Intelligent word recognition

About: Intelligent word recognition is a research topic. Over the lifetime, 2480 publications have been published within this topic receiving 45813 citations.

Method and system for the spotting of arbitrary words in handwritten documents

Abstract: A method and system for the spotting of keywords in a handwritten document, the method comprising the steps of inputting an image of the handwritten document, performing word segmentation on the image to obtain segmented words, performing word matching, and outputting the spotted keywords. The word matching itself consisting in the sub-steps of performing character segmentation on the segmented words, performing character recognition on the segmented characters, performing distance computations on the recognized characters using a Generalized Hidden Markov Model with ergodic topology to identify words based on character models and performing non-keyword rejection using a classifier based on a combination of Gaussian Mixture Models, Hidden Markov Models and Support Vector Machines.

System and method of performing speech recognition based on a user identifier

Abstract: Speech recognition models are dynamically re-configurable based on user information, application information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. Word recognition lattices are generated for each data field of an application and dynamically concatenated into a single word recognition lattice. A language model is applied to the concatenated word recognition lattice to determine the relationships between the word recognition lattices and repeated until the generated word recognition lattices are acceptable or differ from a predetermined value only by a threshold amount. These techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.

Handwritten Chinese character segmentation using a two-stage approach

Abstract: Correct segmentation of handwritten Chinese characters is crucial to the successful recognition. However, because of the many difficulties involved, little work has been done in this area. In this paper, a two-stage approach is addressed to segment unconstrained handwritten Chinese character strings. A string is first coarsely segmented according to the background skeleton and vertical projection after a proper image preprocessing. At the fine segmentation stage that follows, the strokes that may contain segmentation points are first identified. The feature points are then extracted from candidate strokes and taken as segmentation point candidates through each of which a segmentation path may be formed. Geometric features are extracted and fuzzy decision rules learned from examples are used to evaluate the segmentation paths. By using this two-stage segmentation approach, we can achieve both good performance and efficiency in segmenting unconstrained handwritten Chinese characters.

A Study of Nonlinear Shape Normalization for Online Handwritten Chinese Character Recognition: Dot Density vs. Line Density Equalization

Abstract: Nonlinear shape normalization (NSN) approaches based on line density equalization have been the most popular choice for both offline and online handwritten Chinese character recognition (HCCR). However, in a recent study of using 8-directional features for online HCCR, we discovered that an NSN approach based on dot density equalization achieved a much better performance than that of an NSN approach based on line density equalization. In this paper, we present the details of the NSN approaches we studied for online HCCR, and report the comparative experimental results using an in-house developed Chinese handwriting corpus as well as the popular Nakayosi and Kuchibue Japanese character databases. We also present an improved NSN approach based on the equalization of dot densities derived from blurred character image that can be used for offline HCCR.

Word search apparatus, word search method, and recording medium

Abstract: There are provided a word search apparatus, a word search method, and a computer program product. A words dictionary, and a character recognition dictionary for storing coordinate data of a standard character pattern of a handwritten character and a character are used to thereby search for, from the words dictionary, a word including a character corresponding to one or a plurality of character patterns extracted by performing a pattern matching. Only a character string corresponding to one or a plurality of character patterns is extracted from a search result of the words dictionary to generate a part of character string. A selection of one part of character string among the generated parts of character strings is received, and only a word including the selected part of character string is extracted from the search result based on the words dictionary, so that the extracted word is displayed.