Digital cameras reconstruct a full color image using the information captured by different color sensors (RGB) through a Color Filter Array (CFA). A demosaicking algorithm uses the information of the CFA data to restore a full color array. Demosaicking algorithms reconstruct two unknown color components for a pixel at each spatial location. During this stage, the values of other components will be measured mostly by neighboring pixels' information. Since the Fujifilm X-Trans pattern is one of the more successful CFAs, we attempt to model the demosaicking steps using this pattern and simulate both non-adaptive and adaptive demosaicking schemes. Kodak sample images are used in order to validate and compare different approaches with related previous work.

ABSTRACT Cubic panoramas provide an efficient solution to the implementation of immersive and omnidirectional displays for image‐based virtual navigation. To allow unrestricted navigation through the environment, a system must be capable of constructing novel views given the reference images, typically, cubic panoramas. This paper proposes a processing pipeline for cubic panorama synthesis based on 2D/3D triangular reprojection. This pipeline uses matching features to implement the triangular meshing on cubic panoramas, and introduces the matching line constraint, which significantly reduces the artifacts of straight‐line features in the rendered image and accelerates the division process. The modules of this pipeline take the geometrical characteristics of cubic panorama into account. Thus, it can handle the discontinuities of the cube edges when the triangular mesh covers different faces. Furthermore, these modules could be reconfigured to compose customized pipelines and generate virtual cubic panoramas of different quality based on the amount of features. The performance and efficiency of the proposed pipeline is demonstrated with comparison experiments. Copyright © 2013 John Wiley & Sons, Ltd.

This paper adapts the least-squares luma-chroma demultiplexing (LSLCD) demosaicking method to noisy Bayer color filter array (CFA) images. A model is presented for the noise in white-balanced gamma-corrected CFA images. A method to estimate the noise level in each of the red, green, and blue color channels is then developed. Based on the estimated noise parameters, one of a finite set of configurations adapted to a particular level of noise is selected to demosaic the noisy data. The noise-adaptive demosaicking scheme is called LSLCD with noise estimation (LSLCD-NE). Experimental results demonstrate state-of-the-art performance over a wide range of noise levels, with low computational complexity. Many results with several algorithms, noise levels, and images are presented on our companion web site along with software to allow reproduction of our results.

Panoramic images are a new type of visual data that provide many new possibilities as compared to the classic planar images. Tele-presence and virtual navigation are examples of such interesting applications. In order to enable smooth navigation across different view points, we propose a method for aligning cubic-panorama image datasets by using the concept of epipolar geometry. Unlike existing methods, which are limited to a pair of panoramas, our approach is accurate and applicable on datasets with large number of panoramic images. Also for the first time, a measure is introduced in order to enable objective evaluations.

A color filter array (CFA) used in a digital camera is a mosaic of spectrally selective filters, which allows only one color component to be sensed at each pixel. The missing two components of each pixel have to be estimated by methods known as demosaicking. The demosaicking algorithm and the CFA design are crucial for the quality of the output images. In this paper, we present a CFA design methodology in the frequency domain. The frequency structure, which is shown to be just the symbolic DFT of the CFA pattern (one period of the CFA), is introduced to represent images sampled with any rectangular CFAs in the frequency domain. Based on the frequency structure, the CFA design involves the solution of a constrained optimization problem that aims at minimizing the demosaicking error. To decrease the number of parameters and speed up the parameter searching, the optimization problem is reformulated as the selection of geometric points on the boundary of a convex polygon or the surface of a convex polyhedron. Using our methodology, several new CFA patterns are found, which outperform the currently commercialized and published ones. Experiments demonstrate the effectiveness of our CFA design methodology and the superiority of our new CFA patterns.

