, we use low-resolution information to characterize local intensity changes and perform resolution enhancement accordingly. A linear MMSE estimator follows to minimize the estimation error. Local image statistics are involved in determining the spatially adaptive optimal estimator. With knowledge of edge behavior and local signal statistics, the composite estimation is able to enhance important edges and to maintain the intensity consistency along edges. Strong improvement in both the visual quality and the PSNRs of the interpolated images has been achieved by the proposed estimation scheme.

Elastic image registration using parametric deformation models

Abstract: The main topic of this thesis is elastic image registration for biomedical applications. We start with an overview and classification of existing registration techniques. We revisit the landmark interpolation which appears in the landmark-based registration techniques and add some generalizations. We develop a general elastic image registration algorithm. It uses a grid of uniform B-splines to describe the deformation. It also uses B-splines for image interpolation. Multiresolution in both image and deformation model spaces yields robustness and speed. First we describe a version of this algorithm targeted at finding unidirectional deformation in EPI magnetic resonance images. Then we present the enhanced and generalized version of this algorithm which is significantly faster and capable of treating multidimensional deformations. We apply this algorithm to the registration of SPECT data and to the motion estimation in ultrasound image sequences. A semi-automatic version of the registration algorithm is capable of accepting expert hints in the form of soft landmark constraints. Much fewer landmarks are needed and the results are far superior compared to pure landmark registration. In the second part of this thesis, we deal with the problem of generalized sampling and variational reconstruction. We explain how to reconstruct an object starting from several measurements using arbitrary linear operators. This comprises the case of traditional as well as generalized sampling. Among all possible reconstructions, we choose the one minimizing an a priori given quadratic variational criterion. We give an overview of the method and present several examples of applications. We also provide the mathematical details of the theory and discuss the choice of the variational criterion to be used.

Image processing apparatus and method, program, and recording medium

Abstract: Image refreshers enhance edge portions of field data stored in field memories, and then store the result in field memories. An intra-field interpolating unit interpolates a moving image pixel using a single piece of field data stored in the field memory, and then stores the moving image pixel in an interpolation memory. An inter-field interpolating unit interpolates a still image pixel using two pieces of field data stored in the field memories, and then stores the still image pixel in an interpolation memory. A still/moving image area determining unit determines whether each pixel of a progressive picture is a still image pixel or a moving image pixel. On the basis of a result of the determination by the still/moving image area determining unit, a selector reads a pixel value in the interpolation memory for a moving image pixel, and reads a pixel value from the interpolation memory for a still image pixel.

Development of a Robust Algorithm for Transformation of a 3D Object Point onto a 2D Image Point for Linear Pushbroom Imagery

Abstract: expressed in a rigorous mathematical form (Kim, 2000). Further A powe$ul and robust algorithm for the Indirect Method, i.e., studies are required for the understanding of the geometry of the transformation of a 30 object point onto a 2D image point linearpushbroomimages. for linear pushbroom imagery, is proposed. This algorithm This paper addresses the problem of rectification of linear solves the transformation iteratively with an initial estimate pushbroom images so that the images can be geometrically ref- of the 20 image point coordinates. However, this algorithm erenced. There can be two ways to solve this problem. The first does not require any sophisticated procedures to determine a one is referred to as the Direct Method, which solves the prob- "good" initial estimate and it always converges to the correct lem by projecting an image point in two-dimensional (zD) solution. This algorithm works using the following procedures: image coordinates onto an object point in three-dimensional first, with an (random) initial estimate of the 20 image point (3D) object coordinates. A technique called Ray-Tracing coordinates, calculate the attitude of the camera platform; (O'Neilland Dowman, 1988) was developed to solve the prob- second, with the given attitude, calculate the position of the lem by the Direct Method. The other one is referred to as the camera platform and the 20 image point; and third, update Indirect Method, which solves the problem the other way the estimate with the calculated 2D image point coordinates around, i.e-9 by projecting a 3D object point onto a zD image and then go back to the first procedure and continue iteration point. It is known that the Indirect Method has many advan- until the estimated and calculated image point coordinates tages overthe Direct Method (Mayrand Hei~ke, 1988). In Par- converge. Results of the experiment show that this algorithm ticular~ it reduce sthe processing time and amount of memory converges very fast even when the initial estimate has a required for image resampling. huge error. However, a robust numerical solution of the Indirect Method for linear pushbroom images has so far not been devel-

Method and apparatus for de-interlacing video images

Abstract: De-interlacing is effected by determining the motion at each missing pixel and, then, interpolating the missing lines to convert an interlaced field to a progressive frame. The interpolation employed for luminance is determined through motion detection. If motion is detected in the image field based interpolation is used and if no motion of the image is detected frame interpolation is used. Specifically, the interpolation is determined by employing a motion metric. The motion metric at a missing pixel is defined by using a prescribed combination of pixel luminance value differences. A spatial median filter is then used to remove objectionable noise from the pixel luminance value differences and to fill in so-called "holes" in the image. Indeed, the spatial median filter can be considered as providing a measure of the overall effect of all pixels that make up the object of the image.

