Handwritten Chinese Text Recognition by Integrating Multiple Contexts

NANO SCIENTIFIC RESEARCH CENTRE PVT.LTD.,  AMEERPET, HYD

WWW.NSRCNANO.COM, 09640648777, 09652926926

JAVA PROJECTS LIST--2013

JAVA 2013 IEEE PAPERS

Handwritten Chinese Text Recognition by Integrating Multiple Contexts

Abstract

            This paper presents an effective approach for the offline recognition of unconstrained handwritten Chinese texts. Under the general integrated segmentation-and-recognition framework with character oversegmentation, we investigate three important issues: candidate path evaluation, path search, and parameter estimation. For path evaluation, we combine multiple contexts (character recognition scores, geometric and linguistic contexts) from the Bayesian decision view, and convert the classifier outputs to posterior probabilities via confidence transformation. In path search, we use a refined beam search algorithm to improve the search efficiency and, meanwhile, use a candidate character augmentation strategy to improve the recognition accuracy. The combining weights of the path evaluation function are optimized by supervised learning using a Maximum Character Accuracy criterion. We evaluated the recognition performance on a Chinese handwriting database CASIA-HWDB, which contains nearly four million character samples of 7,356 classes and 5,091 pages of unconstrained handwritten texts. The experimental results show that confidence transformation and combining multiple contexts improve the text line recognition performance significantly. On a test set of 1,015 handwritten pages, the proposed approach achieved character-level accurate rate of 90.75 percent and correct rate of 91.39 percent, which are superior by far to the best results reported in the literature.

Existing system

            In the context of handwritten text (character string1) recognition, many works have contributed to the related issues of oversegmentation, character classification, confidence transformation, language model, geometric model, path evaluation and search, and parameter estimation. For oversegmentation, connected component analysis has been widely adopted, but the splitting of connected (touching) characters has been a concern. After generating candidate character patterns by combining consecutive primitive segments, each candidate pattern is classified using a classifier to assign similarity/dissimilarity scores to some character classes. Character classification involves character normalization, feature extraction, and classifier design. For classification of Chinese characters with large number of classes, the most popularly used classifiers are the modified quadratic Discriminant function (MQDF) and the nearest prototype classifier (NPC). The MQDF provides higher accuracy than the NPC but suffers from high expenses of storage and computation.

Proposed System

            This system focuses on the recognition of text lines, which are assumed to have been segmented externally. For the convenience of academic research and benchmarking, the text lines in our database have been segmented and annotated at character level. First, the input text line image is oversegmented into a sequence of primitive segments using the connected component-based method. Consecutive primitive segments are combined to generate candidate character patterns, forming a segmentation candidate lattice. After that, each candidate pattern is classified to assign a number of candidate character classes, and all the candidate patterns in a candidate segmentation path generate a character candidate lattice.

Software Requirement Specification

Software Specification

Operating System       :           Windows XP

Technology                 :           JAVA 1.6

Minimum Hardware Specification

Processor                     :           Pentium IV

RAM                           :           512 MB

Hard Disk                   :           80GB