This paper addresses the problem of interpolating missing color components at the output of a Bayer color filter array (CFA), a process known as demosaicking. A luma-chroma demultiplexing algorithm is presented in detail, using a least-squares design methodology for the required bandpass filters. A systematic study of objective demosaicking performance and system complexity is carried out, and several system configurations are recommended. The method is compared with other benchmark algorithms in terms of CPSNR and S-CIELAB ∆E∗ objective quality measures and demosaicking speed. It was found to provide excellent performance and the best quality-speed tradeoff among the methods studied.

Image-based rendering (IBR) techniques provide a powerful tool for implementing free navigation in image-based virtual environments. Panoramas can produce an immersive and omnidirectional display of the virtual environment. This paper addresses the synthesis of virtual intermediate cubic panoramas given several reference panoramas. A 2D/3D complex triangular meshing representation of the scene is used to render the desired view, which involves solving a variety of visibility issues due to occlusion and disocclusion. The generalized disparity-based homography transformation can project these 2D/3D triangles onto the desired view in the sense of perspective accuracy. In addition, special measures are taken to handle the difficulties caused by the edge discontinuity of cubic panoramas. The experimental results show that we can obtain photo-realistic virtual images with high fidelity, indicating this is an effective approach for virtual cubic-panorama synthesis.

This paper presents an extension to the maximum likelihood estimation sample consensus (MLESAC) algorithm by estimating the prior validity of correspondences using both the measured data and a model hypothesis. Validity is determined based on the data set associated with the model that is considered as the best one so far in the previous random trials. The proposed robust algorithm is applied to estimate the fundamental matrix using randomly generated synthetic test data. Experiment results show that at various outlier ratios the proposed algorithm reduces the Sampson error and is also faster (in terms of the number of trials) in comparison to other conventional algorithms.

The acquisition of digital image data is not just a single step but actually consists of a whole chain of processing tasks. These have been extensively studied over many years. However, with different types of cameras different specific aspects of processing become important. For a panoramic camera setup, specific challenges as well as opportunities derive from the extremely wide angle of view. We present an integrated signal processing chain that mosaics 6 planar simultaneously acquired images into a cubic panorama. Challenges that are unique to omnidirectional imaging are addressed in order to maximize the obtained visual quality with respect to dynamic range, color reproduction, and spatial resolution.

This paper addresses filter design for the adaptive frequency-domain Bayer demosaicking algorithm. Both frequency-domain design and a least-squares design are considered. The effect of filter order on the reconstruction error is studied. A least-squares design method taking into account the characteristics of the adaptive frequency-domain algorithm is shown to lead to better demosaicking results than other published methods with lower complexity than the previous version of the adaptive frequency-domain algorithm.

This paper presents a hybrid disparity estimation algorithm which combines the pixel-based and region-based approaches. In the pixel-based approach, we use the Gabor transform and variational refinement, and the resulting disparities are combined with region information obtained from image segmentation so that a region matching scheme can be further applied. It will be shown with 3D reconstruction that our hybrid algorithm can give disparities with very high quality, in which some standard problems in disparity estimation can be solved.

This paper presents a system for introducing augmented reality (AR) enhancements into an image-based cubic panorama sequence. Panoramic cameras, such as the Point Gray Research Ladybug allow rapid capture and generation of panoramic sequences for users to navigate and view. Our AR system provides the ability for authors to add virtual content into the panoramic sequences. First, a user manually selects a planar region over which to add the content. Then the system automatically finds a matching planar region in all the panoramas, allowing the virtual content to propagate. No preconditioning of the imaged scene through the addition of physical markers is necessary. Instead, 3-D position information is obtained by matching interest-point features across the panoramic sequence. This paper presents an application of augmented reality algorithms to the unique case of pre-captured panoramic sequences.

The paper presents a disparity estimation method that combines a phase-based disparity estimation algorithm with an energy-based variational refinement approach, and adapts the method for the application of intermediate view interpolation. The Gabor transform is implemented using a set of quadrature-pair filters for finding the local phase information as well as for estimating the correspondences in stereo pairs. Then the estimated coarse disparity is applied to the variational refinement process which involves solving a partial differential equation (PDE) associated with an edge-preserving regularization functional. Two stereo image pairs in which the original intermediate images are available were used to test the algorithm, where our interpolated intermediate images could be compared with the originals. The performance with PSNR results shows that our method can provide very good images without introducing obvious artifacts.