Adaptive, optimal-recovery image interpolation

Abstract: We consider the problem of image interpolation using adaptive optimal recovery. We adaptively estimate the local quadratic signal class of our image pixels. We then use optimal recovery to estimate the missing local samples based on this quadratic signal class. This approach tends to preserve edges, interpolating along edges and not across them.

Interpolation apparatus, and video signal processing apparatus including the same

Abstract: An interpolation apparatus that generates interpolation pixel values necessary for converting input video data of interlace scanning into video data of progressive scanning is provided. A plurality of candidate pixel-pairs each of which is composed of two pixels that are symmetric with respected to a pixel that is going to be interpolated are selected from pixels on adjacent two scan lines within one field of the input video data, and a difference between pixel values of each selected pixel-pair is calculated. A pixel-pair to be used for generating the interpolation pixel value is determined, based on the smallest difference and the second smallest difference of the calculated differences. An interpolation pixel value of the pixel that is going to be interpolated is generated based on pixel values of the determined pixel-pair.

Image interpolation method and apparatus

Abstract: An image processing apparatus detects corresponding pixels in object images picked up from a plurality of viewpoints and interpolates object images which are supposed to be seen from viewpoints other than the plurality of viewpoints on the basis of the detected corresponding pixels.

Image processing method and apparatus, and image display method and apparatus, with variable interpolation spacing

Abstract: An image is processed by detecting pixel-to-pixel variations in brightness level, generating high spatial frequency information related to the variations, setting interpolation points with a spacing that varies according to the high spatial frequency information, and generating new pixels by interpolation at the interpolation points. By increasing the zoom ratio in one part and reducing the zoom in another part of each edge in a continuous manner, this method can mitigate edge degradation when an image is enlarged or reduced, without introducing discontinuities or other image artifacts. It also provides a convenient way to adjust edge sharpness in an image.

Intermediate image interpolation using polyphase weighted median filters

Abstract: A new algorithm for the interpolation of temporal intermediate images using polyphase weighted median filters is proposed in this paper. To achieve a good interpolation quality not only in still but also in moving areas of the image, vector based interpolation techniques have to be used. However, motion estimation on natural image scenes always suffers from errors in the estimated motion vector field. Therefore it is of great importance that the interpolation algorithm possesses a sufficient robustness against vector errors. Depending on the input and output frame repetition rate, different cyclically repeated interpolation phases can be distinguished. The new interpolation algorithm uses dedicated weighted median filters for each interpolation phase (polyphase weighted median filters) which are (due to their shift property) able to achieve a correct positioning of moving edges in the interpolated image, even if the estimated vector differs from the true motion vector up to a certain degree. A new design method for these dedicated error tolerant weighted median filters is presented in the paper. Other aspects like e.g. the preservation of fine image details can also be regarded in the design process. The results of the new algorithm are compared to other existing interpolation algorithms.

Models for image interpolation based on the optical flow

Abstract: Optical flow is the 2D motion that needs to be recovered from a video sequence. In this paper we study variational principles for the generation of interpolating sequences between two images. The basic assumption is that there exists an underlying video sequence that solves the optic flow equation and interpolates the two images. The numerical solution of the interpolation problem is reduced to the solution of a system of coupled partial differential equations. Some numerical simulations are presented.

Edge-preserving image resizing using modified B-splines

Abstract: An edge-preserving method for image resizing (decimation and interpolation) is proposed. The decimation is considered as an orthogonal projection with respect to the chosen interpolation basis. The latter one is formed in a spline-like manner as a linear combination of B-splines of different degrees. This combination is optimized in such a way that the small image details are preserved. Considering the strongest edges as step edges, a segmentation procedure preceding the decimation is proposed. It leads to resized images with clearly outlined borders.

Image interpolation using across-scale pixel correlation

Abstract: A novel method is proposed for image interpolation. It is assumed that the pixel correlation between local regions across scales would remain similar. In addition, this a priori similarity could be extracted from a set of available image data that have the same content but different resolutions. A simple architecture is devised to estimate the correlation efficiently, which is then used to predict the unknown pixel values in a high-resolution image. Evaluation shows a promising performance of the proposed algorithm.

Medium whereon image data interpolation program has been recorded, image data interpolation method, and image data interpolation apparatus

Abstract: Generally, it is not easy to judge automatically and correctly whether the attribute of an image is the one belonging to logos and illustrations or the one belonging to natural pictures. Due to incorrect judgment, sometimes, unsuitable interpolation execution has occurred. Pixels interpolated by the first interpolation processing and pixels interpolated by the second interpolation processing are blended, based on a predetermined evaluation function, and placed on a source image. Because the evaluation function depends on the attribute of the image, a blending ratio between the pixels interpolated separately in two modes of interpolation can be determined, according to the image attribute, with high percentages assigned to more suitable interpolation processing. The merit of each mode of interpolation processing becomes more noticeable, whereas the demerit of each becomes mild. Consequently, the invention can prevent an error in selecting an interpolation method, based on the appraised attribute of the image for which interpolation is executed.

An architecture and implementation of image scaling conversion

Abstract: Cubic and bisigmoidal interpolation methods are used for image scaling to improve the video quality. We have developed an efficient architecture with the hardware complexity reduced. The system is implemented and verified by the FPGA-based evaluation board.