Chair: H. Joel Trussell, North Carolina State University (USA)
Davide Corso, D.I.B.E.-University of Genova (ITALY)
Roberto Fioravanti, D.I.B.E.-University of Genova (ITALY)
Stefano Fioravanti, D.I.B.E.-University of Genova (ITALY)
The paper addresses the problem of a real time approach for thin structures identification. The proposed method make use of adaptive morphological operators in which the normal textural properties are brought to bear on the filter design in order to enhance the output related to anomalous structures. We discuss how the introduction of adaptivity criteria is useful to reduce the sensitivity of morphological filtering to local variations typical of textures and to noise. In addition, theoretical investigations related to the problem of the optimization algorithm are presented and discussed. Experimental results have been carried out both on classical natural textures and on ceramic tile images and prove the efficiency of the proposed approach in many practical applications.
Jonathan K. Su, Georgia Institute of Technology(USA)
Russell M. Mersereau, Georgia Institute of Technology(USA)
An iterative post-processing algorithm for reducing blocking and ringing in block transform-coded images is presented. It smooths the decompressed image and uses convex projections theory to preserve the image's edges and bitstream. The algorithm exploits the coding scheme's block structure and ringing behavior. In experiments it outperforms an existing method and gives significant subjective and objective improvement over the decompressed image. Sufficient conditions for algorithm convergence are given.
J.-P. Leduc, IRISA/INRIA (FRANCE)
J.-M. Odobez, IRISA/INRIA (FRANCE)
C. Labit, IRISA/INRIA (FRANCE)
This paper intends to present new approaches in the field of motion-compensated spatio-temporal filters applied to digital image sequences. In time-varying imagery, the temporal correlation information of pixel intensities is folded by motions which may originate from both camera and object displacements. Motion-compensated filters are here defined as temporal filters applied along assumed motion trajectories. As a matter of fact, this paper deals with three-dimensional spatio-temporal filters and aims at generalizing the motion-compensated temporal filtering process as the product of two distinct operators. The first operator depends only on the estimated motion parameters derived from both motion-based image segmentations and parametric affine modelings of regions in motion. The second operator analyzes only the correlations of image-by-image intensities measured along the assumed motion trajectories. Multiresolution filters or wavelets may be consequently applied along the motion trajectories to produce optimum and adaptive resulting procedures for purposes like spatio-temporal prediction, interpolation and smoothing. In this paper, applications are provided to cover the field of image sequence coding and interpolation.
E. Abreu, University of California - Santa Barbara (USA)
S.K. Mitra, University of California - Santa Barbara (USA)
We propose an efficient nonlinear algorithm to suppress impulse noise from highly corrupted images while preserving details and features. The method is applicable to all impulse noise models, including fixed valued (equal height or salt and pepper) impulses and random valued (unequal height) impulses, covering the whole dynamic range. The algorithm is based on a detection-estimation strategy. If a signal sample is detected as a corrupted sample, it is replaced with an estimation of the true value, based on neighborhood information. Otherwise it is kept unchanged. The technique achieves excellent tradeoff between the suppression of noise, and preserving the details and edges without undue increase in computational complexity. Extensive simulation tests indicate that our method performs better than other existing algorithms, including the well known median filters. Illustrative examples included in the paper verify the capability of the proposed approach.
Russell C. Hardie, University of Dayton
Kenneth E. Barner, University of Delaware (USA)
Extended permutation (EP) filters are defined and analyzed in this paper. In particular, we focus on extended permutation rank selection (EPRS) filters. These filters are constrained to output an order statistic from an extended observation vector. This extended vector includes N observation samples and K statistics that are functions of the observation samples. The rank permutations from selected samples in this extended observation vector are used as the basis for selecting an order statistic output. We show that by including the sample mean in the extended observation vector, the filters exhibit excellent edge enhancement properties. We also show that several previously defined classes of rank order based edge enhancers can be formulated as subclasses of EPRS filters. Edge enhancement properties are developed and an L_(eta) norm EPRS filter optimization procedure is presented. Finally, extensive computer simulation results are presented comparing the performance of EPRS and other sharpening filters in edge enhancement applications.
S. Grace Chang, University of California (USA)
Zoran Cvetkovic, University of California (USA)
Martin Vetterli, University of California (USA)
One problem of image interpolation refers to magnifying a small image without loss in image clarity. We propose a wavelet based method which estimates the higher resolution information needed to sharpen the image. This method extrapolates the wavelet transform of the higher resolution based on the evolution of the wavelet transform extrema across the scales. By identifying three constraints that the higher resolution information needs to obey, we enhance the reconstructed image through alternating projections onto the sets defined by these constraints.
Sadik D. Bayrakeri, Georgia Institute of Technology (USA)
Russell M. Mersereau, Georgia Institute of Technology (USA)
A nonlinear method for image interpolation is presented based on spa- tial domain directional interpolation. Existing directional interpo- lation algorithms, which only consider edge regions, are extended to the whole image by interpolating in multiple directions. The interpo- lated values along various directions are combined using directional weights, which depend on the variation in that direction. The inter- polation value for each direction is assigned based on the magnitude of its directional derivative.
Irving Linares, Georgia Institute of Technology (USA)
Our previous work aimed at reducing the blocking distortion of DCT coded images is further improved by introducing the Optimal PSNR Estimated Spectrum Adaptive Postfilter (ESAP) Algorithm. ESAP searches for a one dimensional log- sigmoid weighting function which, when applied to the separable, interpolated local block estimated spectrum of the coded image, minimizes the Mean Square Error (MMSE) with respect to the original image using a 2-D steepest descent search. Convergence is obtained in a few iterations for integer parameters. A unique maximum PSNR is guaranteed given the asymptotic exponential overshoot behavior of the surface generated by ESAP. We obtained PSNR improvement of 1.5 dB over nonpostfiltered JPEG images as well as subjective improvement. ESAP is based on a DFT analysis of the DCT basis functions and uses spatially adaptive separable FIR postfilters.
Ioan Tabus, Tampere University of Technology (FINLAND)
Moncef Gabbouj, Tampere University of Technology (FINLAND)
For the nonlinear classes of filters based on the median archetype, e.g. stack, Weighted Order Statistics (WOS), morphological filters the techniques exploiting recursiveness of embedded structures were not used yet. We investigate in this paper the possibilities of improving the speed of the optimal design techniques restating the optimal design problem as a sequence of optimal problems for embedded structures. The speed up effect of recursive-in-order design techniques is very significant but it is not the only benefit of this principle: it also allows the evaluation of the optimal structure of the filter, comparing the performances of the optimal filters having various structural parameters and observing where the performance index curve starts to flatten with the increase in the structure size
W.W. Cindy Jiang, Columbia University (USA)
A scheme which converts graytone text images of low spatial resolution to bi-level images of higher spatial resolution for character recognition are presented. Higher recognition rates are achieved when text images are processed using the proposed scheme. A good application of the proposed scheme is the recognition of characters in scene images.