A view interpolation algorithm for image-based rendering applications is proposed. In the proposed method, the view change is specified through the motions of feature points, which serve as control points. Triangulation of the images, combined with an affine transformation model, has been applied for texture mapping. Simulation results show that the new synthesized views are of good quality and can well represent the view change due to virtual camera movement.

An algorithm for disparity estimation with energy-based variational regularization using three image views is presented. In the new algorithm, dense disparity maps, including the disparity values for feature points, are first estimated by phase-based methods. Then a set of coupled partial differential equations (PDEs) are solved to refine these disparity maps together with the feature information. The visual appearance of the refined disparity is better than that of the results using similar variational regularization involving only one stereo pair of images. In addition, a new effective occlusion detection scheme is presented with good results.

This letter presents a new and simplified derivation of the frequency-domain representation of color images sampled with the Bayer color filter array. Two new demosaicking algorithms based on the frequency-domain representation are described and shown to give excellent results.

Regularization is one of the most promising methods for image up-sampling, which is an ill-posed inverse problem. A key element of such a regularization approach is the observation model relating the observed lower resolution (LR) image to the desired higher resolution (HR) up-sampled image, used in the data-fidelity term of the regularization cost function. This paper presents an algorithm to determine this observation model based on a model of the physical acquisition process for the LR image, and the ideal acquisition process for the desired HR image, both from the same underlying continuous image. The method is illustrated with typical scenarios corresponding to LR and HR cameras modeled by either Gaussian or rectangular apertures. Experiments with some regularized image up-samplers demonstrate the importance of using the correct, adapted observation model as determined by our algorithm. Index Terms-Camera aperture, data fidelity, image up-sampling, interpolation, multidimensional signal processing, observation model, power spectral density (PSD), super-resolution.

This paper presents a new formulation of the regularized image up-sampling problem that incorporates models of the image acquisition and display processes. We give a new analytic perspective that justifies the use of total-variation regularization from a signal processing perspective, based on an analysis that specifies the requirements of edge-directed filtering. This approach leads to a new data fidelity term that has been coupled with a total-variation regularizer to yield our objective function. This objective function is minimized using a level-sets motion that is based on the level-set method, with two types of motion that interact simultaneously. A new choice of these motions leads to a stable solution scheme that has a unique minimum. One aspect of the human visual system, perceptual uniformity, is treated in accordance with the linear nature of the data fidelity term. The method was implemented and has been verified to provide improved results, yielding crisp edges without introducing ringing or other artifacts.

This paper presents a novel approach to the reconstruction of images from nonuniformly spaced samples. This problem is often encountered in digital image processing applications. Nonrecursive video coding with motion compensation, spatiotemporal interpolation of video sequences, and generation of new views in multicamera systems are three possible applications. We propose a new reconstruction algorithm based on a spline model for images. We use regularization, since this is an ill-posed inverse problem. We minimize a cost function composed of two terms: one related to the approximation error and the other related to the smoothness of the modeling function. All the processing is carried out in the space of spline coefficients; this space is discrete, although the problem itself is of a continuous nature. The coefficients of regularization and approximation filters are computed exactly by using the explicit expressions of B-spline functions in the time domain. The regularization is carried out locally, while the computation of the regularization factor accounts for the structure of the nonuniform sampling grid. The linear system of equations obtained is solved iteratively. Our results show a very good performance in motion-compensated interpolation applications.

