Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/43928
Title: Online handwriting recognition for chemical expressions
Authors: Zhang, Zonglu.
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
Issue Date: 2011
Abstract: Currently, handwriting-based mobile phones, PDAs and computers are very common. Handwriting input provides users an efficient and natural input method. Many handwriting recognition techniques have been proposed in recent years. However, recognizing chemical expressions is still a challenging work. Therefore, this project aims to design an online handwritten chemical expression recognition system. There are two processes for chemical expression recognition. The first process is to identify individual chemical symbols, namely Chemical Symbol Recognition. In order to improve recognition accuracy, this project adopts a Support Vector Machine – Elastic Matching hybrid approach. The second process is to analyze the structure of the chemical expression, namely Chemical Structural Analysis. There are mainly two steps in Structural Analysis, which are Related Symbol Identification and Symbol Relationship Identification. The recognized symbol will be stored in a Chemical Expression Tree based on the result of Structural Analysis. The Chemical Expression Tree can be converted into an ASCII string for displaying the recognition result to the user. This report discusses the proposed approaches for Chemical Symbol Recognition and Chemical Structural Analysis. The performance of the proposed approaches is also evaluated and given. In addition, the future directions for further work are also discussed.
URI: http://hdl.handle.net/10356/43928
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
SCE10-0330.pdf
  Restricted Access
1.84 MBAdobe PDFView/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.