Home   About   Membership   Organization   OAPS Papers   Annual Meeting   Contact 

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/166

Title: Online graphics recognition system
Authors: Yu, Yajie (余亞杰)
Department of Computer Science
Keywords: Pattern recognition systems
Image processing
Online data processing
Issue Date: 2004
Abstract: Online graphics recognition system (OGRS) can provide a natural, convenient,and efficient way for users to input rigid and regular shapes or graphic objects, suchas triangles, rectangles, ellipses, straight lines, arrowheads and so on. All what usershave to do is quickly drawing their sketchy shapes in single or multiple strokes, andthe system will automatically convert the strokes into the user-intended rigid shapebased on the shape similarity and the time constraint of the sketchy line. Two mainstages are involved in the system, one is stroke segmentation and the other iscomposite objects matching. For the stroke segmentation stage, Liu Yin in ourresearch group proposed a new dynamic programming (DP) framework, whichemployed a penalty function to eliminate the requirement of knowing the number ortemplates of the segments in traditional dynamic programming frameworks. In mostsituations, a single stroke contains less than 100 points and experiment shows that thenew DP framework can give an accurate result and respond in real-time. However, inextreme cases, when strokes containing more than 200 points are processed, theresponse time is relatively slow. In order to further improve its performance, Iproposed another DP framework, which was also based on penalty function.Experiments show that my improved DP framework is efficient and accurate. Both ofthe two penalty-based DP frameworks have predefined parameters and I developed aParameter-Optimization system to find the best parameter set for them in theparameter space. For composite objects matching, I conducted a complete experimentto test the accuracy of an algorithm proposed in [25] and in addition, developed an algorithm based on linear programming to match two acyclic graphs in polynomialtime. An experiment was also conducted to show the correctness of the algorithm.
Description: Nominated as OAPS (Outstanding Academic Papers by Students) paper by Department in 2006-07.
Liu W Y. First Reader: Lee K C. Second Reader: Chan Y K
URI: http://hdl.handle.net/123456789/166
Source URI: http://hdl.handle.net/2031/428
Source Fulltext: http://dspace.cityu.edu.hk/bitstream/2031/428/1/fulltext.html
Appears in Collections:OAPS - Dept. of Computer Science 电脑科学系

Files in This Item:

There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback