Showing papers in "Pattern Recognition Letters in 2003"

TL;DR: This paper addresses the problem of information fusion in biometric verification systems by combining information at the matching score level by combining three biometric modalities (face, fingerprint and hand geometry).

TL;DR: A new and efficient steganographic method for embedding secret messages into a gray-valued cover image that provides an easy way to produce a more imperceptible result than those yielded by simple least-significant-bit replacement methods.

TL;DR: A measure for identifying the physical significance of an outlier is designed, which is called cluster-based local outlier factor (CBLOF), which is meaningful and provides importance to the local data behavior.

TL;DR: The algorithm for feature selection is based on an application of a rough set method to the result of principal components analysis (PCA) used for feature projection and reduction.

TL;DR: A palmprint recognition method based on eigenspace technology by means of the Karhunen-Loeve transform, where the original palmprint images are transformed into a small set of feature space, called "eigenpalms", which are the eigenvectors of the training set and can represent the principle components of the palmprints quite well.

TL;DR: This paper describes the texture classification using (i) wavelet statistical features, (ii) wavelets co-occurrence features and (iii) a combination of wavelets statistical features and co- Occurrence features of one level wavelet transformed images with different feature databases.

TL;DR: The proposed similarity measures can provide a useful way for measuring IFSs more effectively and are more reasonable than some existing similarity measures in general cases.

TL;DR: The state of the art on automatic road extraction for GIS update from aerial and satellite imagery is surveyed and a novel classification of road extraction methods according to the criteria is included in order to provide a significant contribution to research in this topic.

TL;DR: The experimental results show that, in the proposed method, the palmprint images with resolution 32 × 32 are optimal for medium security biometric systems while those with resolution 64 × 64 are optimalFor high security biometrics systems.

TL;DR: The objective of this paper is to develop an approach for efficiently and quantitatively evaluating thresholding algorithms for change detection in a surveillance environment and to test the performance of eight different thresholded algorithms using more than 4000 images with two different texture environments.

TL;DR: This paper illustrates the application of the modified similarity measure on a simple pattern recognition problem and shows how a simple modification of the Dengfeng-Chuntian operator may correct this problem.

TL;DR: A new algorithm for digital images unsupervised enhancement with simultaneous global and local effects, called ACE for Automatic Color Equalization, based on a computational model of the human visual system that merges the two basic "Gray World" and "White Patch" global equalization mechanisms.

TL;DR: This paper proposes a fast NCC computation for defect detection using a sum-table scheme, which allows the calculations of image mean, image variance and cross-correlation between images to be invariant to the size of template window.

TL;DR: Experimental results indicate that the proposed MGF enhancement algorithm can reduce the FRR of a fingerprint matcher by approximately 2% at a FAR of 0.01%.

TL;DR: A (k, n)-threshold visual cryptography scheme is proposed to encode a secret image into n shadow images, where any k or more of them can visually recover the secret image, but any k - 1 or fewer of them gain no information about it.

TL;DR: A new fuzzy model structure is proposed where each rule can represent more than one classes with different probabilities, which can be considered as an extension of the quadratic Bayes classifier that utilizes mixture of models for estimating the class conditional densities.

TL;DR: A comprehensive comparative study of artificial neural networks, learning vector quantization and dynamic time warping classification techniques combined with stationary/non-stationary feature extraction for environmental sound recognition shows 70% recognition using mel frequency cepstral coefficients or continuous wavelet transform with dynamic time Warping.

TL;DR: A three-level thresholding method for image segmentation is presented, based on probability partition, fuzzy partition and entropy theory, and a new fuzzy entropy has been defined through probability analysis.

TL;DR: A new adaptive vector median filtering scheme taking the advantage of the optimal filtering situation and the robust order-statistic theory, is provided, based on the set of vector-valued order-Statistics with the smallest distances to other samples in the input set.

TL;DR: Complex filters, applied to the orientation field in multiple resolution scales, are used to detect the symmetry and the type of symmetry and an SP is identified by its symmetry properties.

TL;DR: The performance of this segmentation algorithm is superior to traditional single resolution techniques such as texture spectrum, co-occurrences, local linear transforms, etc.

TL;DR: In this article, a generalized version of the conventional k-means clustering algorithm is presented, which is applicable to ellipse-shaped data clusters without dead-unit problem, and performs correct clustering without pre-assigning the exact cluster number.

TL;DR: A pattern recognition approach to network intrusion detection based on the fusion of multiple classifiers is proposed, and five decision fusion methods are assessed by experiments and their performances compared.

TL;DR: It is shown that good pivots are outliers, but that selecting outliers does not ensure that good pivotots are selected, and an efficiency measure to compare two pivot sets is proposed.

TL;DR: The proposed face recognition technique is based on the implementation of the principal component analysis algorithm and the extraction of depth and colour eigenfaces and Experimental results show significant gains attained with the addition of depth information.

TL;DR: This result indicates that the best practical combination is PCA with SVM for face recognition, as the training time for ICA is much larger than that of PCA.

TL;DR: A comparison between NMF, WNMF and the well-known principal component analysis (PCA) in the context of image patch classification has been carried out and it is claimed that all three techniques can be combined in a common and unique classifier.

TL;DR: It is shown that beam angle statistics (BAS) is invariant to translation, rotation, scale and is insensitive to distortions.

TL;DR: This paper reviews various segmentation proposals that integrate edge and region information and highlights different strategies and methods for fusing such information.

TL;DR: The effects of spatial, fixed and adaptive, filtering, as well as, of grey scale encoding on the detection of lesions are analytically described and illustrated with representative images.