A modified direction feature for cursive character recognition
Authors:
- M. BlumensteinMBM. Blumenstein
- X.Y. LiuXLX.Y. Liu
- B. VermaBVB. Verma
- 1 January 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4, 2983-2987
Abstract
This paper describes a neural network-based technique for cursive character recognition applicable to segmentation-based word recognition systems. The proposed research builds on a novel feature extraction technique that extracts direction information from the structure of character contours. This principal is extended so that the direction information is integrated with a technique for detecting transitions between background and foreground pixels in the character image. The proposed technique is compared with the standard direction feature extraction technique, providing promising results using segmented characters from the CEDAR benchmark database.Griffith Sciences, School of Information and Communication TechnologyFull TexKeywords
This publication has 8 references indexed in Scilit:
- Combining neural gas and learning vector quantization for cursive character recognitionNeurocomputing, 2003
- Machine and human recognition of segmented characters from handwritten words2002
- Cursive digit and character recognition in CEDAR database2002
- A new segmentation algorithm for handwritten word recognition1999
- Neural-based solutions for the segmentation and recognition of difficult handwritten words from a benchmark database1999
- Handwritten word recognition with character and inter-character neural networksIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 1997
- Neural-network classifiers for recognizing totally unconstrained handwritten numeralsIEEE Transactions on Neural Networks, 1997
- A database for handwritten text recognition researchIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994