The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques and methods imported from statistical learning theory have bean receiving increasing attention. The design of a recognition system requires careful attention to the following issues: definition of pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design and learning, selection of training and test samples, and performance evaluation. In spite of almost 50 years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition, and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.

A graduated assignment algorithm for graph matching is presented which is fast and accurate even in the presence of high noise. By combining graduated non-convexity, twoway (assignment) constraints, and sparsity, large improvements in accuracy and speed are achieved. Its low order computational complexity [O(lm), where l and m are the number of links in the two graphs] and robustness in the presence of noise offer advantages over traditional combinatorial approaches. The algorithm, not restricted to any special class of graph, is applied to subgraph isomorphism, weighted graph matching, and attributed relational graph matching. To illustrate the performance of the algorithm, attributed relational graphs derived from objects are matched. Then, results from twenty-five thousand experiments conducted on 100 node random graphs of varying types (graphs with only zero-one links, weighted graphs, and graphs with node attributes and multiple link types) are reported. No comparable results have...

A new approach to the recognition of temporal behaviors and activities is presented. The fundamental idea, inspired by work in speech recognition, is to divide the inference problem into two levels. The lower level is performed using standard independent probabilistic temporal event detectors such as hidden Markov models (HMMs) to propose candidate detections of low level temporal features. The outputs of these detectors provide the input stream for a stochastic contextfree grammar parsing mechanism. The grammar and parser provide longer range temporal constraints, disambiguate uncertain low level detections, and allow the inclusion of a priori knowledge about the structure of temporal events in a given domain. To achieve such a system we provide techniques for generating a discrete symbol stream from continuous low level detectors, for enforcing temporal exclusion constraints during parsing, and for generating a control method for low level feature application based upon the current parsing state. We demonstrate the approach in several experiments using both visual and other sensing data.

Despite the success of many pattern recognition systems in constrained domains, problems that involve noisy input and many classes remain difficult. A promising direction is to use several classifiers simultaneously, such that they can complement each other in correctness. This thesis is concerned with decision combination in a multiple classifier system that is critical to its success. A multiple classifier system consists of a set of classifiers and a decision combination function. It is a preferred solution to a complex recognition problem because it allows simultaneous use of feature descriptors of many types, corresponding measures of similarity, and many classification procedures. It also allows dynamic selection, so that classifiers adapted to inputs of a particular type may be applied only when those inputs are encountered. Decisions by the classifiers are represented as rankings of the class set that are derivable from the results of feature matching. Rank scores contain more ...

Recent scientific and technological advances have witnessed an abundance of structural patterns modeled as graphs. As a result, it is of special interest to process graph containment queries effectively on large graph databases. Given a graph database G, and a query graph q, the graph containment query is to retrieve all graphs in G which contain q as subgraph(s). Due to the vast number of graphs in G and the nature of complexity for subgraph isomorphism testing, it is desirable to make use of high-quality graph indexing mechanisms to reduce the overall query processing cost. In this paper, we propose a new cost-effective graph indexing method based on frequent tree-features of the graph database. We analyze the effectiveness and efficiency of tree as indexing feature from three critical aspects: feature size, feature selection cost, and pruning power. In order to achieve better pruning ability than existing graph-based indexing methods, we select, in addition to frequent tree-features (Tree), a small number of discriminative graphs (∆) on demand, without a costly graph mining process beforehand. Our study verifies that (Tree+∆) is a better choice than graph for indexing purpose, denoted (Tree+ ∆ ≥Graph), to address the graph containment query problem. It has two implications: (1) the index construction by (Tree+∆) is efficient, and (2) the graph containment query processing by (Tree+∆) is efficient. Our experimental studies demonstrate that (Tree+∆) has a compact index structure, achieves an order of magnitude better performance in index construction, and most importantly, outperforms up-to-date graphbased indexing methods: gIndex and C-Tree, in graph containment query processing. 1.

