Type classification of fingerprints: a syntactic approach

A fingerprint classification procedure using a computer is described. It classifies the prints into one of ten defined types. The procedure is implemented using PICAP (picture array processor). The picture processing system includes a TV camera input and a special picture processor. The first part of the procedure is a transformation of the original print to a sampling matrix, where the dominant direction of the ridges for each subpicture is indicated. After smoothing, the lines in this pattern are traced out and converted to strings of symbols. Finally, a syntactic approach is adopted to make the type classification based on this string of symbols.     

A multidimensional approach to syntactic pattern recognition

This paper describes a syntactic method for representing the primitive parts of a pattern as nodes of a type of directed graph. A linear representation of the digraph can then be presented to a regular unordered tree automaton for classification. Regular unordered tree automata can be simulated by deterministic pushdown automata, so this procedure can be implemented easily. Regular u-tree automata and the corresponding generative systems, regular u-tree grammars are formally defined. Several results are shown which are applicable to all syntactic pattern recognition schemes involving the use of primitives.     

A syntactic approach to scale-space-based corner description

Planar curves are described by information about corners integrated over various levels of resolution. The detection of corners takes place on a digital representation. To compensate for ambiguities arising from sampling problems due to the discreteness, results about the local behavior of curvature extrema in continuous scale-space are employed     

A syntactic approach to twig-query matching on XML streams

Query matching on XML streams is challenging work for querying efficiency when the amount of queried stream data is huge and the data can be streamed in continuously. In this paper, the method Syntactic Twig-Query Matching (STQM) is proposed to process queries on an XML stream and return the query results continuously and immediately. STQM matches twig queries on the XML stream in a syntactic manner by using a lexical analyzer and a parser, both of which are built from our lexical-rules and grammar-rules generators according to the user's queries and document schema, respectively. For query matching, the lexical analyzer scans the incoming XML stream and the parser recognizes XML structures for retrieving every twig-query result from the XML stream. Moreover, STQM obtains query results without a post-phase for excluding false positives, which are common in many streaming query methods. Through the experimental results, we found that STQM matches the twig query efficiently and also has good scalability both in the queried data size and the branch degree of the twig query. The proposed method takes less execution time than that of a sequence-based approach, which is widely accepted as a proper solution to the XML stream query.     

A systems approach to organizational error

Abstract

The past two decades have seen a significant number of large-scale disasters in a wide range of hazardous, well-defended technologies. Despite their differences, the root causes of these accidents have been traced to latent failures and organizational errors arising in the upper echelons of the system in question. A model of the aetiology of these organizational accidents is outlined. The model describes two interrelated causal sequences: (a) an active failure pathway that originates in top-level decisions and proceeds via error-producing and violation-promoting conditions in the various workplaces to unsafe acts committed by those at the immediate human-system interface and (b) a latent failure pathway that runs directly from the organizational processes to deficiencies in the system's defences. The paper goes on to identify two sets of dependencies associated with latent failures and violations. Organizational errors increase the likelihood of operator error through the active failure pathway and, at the same time, enhance the possibility of adverse outcomes through defensive weaknesses. Violations have a narrower range of consequences. Non-compliance with safe operating procedures increases the likelihood of error by taking perpetrators into regions of operation in which neither the physical regime nor the hazards are well understood. Violations, by definition, also take perpetrators ‘closer to the edge,’ and thus increase the chance that subsequent errors will have damaging outcomes. The paper concludes by indicating two ways in which the model has been applied in industrial settings: (a) through the development of proactive measures for diagnosing and remedying organizational processes known to be implicated in accident causation, and (b) an accident investigation technique that guides investigators and analysts to the organizational root causes of past accidents.     

A general approach to classification problems

In the classification problem the value of the discrete random variable I must be estimated from that of a random variable X given a random sample (I₁, X₁),(I₂, X₂),…,(I_n, X_n). A classification rule giving such an estimate can be based on most nonparametric density estimators which are based on delta sequences. The rate at which the nonerror rate for this rule converges to the optimal is calculated. It is found to be $\text{O(n}{}^{\mathrm{<mtext>?1</mtext><mtext>(2d\; +\; 2)</mtext>}}\text{log}{}^{\mathrm{d}}\text{n}$, where d is the dimension of the variable X under mild assumptions on the densities. A procedure for choosing an associated parameter for small samples is given.     

