Showing papers in "Pattern Recognition in 2000"

TL;DR: A new method is presented for adaptive document image binarization, where the page is considered as a collection of subcomponents such as text, background and picture, which adapts and performs well in each case qualitatively and quantitatively.

TL;DR: It is proved that the most expressive vectors derived in the null space of the within-class scatter matrix using principal component analysis (PCA) are equal to the optimal discriminant vectorsderived in the original space using LDA.

TL;DR: The superiority of the GA-clustering algorithm over the commonly used K-means algorithm is extensively demonstrated for four artificial and three real-life data sets.

TL;DR: It is shown that sequential floating search methods are suitable for small- and medium-scale problems and genetic algorithms are suitableFor large-scale feature selection algorithms, the goodness of a feature subset is measured by leave-one-out correct-classification rate of a nearest-neighbor (1-NN) classifier.

TL;DR: A simple method of replacing costly computation of nonlinear (on-line) Bayesian similarity measures by inexpensive linear subspace projections and simple Euclidean norms is derived, thus resulting in a significant computational speed-up for implementation with very large databases.

TL;DR: A distribution-based classification approach and a set of recently developed texture measures are applied to rotation-invariant texture classification and the results show much better performance than for the CSAR features.

TL;DR: Differences and similarities are shown between this mean combination rule and the product combination rule in theory and in practice.

TL;DR: Overall, using the Gabor "lter magnitude response given a frequency bandwidth and spacing of one octave and orientation bandwidth and spaced of 303 augmented by a measure of the texture complexity generated preferred results.

TL;DR: It is shown that for any set of invariants there exists a relatively small basis by means of which all other invariants can be generated and how to prove its independence and completeness is presented.

TL;DR: This paper attempts to eliminate the assumption that the written text is fixed by presenting a novel algorithm for automatic text-independent writer identification by taking a global approach based on texture analysis, where each writer's handwriting is regarded as a different texture.

TL;DR: The novelty of the approach is to adapt the connected component operator to solve the watershed segmentation problem and the resulting algorithm is independent of the number of grey-levels, employs simple data structures, requires less error prone memory management, and issues a lower complexity and a short running time.

TL;DR: This method can threshold various poor quality gray-scale document images automatically without need of any prior knowledge of the document image and manual fine-tuning of parameters and has a wider range of applications compared with other methods.

TL;DR: The formulation of the best error-reject trade-off and the related optimal reject rule was given by Chow and can be obtained using the so-called “reject” option.

TL;DR: A novel adaptive direct fuzzy contrast enhancement method based on the fuzzy entropy principle and fuzzy set theory is proposed that is very effective in contrast enhancement as well as in preventing over-enhancement.

TL;DR: It is argued that a fruitful direction for future research may lie in weighing information about facial features together with localized image features in order to provide a better mechanism for feature selection.

TL;DR: Both Bayesian classifiers and neural networks are employed to test the efficiency of the proposed feature and the achieved identification success using a long word exceeds 95%.

TL;DR: An active-camera real-time system for tracking, shape description, and classi"cation of the human face and mouth expressions using only a PC or equivalent computer based on use of 2-D blob features.

TL;DR: A new approach to automatic face location on gray-scale static images with complex backgrounds by approximately detects the image positions where the probability of finding a face is high and improves the location accuracy of the candidate faces.

TL;DR: A combined approach for facial feature extraction and determination of gaze direction is proposed that employs some improved variations of the adaptive Hough transform for curve detection, minima analysis of feature candidates, template matching for inner facial feature localization and active contour models for inner face contour detection.

TL;DR: Experimental results show that this approach for segmenting cell images is both fast and robust, and that this methodology may be used for automatic classification as part of a computer-aided medical decision making technique.

TL;DR: A combination of signal normalization preprocessing and the use of invariant features makes the HMM based writer independent handwriting recognition system robust with respect to variability among di!erent writers as well as di?erent writing environments and ink collection mechanisms.

TL;DR: This paper considers a clustering problem where a given data set is partitioned into a certain number of natural and homogeneous subsets such that each subset is composed of elements similar to one another but different from those of any other subset.

TL;DR: A set of axioms is proposed and discussed to distinguish different cluster compactness and separation criteria, and is extended to nested partitionings for hierarchical data clustering in the case of sparse proximity matrices.

TL;DR: In this detection system, the morphology-based eye-analogue segmentation process is able to reduce the background part of a cluttered image by up to 95%.

TL;DR: In this article, a class of iterative morphological image operators with applications to sharpen digitized gray-scale images was introduced, and the underlying partial differential equation that governs this class of image operators was introduced.

TL;DR: A system of two separate phases for signature recognition and verification is developed based on a multistage classifier and a combination of global and local features.

TL;DR: A general class of multilayer feed-forward neural networks where the combination of inputs in every node is formed by hybrid linear and nonlinear (of the morphological/rank type) operations is proposed.

TL;DR: A novel vision-based system for automatic interpretation of a limited set of dynamic hand gestures using a finite state machine to translate the richness of the human gestural communication power to a machine for a better man–machine interaction.

TL;DR: This chain code is inspired by the work of Guzman (MCC Technical Report Number: ACA-254-87, July 13, 1987) for representing 3D Stick Bodies and allows for a curve descriptor invariant under translation and rotation and for representing solids composed of voxels.

TL;DR: A new technique for the recognition of Arabic text using the C4.5 machine learning system that can generalize over the large degree of variations between different fonts and writing style and recognition rules can be constructed by examples.