This paper presents a new method for registration and stitching of adjacent views in an image mosaic, specifically for cylindrical panoramas. The possible registration errors for the overlap area of two pictures, captured by a camera mounted on a tripod at different rotation angles, have been analyzed, based on a general camera rotation model. Then a novel algorithm is proposed, based on affine and refocusing adjustments on the optimal stripe block where the stitching is to be implemented. The registration errors can be largely reduced because a narrow stripe block is selected instead of the whole overlap area. Simulations on the overlap area registration algorithm show that our method can significantly reduce the stitching errors.

We show how to design an optimal camera aperture for a given display aperture defined on an arbitrary spatial or spatio-temporal sampling lattice. This is achieved by determining the reciprocal Riesz basis for the vector space of displayed images. The result can be used to specify ideal virtual cameras for a number of applications such as regularized image up-sampling. A case study for displaying a hexagonally sampled image on a cathode ray tube is presented.

The moving average is often used as an observation model for image interpolation, but is it a correct and accurate model for most circumstances? Are there other options? We present a novel theoretical analysis of the regularized image up-sampling problem focusing on the data fidelity term. We start with formulation of the physical acquisition processes the image has undergone and develop a generalized design for the correct and accurate data fidelity term for regularized image up-sampling.

This paper presents a new formulation of the regularized image up-sampling problem that incorporates models of the image acquisition and display processes. This approach leads to a new data fidelity term that has been coupled with a bounded-total-variation regularizer to yield our objective function. This objective function is minimized using the level-set method with two types of motion that interact simultaneously. The method was implemented and has been verified to provide good results, yielding crisp edges without introducing ringing or other artifacts.

The purpose of this paper is to introduce a fast automated white-noise estimation method which gives reliable estimates in images with smooth and textured areas. This method is a block-based method that takes the image structure into account and uses a measure other than the variance to determine if a block is homogeneous. It uses no thresholds and automates the way that block-based methods stop the averaging of block variances. The proposed method selects intensity-homogeneous blocks in an image by rejecting blocks of structure using a new structure analyzer. The analyzer used is based on high-pass operators and special masks for comers to allow implicit detection of structure and to stabilize the homogeneity estimation. For a typical image quality (PSNR of 20-40 dB) the proposed method outperforms other methods significantly and the worst-case estimation error is 3 dB which is suitable for real applications such as video surveillance or broadcasts. The method performs well even in images with few smooth areas and in highly noisy images.

Sufficient sampling for scene representation is an important topic in Image-Based Rendering (IBR). It determines both the rendered image quality and the quantity of image data that must be recorded, and it depends on the method by which the image samples are taken. The Concentric Mosaics technique is one way to get image samples of a scene, to which the sampling theory of the plenoptic function is applied in this paper to determine the required number of image samples at the different lengths of the rotation beam. The general spectral filtering model of image based rendering techniques is analyzed based on the Concentric Mosaics technique to provide a frequency domain interpretation. Simulations on both a numerical design example for a typical scene and the illustration of the different rendered image qualities with different number of image samples per rotation circle are provided.

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 cubic spline representation of the image. An objective function taking into account the similarity to the known samples and the regularity of the function is minimized in order to obtain a good approximation. We apply the developed algorithm to motion-compensated image interpolation. Under motion compensation, the resulting sampling grids are irregular and require irregular/regular interpolation. We show experimental results on real-world images and we compare our results with other methods proposed in the literature.

An anaglyph image allows the perception of depth when observed through colored glasses such as the familiar red/blue glasses. Although the method is very old, the techniques used to generate anaglyph images are very empirical. This paper describes a projection method to generate anaglyph stereoscopic images using the spectral absorption curves of the glasses, the spectral density functions of the display primaries and the colorimetric properties of the human observer.

The discrete wavelet transform (DWT) is a tool extensively used in image processing algorithms. It can be used to decorrelate information from the original image, which can thus help in compressing the data for storage, transmission or other post-processing purposes. However, the finite nature of such images gives rise to edge artifacts in the reconstructed data. A commonly used technique to overcome this problem is a symmetric extension of the image, which can preserve zeroth order continuity in the data. This still produces undesirable edge artifacts in derivatives and subsampled versions of the image. To overcome this problem, we present a new method of wavelet analysis by extrapolating the image based on a method presented by Therrien and El-Shaer (1989) and extended by Kuduvalli and Rangayyan (1993).