A multichannel approach to fingerprint classification

Fingerprint classification provides an important indexing mechanism in a fingerprint database. An accurate and consistent classification can greatly reduce fingerprint matching time for a large database. We present a fingerprint classification algorithm which is able to achieve an accuracy better than previously reported in the literature. We classify fingerprints into five categories: whorl, right loop, left loop, arch, and tented arch. The algorithm uses a novel representation (FingerCode) and is based on a two-stage classifier to make a classification. It has been tested on 4000 images in the NIST-4 database. For the five-class problem, a classification accuracy of 90 percent is achieved (with a 1.8 percent rejection during the feature extraction phase). For the four-class problem (arch and tented arch combined into one class), we are able to achieve a classification accuracy of 94.8 percent (with 1.8 percent rejection). By incorporating a reject option at the classifier, the classification accuracy can be increased to 96 percent for the five-class classification task, and to 97.8 percent for the four-class classification task after a total of 32.5 percent of the images are rejected     

On the role of error productions in syntactic error correction

Error productions are presented as a means of augmenting syntactic error correctors. Such productions are able to simply and efficiently handle a wide variety of difficult error situations. Furthermore, they can be employed without changing (or even knowing) the structure of the error correction algorithm being used. Examples of the use of error productions with the programming language Pascal are presented. Implementation and test results are included.     

A syntactic approach to robot imitation learning using probabilistic activity grammars

This paper describes a syntactic approach to imitation learning that captures important task structures in the form of probabilistic activity grammars from a reasonably small number of samples under noisy conditions. We show that these learned grammars can be recursively applied to help recognize unforeseen, more complicated tasks that share underlying structures. The grammars enforce an observation to be consistent with the previously observed behaviors which can correct unexpected, out-of-context actions due to errors of the observer and/or demonstrator. To achieve this goal, our method (1) actively searches for frequently occurring action symbols that are subsets of input samples to uncover the hierarchical structure of the demonstration, and (2) considers the uncertainties of input symbols due to imperfect low-level detectors.We evaluate the proposed method using both synthetic data and two sets of real-world humanoid robot experiments. In our Towers of Hanoi experiment, the robot learns the important constraints of the puzzle after observing demonstrators solving it. In our Dance Imitation experiment, the robot learns 3 types of dances from human demonstrations. The results suggest that under reasonable amount of noise, our method is capable of capturing the reusable task structures and generalizing them to cope with recursions.     

A syntactic approach for handwritten mathematical formula recognition

Mathematical formulas are good examples of two-dimensional patterns as well as pictures or graphics. The use of syntactic methods is useful for interpreting such complex patterns. In this paper we propose a system for the interpretation of 2-D mathematic formulas based on a syntactic parser. This system is able to recognize a large class of 2-D mathematic formulas written on a graphic tablet. It starts the parsing by localization of the ``principal' operator in the formula and attempts to partition it into subexpressions which are similarly analyzed by looking for a starting character. The generalized parser used in the system has been developed in our group for continuous speech recognition and picture interpretation.     

A fuzzy approach to some classification problems

An appropriate algebraic structure was previously defined which can be regarded as a possible alternative to the theory of approximate reasoning, [A. Gisolfi, Fuzzy Sets Syst. 44 , 37–43 (1992)]. In this article we aim at extending the operations of the structure in order to cope with classification problems. the theoretical aspects are emphasized in order to give an adequate background to the possible applications. After defining the basic elements and the related operations, the structure is implemented by means of Prolog. Finally the relationship between the structure and the problems of classification is discussed in some detail. © 1993 John Wiley & Sons, Inc.     

A maximum likelihood approach to texture classification

A new approach to texture classification is described which is based on measurements of the spatial gray-level co-occurrence probability matrix. This approach can make use of assumed stochastic models for texture in imagery and is an approximation to the statistically optimum maximum likelihood classifier. The efficacy of the approach is demonstrated through experimental results obtained with real-world texture data.     

Applications of regularized least squares to pattern classification

We survey a number of recent results concerning the behaviour of algorithms for learning classifiers based on the solution of a regularized least-squares problem.