A pattern recognition apparatus that comprises an input section, a feature extraction module, a feature transform module, a recognition section that includes a recognition dictionary, and a categorizer. The input section receives input patterns that include a pattern belonging to one of plural categories constituting a category set. The feature extraction module that expresses features of the pattern as a feature vector. The feature transform module uses transform vector matrices to transform at least part of the feature vector to generate an at least partially transformed feature vector corresponding to each of the categories. The transform vector matrices include a transform vector matrix generated in response to a rival pattern set composed of rival patterns misrecognized as belonging to plural ones of the categories. The plural ones of the categories constitute a category subset. The at least partially transformed feature vector is common to the ones of the categories constituting the category subset. The recognition dictionary stores both matching information and transformed matching information for each of the categories. The first transformed matching information has been transformed using the transform vector matrices. The recognition section generates at least one difference value for each of the categories by performing a matching operation between the matching information and the first transformed matching information on one hand, and at least one matching vector derived at least from the at least partially transformed feature vector corresponding to each of the categories on the other hand. The categorizer identifies the category to which the pattern belongs in response to the at least one difference value.
A method for processing digitized speech signals by analyzing redundant features to provide more robust voice recognition. A primary transformation is applied to a source speech signal to extract primary features therefrom. Each of at least one secondary transformation is applied to the source speech signal or extracted primary features to yield at least one set of secondary features statistically dependant on the primary features. At least one predetermined function is then applied to combine the primary features with the secondary features. A recognition answer is generated by pattern matching this combination against predetermined voice recognition templates.
A personal identity authentication process and system use a class specific linear discriminant transformation to test authenticity of a probe face image. A `client acceptance` approach, an `imposter rejection` approach and a `fused` approach are described.
In order to improve pattern recognition, various kinds of transformations are performed on an input object. One or more recognition algorithms are then performed on the input object transforms in addition to the input object itself. By performing recognition algorithms on an input object and its transforms, a more comprehensive set of recognition results are generated. A final recognition decision is based upon an input object and its transforms by aggregating the recognition results.
A method and apparatus for generating a degraded dictionary automatically is presented in this invention. Herein, a degraded pattern generating means generates a plurality of degraded patterns from an original character image, based on a plurality of degradation parameters. A degraded dictionary generating means generates a plurality of degraded dictionaries corresponding to the plurality of degradation parameters, based on the plurality of degradation patterns. Finally, a dictionary matching means selects one of the plurality of dictionaries which matches the degradation level of a test sample set best, as the final degraded dictionary. In this invention, various degraded patterns can be generated by means of simple scaling and blurring process for establishing degraded dictionaries. Therefore, the invention can be implemented simply and easily. The method and apparatus of the invention can not only be used in character recognition field, but also can be used in other fields such as speech recognition and face recognition.
A method and apparatus for generating a degraded character image at various levels of degradation automatically is presented in this invention. The method comprises rendering the character image on a scene plane; translating and rotating the scene plane according to various parameters; determining a projection region of the character image on an image plane according to various parameters; generating a pixel region mask; and generating a final degraded image by super-sampling. Thus various degraded character images are generated on various conditions of degradation. The generated synthetic characters can be used for performance evaluation and training data augmentation in optical character recognition (OCR).