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.

Stereoscopic visualization systems based on liquid crystal shutter (LCS) eyewear and cathode-ray tube (CRT) displays provide today the best overall quality of three-dimensional (3-D) images and therefore have a dominant position in commercial as well as professional markets. Due to the CRT and LCS characteristics, however, such systems suffer from perceptual crosstalk ("shadows") at object boundaries that can reduce, and at times inhibit, the ability to perceive depth. In this paper, we propose a method to reduce such crosstalk. We present a simple model for intensity leak, we assess model parameters for a time-sequential LCS/CRT system and we propose a computationally efficient algorithm to eliminate the crosstalk. Since the full crosstalk elimination implies an unacceptable image degradation (reduction of contrast), we study the tradeoff between crosstalk elimination and image contrast. We describe experiments on synthetic and natural stereoscopic images and we discuss informal subjective viewing of processed images. Overall, the viewer response has been very positive; 3-D perception of many objects became either much easier or even effortless. Since the proposed algorithm can be easily implemented in real time (only linear scaling and table look-up are needed), we believe that it can be successfully used today in various stereoscopic applications suffering from image crosstalk. This is particularly true for PC-based 3-D viewing where the algorithm can be executed by the CPU or by an advanced graphics board.

This paper presents a novel method for representing motion information based on a dictionary of motion models and a tag image which indicates which motion model is used at any given image position. Each model is composed of low-order polynomial-based motion fields. The motion in most sequences can be adequately represented by a very small number of such motion models. We further present an efficient way of estimating and coding this representation. Comparative results are presented which indicate a performance superior to that of motion representations found in classical block-based codecs.

We propose a new method for the encoding of label images (also known as segmentation maps or alpha planes) that are often used to identify object location in region-based image and video coders. The method is contour-based and lossless with a contour model composed of two parts: a contour graph describing the topology of the contour network and a directional chain code to deal with the geometric part of the label image (internal contour points). The graph-based description of the topology is designed to minimize the cost of encoding the nodes, while the directional chain codes are compressed by arithmetic coding. The approach is flexible since separating the contour network into topological and geometrical parts allows the use of other lossless or lossy methods to encode the geometric part without changing the graph representation. The proposed method has been compared with an arithmetic encoder used in MPEG-4.

This article addresses the problem of designing two-channel near-perfect-reconstruction filter banks over multidimensional lattices. First, a cosine-modulated filter structure having arbitrary spatial shift and phase parameters is considered. The use of this structure leads to many possible two-channel multirate systems. The perfect reconstruction conditions are studied and appropriate choices for the parameters of the cosine-modulated structure are obtained. A simple but efficient unconstrained procedure for designing (possibly linear phase) near-perfect-reconstruction filter banks having good frequency responses (with arbitrary shapes) is proposed. A 2D design example is presented. The filter banks obtained can be used in transmultiplexers as well as in subband coders.

This paper presents a supervised algorithm for estimating the unknown nonlinearity undergone by the three color components of an image in the image acquisition process. The algorithm is based on the rank-one hypothesis, which postulates that the linear tristimulus values in a region of uniform surface color lie on a straight line through the origin. An objective function is formulated whose minimization yields the estimate of the unknown nonlinearity. Images corrected with the estimated inverse nonlinearity are shown to exhibit chromatic properties that are much more piecewise constant that in the original image. This property will be very useful in quantization and segmentation applications.

This paper presents some applications to FIR filter design of multi-D windows over arbitrary lattices and with arbitrary center of spatial symmetry. First, classic windows (such as Hamming, Blackman, etc.) are extended to windows over 1D and multi-D lattices with arbitrary spatial symmetry centers (which multirate applications sometimes require). Then the problem of obtaining a target frequency response with a good transition band from an ideal frequency response (made of a passband having constant gain and a stopband for which the gain is zero) is studied. A method to obtain a described target response using multi-D windows with application to the design of FIR filters is presented. Finally, a procedure for designing multi-D FIR filters by windowing is explained.

