Showing papers on "Image scaling published in 2000"

Effective color interpolation in CCD color filter array using signal correlation

Abstract: We propose an effective color filter array interpolation method for digital still cameras using a simple image model that correlates the R, G, B channels. The main contribution of this paper is that we propose a low complexity interpolation method to improve the image quality. Simulation results verify that the proposed method obtain superior image quality. The G channel of the proposed method outperforms the bilinear method by 6.34 dB, and the R, B channels have a 7.69 dB improvement in average. Furthermore, the complexity of the proposed method is acceptable compared with the existing methods.

Efficient image scaling for scan rate conversion

Abstract: The present invention is a method and apparatus for converting scan rates of image data in a memory (140). A buffer (240) stores a source image data. A scaling filter (210) is coupled to the buffer (240) to scale the source image data.

Interpolation using neural networks for digital still cameras

Abstract: In this paper we present a color interpolation technique based on artificial neural networks for a single-chip CCD (charge-coupled device) camera with a Bayer color filter array (CFA) Single-chip digital cameras use a color filter array and an interpolation method in order to produce high quality color images from sparsely sampled images We have applied 3-layer feedforward neural networks in order to interpolate a missing pixel from surrounding pixels And we compare the proposed method with conventional interpolation methods such as the bilinear interpolation method and cubic spline interpolation method Experiments show that the proposed interpolation algorithm based on neural networks provides a better performance than the conventional interpolation algorithms

A high quality 2 x image interpolator

Abstract: In this paper, we present an innovative interpolator that performs high quality 2/spl times/interpolation on both synthetic and real world images. Its structure, which is based on a rational operator, provides edge-sensitive data interpolation so that sharp- and artifacts-free images are obtained at a reasonable computational cost.

New edge directed interpolation

Abstract: This paper presents a novel edge orientation adaptive interpolation scheme for resolution enhancement of still images. In order to achieve ideal orientation adaptation, we propose to estimate the local covariance characteristics at low resolution but cleverly use them to direct the interpolation at high resolution based on the resolution invariant property of edge orientation. The orientation adaptive property guarantees the interpolation always go along the edge orientation but not across it. Our new interpolation scheme can generate images with dramatically higher visual quality than linear interpolation techniques while keeping the computational complexity still modest.

Robust dense matching using local and global geometric constraints

Abstract: A new robust dense matching algorithm is introduced. The algorithm starts from matching the most textured points, then a match propagation algorithm is developed with the best first strategy to dense matching. Next, the matching map is regularised by using the local geometric constraints encoded by planar affine applications and by using the global geometric constraint encoded by the fundamental matrix. Two most distinctive features are a match propagation strategy developed by analogy to region growing and a successive regularisation by local and global geometric constraints. The algorithm is efficient, robust and can cope with wide disparity. The algorithm is demonstrated on many real image pairs, and applications on image interpolation and a creation of novel views are also presented.

Prediction of image detail

Abstract: In the problem of image interpolation, most of the difficulties arise in areas around edges and sharp changes. Around edges, many interpolation methods tend to smooth and blur image detail. Fortunately, most of the signal information is often carried around edges and areas of sharp changes and can be used to predict these missing details from a sampled image. A method for adding image detail based on the cone of influence, the evolution of the wavelet coefficients across scales, is presented.

System and method of color interpolation

Abstract: A system and method of processing data associated with an array of pixels arranged in two dimensions are disclosed. The data includes a value representative of an intensity of photoexposure in one of a plurality of distinct spectral regions for each pixel location in the array. For at least one pixel location being associated with a first one of the plurality of spectral regions, a direction for interpolation is selected based upon an intensity gradient at the pixel location and an intensity continuity bias representative of a trend in changes in intensity in a neighborhood in the array of pixels about the pixel location. An interpolation among values representative of an intensity of photoexposure at two or more neighboring pixel locations in at least one set of neighboring pixels aligned along the selected direction to provides a value associated with the at least one pixel location which is representative of an intensity in a second one of the plurality of spectral regions distinct from the first spectral region.

Global/local motion-compensated frame interpolation for low-bit-rate video

Abstract: A new motion-compensated frame interpolation scheme for low bitrate video based on the ITU-T H.263/H.263+ standard is investigated in this research. The proposed scheme works solely on the decoded bitstream with a block-based approach to achieve interpolation results. It is composed of two main modules: the background/foreground segmentation module and the hybrid motion compensated frame interpolation module. The background/foreground segmentation module uses a global motion model to estimate the background motion and an iterative background update to refine the segmentation. The hybrid motion compensated frame interpolation module is employed to reconstruct background and foreground, respectively. Global motion compensation and frame interpolation is applied to background blocks where either the 6-parameter affine or the 8-parameter perspective model is used to reduce the computational complexity and implement perspective correction, while local motion compensation and frame interpolation with localized triangular patch mapping is applied to the foreground area. Experiments show that the proposed scheme can achieve higher overall visual quality compared to conventional block-based frame interpolation schemes.

Image interpolation device

Abstract: PROBLEM TO BE SOLVED: To provide an image interpolation device capable of obtaining the high definition one as an image after interpolation. SOLUTION: In the case that strongest correlation is present in a diagonal direction H3, at the time of defining the signs (+, - or 0) of differences A(-1)-A(-2) and B(1)-B(0) between two input pixels A(-1) and B(1) facing each other across an interpolation pixel X in the diagonal direction H3 and the input pixels A(-2) and B(0) adjacent in the left direction of the respective two pixels respectively as (a) and c and defining the signs of the differences A(-1)-A(0) and B(1)-B(2) between the input pixels A(-1) and B(1) and source pixels A(0) and B(2) adjacent in the right direction of the respective two pixels respectively as (b) and (d), when the condition of a=c or b=d is satisfied, the interpolation is performed in the diagonal direction H3 where the strongest correlation is present.

Method and apparatus for interpolating a digital image

Abstract: A method and apparatus is provided for interpolating a digital image in response to a requested degree of sharpness. An adjusting signal representing the requested degree of sharpness will then be generated. The interpolated pixel data are computed based on a two-order interpolation function for two sampling input pixels with an adjustable weight coefficient representing the selected degree of sharpness. The apparatus of the present invention mainly includes: a control interface, a control unit, a vertical interpolation computation module, and a horizontal interpolation module. The vertical interpolation computation module and the horizontal interpolation module are implemented according to an interpolation function derived by the present invention. The control unit comprises a lookup table built according to a scaling function of the present invention. The vertical scaling factor and the horizontal scaling factor required for the interpolation function can be obtained by looking up the lookup table according to the adjusting signal, and the position of the interpolated pixel. Accordingly, the present invention can control the degree of sharpness without having to implement an additional sharp control circuit.

Efficient edge line average interpolation algorithm for deinterlacing

Abstract: In this paper, an efficient interpolation approach is proposed for deinterlacing within a single frame. On the basis of the edge-based line average (ELA) algorithm, two useful measurements are introduced within the analysis window in order to alleviate misleading decisions in determining the direction in which the interpolation is to be made. By efficiently estimating the directional spatial correlations of neighboring pixels, increased interpolation accuracy has been achieved. Additionally, the new method possesses a simple computation structure and is therefore easy to implement. Extensive simulations conducted for different images and video sequences have shown the efficacy of the proposed interpolator with significant improvement over previous ELA based algorithms in terms of both quantitative and perceived image quality.

Polynomial and rational operators for image processing and analysis

Abstract: Publisher Summary This chapter discusses polynomial and rational operators for image processing and analysis. A wide class of nonlinear operators can be devised for image processing applications, based on polynomial and rational functions of the pixels of an image. This chapter shows that this approach can be exploited successfully for image enhancement, image analysis, and image format conversion. The chapter also presents the properties of polynomial-based functions. It deals with the formalisms of polynomial and rational filters and provides an overview of the theoretical properties of operators that are based on the Volterra series and of their extensions to two dimensions. The chapter then discusses the applications of polynomial filters for contrast enhancement, texture segmentation, and edge extraction. Finally, it takes a look at rational operators with a number of applications such as noise smoothing, image interpolation and contrast enhancement.

