In this paper a methodology for feature selection for the handwritten digit string recognition is proposed. Its novelty lies in the use of a multi-objective genetic algorithm where sensitivity analysis and neural network are employed to allow the use of a representative database to evaluate tness and the use of a validation database to identify the subsets of selected features that provide a good generalization. Some advantages of this approach include the ability to accommodate multiple criteria such as number of features and accuracy of the classier, as well as the capacity to deal with huge databases in order to adequately represent the pattern recognition problem. Comprehensive experiments on the NIST SD19 demonstrate the feasibility of the proposed methodology.

This paper describes our prototype which can differentiate between cheques and

Feature selection for ensembles has shown to be an effective strategy for ensemble creation. In this paper we present an ensemble feature selection approach based on a hierarchical multi-objective genetic algorithm. The first level performs feature selection in order to generate a set of good classifiers while the second one combines them to provide a set of powerful ensembles. The proposed method is evaluated in the context of handwritten digit recognition, using three different feature sets and neural networks (MLP) as classifiers. Experiments conducted on NIST SD19 demonstrated the effectiveness of the proposed strategy.

A new paradigm, which models the relationships between handwriting and topic categories, in the context of medical forms, is presented. The ultimate goals are (i) the recognition of medical handwriting, and (ii) the use of such information for practical applications such as a medical form search engine. Medical forms have diverse, complex and large lexicons consisting of English, Medical and Pharmacology corpus. Our technique shows that a few recognized characters, returned by handwriting recognition, can be used to construct a linguistic model capable of representing a medical topic

Without an independence assumption, combining multiple classifiers deals with a high order probability distribution composed of classifiers and a class label. Storing and estimating the high order probability distribution is exponentially complex and unmanageable in theoretical analysis, so we rely on an approximation scheme using the dependency. In this paper, as an extension of the second-order dependency approach, the probability distribution is optimally approximated by the third-order dependency and multiple classifiers are combined. The proposed method is evaluated on the recognition of unconstrained handwritten numerals from Concordia University and the University of California, Irvine. Experimental results support the proposed method as a promising approach.

In this thesis, an HMM-MLP hybrid system for segmenting and recognizing unconstrained handwritten dates written on Brazilian bank cheques is presented. The system evolves by dealing with many sources of variability, such as heterogeneous data types and styles, variations present in the date field, and difficult cases of segmentation that make the recognizer task particular hard to do. The system takes an HMM-based...

In this paper, we present a novel approach to accurately detect text regions including shop name in signboard images with complex background for mobile system applications. The proposed method is based on the combination of text detection using edge profile and region segmentation using fuzzy c-means method. In the first step, we perform an elaborate canny edge operator to extract all possible object edges. Then, edge profile analysis with vertical and horizontal direction is performed on these edge pixels to detect potential text region existing shop name in a signboard. The edge profile and geometrical characteristics of each object contour are carefully examined to construct candidate text regions and classify the main text region from background. Finally, the fuzzy c-means algorithm is performed to segment and detected binarize text region. Experimental results show that our proposed method is robust in text detection with respect to different character size and color and can provide reliable text binarization result.

This paper discusses a two-level approach to optimize classification systems with multi-objective genetic algorithms. The first level creates a set of representations through feature extraction, which is used to train a classifier set. At this point, the most performing classifier can be selected for a single classifier system, or an ensemble of classifiers can be optimized for improved accuracy. Two zoning strategies for feature extraction are discussed and compared using global validation to select optimized solutions. Experiments conducted with isolated handwritten digits and uppercase letters demonstrate the effectiveness of this approach, which encourages further research in this direction.

The performance of multiple classifier systems varies with the performance of component classifiers as well as the method of combination. In this paper, informationtheoretic methods are proposed for constructing multiple classifier systems, provided that the number of component classifiers is constrained in advance. These proposed methods are applied to a classifier pool and examine the possible classifier sets by the selected information-theoretic criteria. One of them is then selected as the candidate and is evaluated together with the other multiple classifier systems on the recognition of unconstrained handwritten numerals from Concordia University and the University of California, Irvine. Experimental results support the approach.

In this paper we discuss a strategy to create ensemble of classifiers based on unsupervised features selection. It takes into account a hierarchical multi-objective genetic algorithm that generates a set of classifiers by performing feature selection and then combines them to provide a set of powerful ensembles. The proposed method is evaluated in the context of handwritten month word recognition, using three different feature sets and Hidden Markov Models as classifiers. Comprehensive experiments demonstrates the effectiveness of the proposed strategy.