We consider the transmission of QCIF resolution (176/spl times/144 pixels) video signals over wireless channels at transmission rates of 64 kb/s and below. The bursty nature of the errors on the wireless channel requires careful control of transmission performance without unduly increasing the overhead for error protection. A dual-rate source coder is presented that adaptively selects a coding rate according to the current channel conditions. An automatic repeat request (ARQ) error control technique is employed to retransmit erroneous data-frames. The source coding rate is selected based on the occupancy level of the ARQ transmission buffer. Error detection followed by retransmission results in less overhead than forward error correction for the same quality. Simulation results are provided for the statistics of the frame-error bursts of the proposed system over code division multiple access (CDMA) channels with average bit error rates of 10/sup -3/ to 10/sup -4/.

The goal of this paper is to study the application of two-dimensional (2-D) finite-precision infinite impulse response (IIR) filters to enhanced National Television System Committee (NTSC) coding. It is well-known that suitable twoor three-dimensional (3-D) digital filtering greatly improves the quality of NTSC pictures by suppressing the interference between the luminance Y and the chrominances I, Q. Thus far, 2-D and 3-D finite impulse response (FIR) filters have been used to reduce or eliminate these cross effects. To achieve good performance, however, they require many coefficients. Since, in general, IIR filters need fewer coefficients to approximate a given magnitude response, we investigate the possibility of applying 2-D IIR filters to the NTSC encoding/decoding. We also study the feasibility of using digital filters for NTSC channel filtering; this would permit a digital-only encoder. To design suitable filters, we use a method based on multiple constraint optimization and simulated annealing. We propose a new implementation structure for the IIR filters that differs from the zero-phase FIR structure. We simulate the full NTSC coding chain and compare the resulting images for both filter types.

We present a block-oriented method for design of a finite set of linear predictors using a clustering algorithm on a training set. The predictors are tested with DPCM and 2-D code excited linear prediction (2-D-CELP) coding systems using switched prediction. An improved 2-D-CELP system using variable block size (VB 2-D-CELP) is proposed.

This research addresses the frequency multiplexing of multidimensional signals, having different bandwidths or defined on different lattices, with perfect or near perfect reconstruction i.e. zero or low crosstalk between signals and zero or low distortion of individual signals. The paper discusses important issues such as: what are the valid modulating frequencies, how to manage quadrature modulation, what are the conditions for perfect reconstruction, how many signals can be transmitted perfectly, when and why linear periodically time-varying (LPTV) filters must be used in the system, structure of filters to preserve compatibility with conventional frequency multiplexers and design procedures.

We show how the Pade table can be utilized to develop a new lattice structure for general two-channel bi-orthogonal perfect reconstruction (PR) filter banks. This is achieved through characterization of all two-channel bi-orthogonal PR filter banks. The parameter space found using this method is unique for each filter bank. Similarly to any other lattice structure, the PR property is achieved structurally and quantization of the parameters of the lattice does not effect this property. Furthermore, we demonstrate that for a given filter, the set of all complementary filters can be uniquely specified by two parameters, namely the end-to-end delay of the system and a scalar quantity.