Image scaling method and apparatus using continuous domain filtering and interpolation method

Abstract: An image scaling method and apparatus converting an input image into an aimed image with suitable resolution for application, in the case that the input image and the aimed image are different in resolution, using continuous domain filtering and interpolation method. The invention utilizes domain filter of regular square type and generates filter coefficient based on area occupancy ratio of filter window according to the area of corresponding pixel extending the applied filter window and then obtains scaled image through the method of computing weighted average value, which is determined through multiplication of the generated filter coefficient and value of each pixel. By using image scaling method of the invention, there is an advantage of much lower cost in hardware implementation of an image scaler in comparison with conventional image scaling methods.

Compression method for real-time systems of remote sensing

Abstract: In this paper the method of image compression is presented. It is designed for data processing in real-time systems of remote sensing. In the midpoint there are compression algorithm based on component transformation with pixel interpolation and algorithm of stabilization of encoded image forming speed, which provide high compression ratio, stable speed of an output data flow and controlled error of image reconstruction.

DEM extraction from stereo SAR satellite imagery

Abstract: Automatic Digital Elevation Model (DEM) extraction from stereo SAR satellite images continues to be a challenge. The main difficulty is in obtaining the largest possible number of correct matches, at the same time minimizing the number of false matches and other artifacts and distortions in the derived DEMs. For accepted matches, the highest possible geometric accuracy is desired. This paper describes a number of enhancements in an operational module for DEM extraction: (1) image resampling to quasi-epipolar geometry, (2) uniform area detection and skipping, (3) matching window warping, (4) automatic screening and editing of matches, and (5) large hole interpolation. As a result of these enhancements, more detailed and more accurate DEMs with fewer blunders are extracted with minimum user interaction. This is illustrated by examples for various terrain and land cover conditions.

Digital still camera system and complementary-color-filtered array interpolation method

Abstract: A complementary-color-filtered array interpolation by first interpolate each color subarray so each pixel has four colors and then adjust each color at a pixel with addition or subtraction of a color imbalance factor for the pixel. For yellow and cyan the adjustment is subtraction but for magenta and green the adjustment is addition.

Wavelet-based reconstruction of irregularly-sampled images: application to stereo imaging

Abstract: We are concerned with the reconstruction of a regularly-sampled image based on irregularly-spaced samples thereof. We propose a new iterative method based on a wavelet representation of the image. For this representation we use a biorthogonal spline wavelet basis implemented on an oversampled grid. We apply the developed algorithm to disparity-compensated stereoscopic image interpolation. Under disparity compensation, the resulting sampling grids are irregular and require the irregular/regular interpolation. We show the experimental results on real-world images and we compare our results with other methods proposed in the literature.

Edge-constrained joint view triangulation for image interpolation

Abstract: Image-based-interpolation creates smooth and photo-realistic views between two view points. The concept of joint view triangulation (JVT) has been proven to be an efficient multi-view representation to handle visibility issue. However, the existing JVT built only on a regular sampling grid, often produces undesirable artifacts for artificial objects. To tackle these problems, a new edge-constrained joint view triangulation is developed in this paper to integrate contour points and artificial rectilinear objects as triangulation constraints. Also a super-sampling technique is introduced to refine visible boundaries. The new algorithm is successfully demonstrated on many real image pairs.

Image sequence processing for videowall visualization

Abstract: A new processing scheme for large high-resolution displays such as Videowalls is proposed in this paper. The scheme consists in a deinterlacing, an interpolation and an optional enhancement algorithm; its hardware implementation requires a low computational cost. The deinterlacing algorithm is motion- adaptive. A simple hierarchical three-level motion detector provides indications of static, slow and fast motion to activate a temporal FIR filter, a three-tap vertico-temporal median operator and a spatial FIR filter respectively. This simple algorithm limits the hardware requirements to three field memories plus a very reduced number of algebraic operations per interpolated pixel. Usually linear techniques such as pixel repetition or the bilinear method are employed for image interpolation, which however either introduce artifacts (e.g. blocking effects) or tend to smooth edges. A higher quality rendition of the image is obtained by the concept of the Warped Distance among the pixels of an image. The computational load of the proposed approach is very small if compared to that of state-of-the-art nonlinear interpolation operators. Finally the contrast enhancement algorithm is a modified Unsharp Masking technique: a polynomial function is added to modulate the sharpening signal, which allows to discriminate between noise and signal and, at the same time, provides an appropriate amplification to low-contrast image details.

Comparison of adaptive linear interpolation and conventional linear interpolation for digital radiography systems

Abstract: Some large field digital radiography systems are currently under development using multiple detectors. These small size two dimension detectors are abutted together to cover a large field. Physical gaps existing between adjacent detectors produce seams between the resultant subimages. In this paper, an adaptive linear interpolation algorithm was introduced for estimating missing information at the seams and was compared with conventional linear interpolation and with nonlinear interpolation. The effectiveness of the algorithms was evaluated for relative/absolute errors. Phantom images were acquired using prototype digital radiography systems. Seams with width ranging from two pixel to six pixel were introduced and adaptive linear interpolation algorithms were applied to estimate missing information at the seams. Quantitatively, the adaptive interpolation offers at least equivalent or less error than that of the linear interpolation algorithm. The experimental results prove that there is a significant difference between two algorithms for tested radiographic images. The comparison results also show that the adaptive interpolation offers better performance than nonlinear interpolation. When developing large field digital radiography imaging systems, gaps between adjacent detectors should be minimized. For narrow seams, the adaptive linear interpolation algorithm is a practical solution because of its simplicity and effectiveness.

Adaptive regularized image interpolation using data fusion and steerable constraints

Abstract: Adaptiveregularizedimageinterp olationusingdatafusionandsteerableconstraintsJeong-HoShina,Jo on-Ki Paik, Je eryR. PricebandMongiA. AbidiaDepartmentofImage Engineering,Graduate Scho ol of AdvancedImaging Science,Multimedia, andFilm, Chung-AngUniversity221 Huksuk-Dong,Tong jak-Ku, Seoul156-756, KoreabDepartmentof Electrical andComputerEngineering,Universityof Tennessee,Knoxville, TN,USAABSTRACTThispap erpresentsanadaptiveregularizedimageinterp olationalgorithmfromblurredandnoisylowresolutionimage sequence, whichisdevelop ed in a general framework based on data fusion.This framework can preservethehigh frequency comp onents along the edge orientation in a restored high resolution image frame.This multiframeimage interp olation algorithm is comp osed of twolevels of fusion algorithm.One is to obtain enhanced low resolutionimages as an input data of the adaptive regularized image interp olation based on data fusion.The other one is toconstruct the adaptive fusion algorithm based on regularized image interp olation using steerable orientation analysis.In order to apply the regularization approachto the interp olation pro cedure, we rst present an observation mo deloflowresolutionvideoformationsystem.Basedontheobservationmo del,ecanhavaninterp olatedimagewhichminimizesbothresidualbetweenthehighresolutionandinterp olatedimageswithapriorconstraints.In addition, bycombining spatially adaptive constraints, directional high frequency comp onents are preserved witheciently suppressed noise.In the exp erimental results, interp olated images using the conventional algorithms areshowntocomparetheconventional algorithms withprop osedadaptive fusionbasedalgorithm.Exp erimentalresults show that the prop osed algorithm has the advantage of preserving directional high frequency comp onents andsuppressing undesirable artifacts such as noise.Keywords:Image interp olation, data fusion, steerable lter, regularization, resolution enhancement, adaptive edgepreserving1.INTRODUCTIONHighresolution(HR)restorationhasmanyapplicationsinimagepro cessing.Therearetwocategorieshighresolution restoration.One is the traditional image restoration concerned with the reconstruction of an uncorruptedimage from a blurred and noisy one.1The other one is the image interp olation asso ciated with increase of the spatialresolution of a single or a set of image frames.2,3HR image pro cessing applications such as digital high-de nitiontelevision (HDTV), aerial photo, medical imaging, surveillance video and military purp ose images, need HR imageinterp olation algorithms.In this pap er, we mainly deal with image interp olation algorithms to enhance the image quality in the sense ofresolution.By intro ducing image fusion and adaptive regularization algorithms, the prop osed algorithm can restoreHR image from low-resolution (LR) video.Originally, the ob jective of image fusion is to combine information frommultiple images of the same scene.As a result of image fusion, a single image which is more suitable for human andmachine p erception or further image-pro cessing tasks can b e obtained.4Data fusion algorithms are usually used inapplications ranging from Earth resource monitoring, weather forecasting, and vehicular trac control to militarytarget classi cation and tracking.5By utilizing the nature of image fusion mentioned ab ove, we can not only makeFurther author information:(Send corresp ondence to J. K. Paik)J. H. Shin :E-mail:shinj@ms.cau.ac.krJ. K. Paik :E-mail:paikj@cau.ac.krJ. R. Price :E-mail:jrp@utk.eduM. A. Abidi :E-mail:abidi@utk.edu

Context sensitive scaling device and method

Abstract: A method is provided for scaling a source image to produce a destination image. According to the method, a local context metric is calculated from a local portion of the source image. A convolution kernel is generated from a plurality of available convolution kernels based on the calculated local context metric, and the generated convolution kernel is used to generate at least one pixel in the destination image. Also provided is an image scaling device that receives pixels of a source image and outputs pixels of a scaled destination image. The image scaling device includes a context sensor, a kernel generator, and a scaler. The context sensor calculates a local context metric based on local source image pixels, and the kernel generator generates a current convolution kernel from a plurality of available convolution kernels based on the local context metric calculated by the context sensor. The scaler receives the coefficients of the current convolution kernel from the kernel generator, and uses the coefficients to generate at least one pixel of the destination image from pixels of the source image. Additionally, a display device that includes such an image scaling engine is provided.

Benefits of hybrid DCT domain image matching

Abstract: An enhancement to least squares image matching is proposed which combines a Discrete Cosine Transform (DCT) domain solution of the linearized normal equations, and resampling between iterations in the pixel domain. This approach reduces the size of the normal equations by discarding higher frequency DCT coefficients, while avoiding the overhead of image resampling in the DCT domain. A method for computing the DCT of the sampled derivative of a function from the DCT of its samples is given, and the least squares problem is framed in the DCT domain. In an experimental comparison between the proposed algorithm and an equivalent pixel domain algorithm, we find that the match time can be halved for 32 × 32 pixel windows, and reduced to 75% for 16 × 16 windows, while measures of match quality remain comparable or improve. The measures of much quality considered were the mean and standard deviation of the disparity error, and the number of match windows that converged. The optimum percentages of DCT coefficients for these window sizes were 20% for the 16 × 16 window and 10% for the 32 × 32 window. An 8 × 8 window size was also tested, but showed no speed-up over the pixel domain algorithm. The approach incorporates derivative estimates that result in better accuracy than can be achieved using the first differences of a pixel domain approach.

Two-dimensional Teager filters

Abstract: Publisher Summary This chapter presents a framework that describes a category of homogeneous quadratic Volterra filters to which the Teager filter belongs. The Teager filter is a homogeneous quadratic Volterra filter. The chapter analyzes their properties and derives a two-dimensional version of the Teager filter, stating that the Teager filter has the property that sinusoidal inputs generate constant outputs that are approximately proportional to the square of the input frequency. The chapter presents an intuitive interpretation of the frequency response of these filters, which facilitates a better understanding of their properties. This filter and modifications of it have been used successfully in image enhancement applications. Two-dimensional Teager filters have properties that are desirable for image enhancement since they can be approximated as mean-weighted highpass filters. Finally, the chapter illustrates some examples of the application of image enhancement, image interpolation, and image halftoning.

Method and device for image processing

Abstract: PROBLEM TO BE SOLVED: To perform optimum image interpolation on the basis of the edge information of image data prepared by mapping a discrete sample value SOLUTION: An edge sample value is identified by calculating an edge intensity value for every discrete sample value of the image data, an azimuth angle is stored for each edge sample value, and the edge sample value and the stored azimuth angle are processed by using a format process for each discrete sample value Thus, it is possible to select a kernel that is optimum to image interpolation

Improved multi-image resolution enhancement for colour images captured by single-CCD cameras

Abstract: Most consumer cameras use a single CCD sensor and a colour filter array (CFA) to take colour pictures. The array samples the red, green, and blue colour planes on sparse sampling lattices that are not coincident and not necessarily orthogonal. The "raw" images are then interpolated and post-processed to prepare them for display. Existing multiframe enhancement methods use the post-processed colour frames. We present a new multi-image enhancement algorithm which uses the raw image sequence directly to produce improved enhanced-resolution enlargements. We also demonstrate the limitations to resolution enhancement that occur when performing enhanced enlargement on low-resolution CFA-interpolated sequences.