Abstract—This paper describes a novel solution to the rigid point pattern matching problem in Euclidean spaces of any dimension. Although we assume rigid motion, jitter is allowed. We present a noniterative, polynomial time algorithm that is guaranteed to find an optimal solution for the noiseless case. First, we model point pattern matching as a weighted graph matching problem, where weights correspond to Euclidean distances between nodes. We then formulate graph matching as a problem of finding a maximum probability configuration in a graphical model. By using graph rigidity arguments, we prove that a sparse graphical model yields equivalent results to the fully connected model in the noiseless case. This allows us to obtain an algorithm that runs in polynomial time and is provably optimal for exact matching between noiseless point sets. For inexact matching, we can still apply the same algorithm to find approximately optimal solutions. Experimental results obtained by our approach show improvements in accuracy over current methods, particularly when matching patterns of different sizes. Index Terms—Point pattern matching, graph matching, graphical models, Markov random fields, junction tree algorithm. 1

this paper to structural and syntactic methods does of course in no ways mean that we despise or reject the statistical approach to pattern recognition. Actually, the complementarity of these two families has already been extensively proved [24], attributed and stochastic grammars have been proposed [36, 58] and used in various applications [53, 55], probabilistic relaxation is often used in connection with structural matching [9], etc. But in this paper, we concentrate on the use of structural and syntactic pattern recognition methods when dealing with the analysis of graphics and line drawings. 2 Structural and syntactic methods in document and graphics analysis

Given a set of model graphs D and a query graph q, containment search aims to find all model graphs g ∈ D such that q contains g (q ⊇ g). Due to the wide adoption of graph models, fast containment search of graph data finds many applications in various domains. In comparison to traditional graph search that retrieves all the graphs containing q (q ⊆ g), containment search has its own indexing characteristics that have not yet been examined. In this paper, we perform a systematic study on these characteristics and propose a contrast subgraph-based indexing model, called cIndex. Contrast subgraphs capture the structure differences between model graphs and query graphs, and are thus perfect for indexing due to their high selectivity. Using a redundancy-aware feature selection process, cIndex can sort out a set of significant and distinctive contrast subgraphs and maximize its indexing capability. We show that it is NP-complete to choose the best set of indexing features, and our greedy algorithm can approximate the one-level optimal index within a ratio of 1 − 1/e. Taking this solution as a base indexing model, we further extend it to accommodate hierarchical indexing methodologies and apply data space clustering and sampling techniques to reduce the index construction time. The proposed methodology provides a general solution to containment search and indexing, not only for graphs, but also for any data with transitive relations as well. Experimental results on real test data show that cIndex achieves near-optimal pruning power on various containment search workloads, and confirms its obvious advantage over indices built for traditional graph search in this new scenario. 1.

The more a priori knowledge we encode into a signal processing algorithm, the better performance we can expect. In this paper, we overview several approaches to capturing the structure of singularities (edges, ridges, etc.) in wavelet-based signal processing schemes. Leveraging results from approximation theory, we discuss nonlinear approximations on trees and point out that an optimal tree approximant exists and is easily computed. The optimal tree approximation inspires a new hierarchical interpretation of the wavelet decomposition and a tree-based wavelet denoising algorithm that suppresses spurious noise bumps. Keywords: Wavelets, trees, nonlinear approximation, Besov space,optimization 1. INTRODUCTION The wavelet transform provides a natural setting for developing new signal and image processing algorithms, especially for signals and images rich in singularities (edges, ridges, and other transients). Since wavelets form a basis, 1,2 they can reproduce arbitrary functions, fro...

This report addresses the problem of using probabilistic formal languages to describe and understand actions with explicit structure. The paper explores a probabilistic mechanisms of parsing the uncertain input string aided by a stochastic context-free grammar. This method, originating in speech recognition, allows for combination of a statistical recognition approach with a syntactical one in a unified syntactic-semantic framework for action recognition. The basic approach is to design the recognition system in a two-level architecture. The first level, a set of independently trained component event detectors, produces the likelihoods of each component model. The outputs of these detectors provide the input stream for a stochastic context-free parsing mechanism. Any decisions about supposed structure of the input are deferred to the parser, which attempts to combine the maximum amount of the candidate events into a most likely sequence according to a given Stochastic Context-Free Grammar (SCFG). The grammar and parser enforce longer range temporal constraints, disambiguate or correct uncertain or mis-labeled low level detections, and allow the inclusion of a priori knowledge about the structure of temporal events in a given domain. The method takes into consideration the continuous character of the input and performs “structural rectification” of it in order to account for misalignments and ungrammatical symbols in the stream. The presented technique of such a rectification uses the structure probability maximization to drive the segmentation.