A method is developed to compute the weight distribution of a set of points obtained from a lattice. The lattice is scaled (with possibly nonequal factors) along different dimensions, is shifted to an arbitrary point, and its lower dimensional subspaces are truncated within given shaping regions. Each branch in the lattice trellis diagram is labeled by the weight distribution of the corresponding coset incorporating the effects of scaling, shifting, and truncation. The weight distribution is obtained by multiplying the weight distribution of the serial branches and then adding the result over parallel paths.>

The goal of this work is to investigate a high quality, low bit rate image sequence coding scheme which takes full advantage of the temporal redundancies. Such schemes will be important for high-compression video coding, such as the one studied in MPEG-4. In conventional image sequence coding, a motion-compensated DPCM configuration is used to remove the temporal redundancies. The performance of the DPCM for such highly correlated sources at rates below one bps degenerates substantially. We have recently shown that for a highly correlated Gauss-Markov source, practical near rate-distortion function performance is possible. We used the entropy-constrained code-excited linear predictive (EC-CELP) quantization scheme to obtain such performances. Hence for low rate image sequence coding, to improve the rate-distortion performance, an EC-CELP configuration can replace the DPCM configuration to quantize the image intensities along motion trajectories. To apply EC-CELP quantization, we first investigate the methods of motion trajectory estimation and propose a temporal block motion compensation configuration. Then, the performance advantage of EC-CELP configuration over the conventional DPCM structures is shown.>

Visual communications over wireless networks require the efficient and robust coding of video signals for transmission over wireless links having time-varying channel capacity. The authors compare several schemes for encoding video data into two priority streams, thereby enabling the transmission of video data over wireless links to be switched between two bit rates. An H.261 (p/spl times/64) algorithm is modified to implement each candidate scheme. The algorithms are evaluated for a microcellular wireless environment and a clear-channel bit rate of 65 kb/s. The results show that by combining layering with automatic-repeat request (wireless-)link control, almost-wireline visual quality can be achieved. >

Results on block adaptive quantization of images using the adaptive dynamic range coding (ADRC) method are presented. In this method, the minimum and maximum intensities for each block are determined, and the pixels in the block are quantized with a uniform quantizer covering this range. The minimum and maximum values are transformed to midpoint and range values and then quantized and coded for transmission as side information. In order to reduce the visibility of blocking effects, ADRC can be applied to a subband decomposition of the image. This extension is called SADRC. Experimental results are presented for ADRC applied to the original image, as well as for four-band SADRC.>

The problem of designing finite precision one-dimensional infinite impulse response (IIR) digital filters for video processing is addressed. The design algorithm is based on simultaneous minimization of magnitude, phase, and stability errors in a discrete space of solutions using simulated annealing. It is demonstrated that the approach results in filters characterized by a substantially reduced nonlinearity of the phase response in filter passband, which is critical in any video processing application. To reduce image degradations due to ripples of the filter step response, another error term is introduced into the cost function. It is demonstrated that this additional term permits significant reduction of step response overshoots, and thus the visibility of degradations in a filtered image. The designed IIR filters are compared with their finite impulse response (FIR) counterparts in terms of characteristic parameters as well as distortion visibility in processed images.>

A stochastic approach to the estimation of 2D motion vector fields from time-varying images is presented. The formulation involves the specification of a deterministic structural model along with stochastic observation and motion field models. Two motion models are proposed: a globally smooth model based on vector Markov random fields and a piecewise smooth model derived from coupled vector-binary Markov random fields. Two estimation criteria are studied. In the maximum a posteriori probability (MAP) estimation, the a posteriori probability of motion given data is maximized, whereas in the minimum expected cost (MEC) estimation, the expectation of a certain cost function is minimized. Both algorithms generate sample fields by means of stochastic relaxation implemented via the Gibbs sampler. Two versions are developed: one for a discrete state space and the other for a continuous state space. The MAP estimation is incorporated into a hierarchical environment to deal efficiently with large displacements.>

The problem of designing finite precision one-dimensional infinite impulse response (IIR) filters with prescribed magnitude, phase, and stability constraints is studied. The design problem is formulated as the minimization of a cost function incorporating these conflicting requirements. The first two elements of the cost function express magnitude and group delay errors between the desired and the actual frequency responses of a filter, while the third one is related to its stability margin. This cost function is minimized using simulated annealing based on the Metropolis (1953) algorithm. Examples of several finite wordlength filters designed by the above method are presented and compared with Chebyshev and elliptic filters with rounded coefficients.>

A study of block-oriented motion estimation algorithms is presented, and their application to motion-compensated temporal interpolation is described. In the proposed approach, the motion field within each block is described by a function of a few parameters that can represent typical local motion vector fields.. A probabilistic formulation is then used to develop maximum-likelihood (ML) and maximum a posteriori probability (MAP) estimation criteria. The MAP criterion takes into account the dependence of the motion fields in adjacent blocks. A procedure for minimizing the resulting objective function based on the Gauss-Newton algorithm is presented. The use of multiresolution techniques, essential for satisfactory performance, is discussed. Experimental results evaluating the algorithms for the task of motion-compensated temporal interpolation are presented. The relative complexity of the algorithms is also discussed.>

An extension of the Bayesian estimation of 2-D motion in spatiotemporally sampled image sequences is presented by incorporating the color cue into the estimation process. Instead of scalar image intensity, three-component vector representation of color images is used, thus allowing Y-C1-C2, RGB, or other formats. The maximum a posteriori probability estimation is shown to result in a three-term energy minimizationj. A white Gaussian noise model is used for the displaced pel differences of each image component, and a coupled vector-binary Markov random field model is used for displacement and discontinuity fields. The resulting criterion is optimized using the method of discrete state-space simulated annealing. Improvements in the quality of estimated displacement fields due to additional color information are demonstrated through several experimental results.>

A general block differential method for estimating two-dimensional motion in video sequences is presented. A probabilistic interpretation is given, and other existing methods in the same category are generalized by the introduction of a parametric model to compute motion inside blocks as well as an explicit communication between disjoint blocks. The solution algorithm also generalizes other algorithms which formulate the estimation through a linearized version of the displaced pel difference. The method is applied to motion-compensated interpolation of a video sequence. Very good results are obtained for specific implementations of the general method. A heuristic block-splitting scheme used to enhance block estimation around moving boundaries is discussed.>

Presents a probabilistic formulation for motion estimation in images and a stochastic algorithm for minimization of the associated objective function. It is shown that motion estimation, an ill-posed problem, can be regularized by means of a Bayesian estimation approach. The unknown motion field is modeled as a two-dimensional vector Markov random field with a certain neighbourhood system. The posterior distribution of the motion field given image observations is then a Gibbs distribution. Maximization of this a posteriori probability to obtain the MAP estimate of the motion field is achieved by simulated annealing. Results of the estimation procedure applied to television sequences with natural motion are presented.>

Coding schemes suitable for progressive transmission of still pictures in NTSC composite format are studied. Transform coding methods based on the discrete cosine transform (DCT) and the WalshHadamard transform (WHT) are used. The transform coefficients are segmented into groups having similar properties. Variable blocklength to variable length codes are used to encode the quantized coefficients for each of the groups. Progressive transmission is achieved by sending the coefficients belonging to particular groups in successive passes over the image. Results are presented for coding the NTSC signal sampled at four times the color subcarrier frequency with an orthogonal structure, and at twice the subcarrier frequency with a hexagonal structure.

Number theoretic transforms using the modulus2^{2_{q}} - 2^{q} + 1, which require no multiplications, are introduced. These transforms have 50% greater sequence lengths for a given wordlength than Fermat transforms, and there is greater flexibility in the choice of wordlength. Furthermore, arithmetic modulo2^{2_{q}} - 2^{q} + 1is not significantly more complex than arithmetic modulo a Fermat number. Complex number theoretic transforms using these moduli are also discussed and techniques for using them to convolve real data are described.

Necessary and sufficient conditions for a direct sum of local rings to support a generalized discrete Fourier transform are derived. In particular, these conditions can be applied to any finite ring. The function O(N) defined by Agarwal and Burrus for transforms over ZN is extended to any finite ring R as O(R) and it is shown that R supports a length m discrete Fourier transform if and only if m is a divisor of O(R) This result is applied to the homomorphic images of rings-of algebraic integers.