Hybrid SVM-HMM based recognition algorithm for pen-based tutoring system

Zhenming Yuan; Hong Pan

doi:10.1117/12.774813

14 November 2007 Hybrid SVM-HMM based recognition algorithm for pen-based tutoring system

Zhenming Yuan, Hong Pan

Proceedings Volume 6790, MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications; 679058 (2007) https://doi.org/10.1117/12.774813
Event: International Symposium on Multispectral Image Processing and Pattern Recognition, 2007, Wuhan, China

Abstract

Pen-based computing takes advantage of human skill with the pen, which is more than a substitute for the mouse. A hybrid SVM-HMM based recognition algorithm is presented for pen-based single stroke diagram. The algorithm includes five steps: sampling and pre-processing, segmentation, formal feature computing, SVM based feature classification, and HMM based symbol recognition. The formal feature of a stroke is composed of five static features and one dynamic feature. A group of one-to-one combinations of binary SVMs are used as feature classifiers to produce fixed length feature vectors, each of which is produced by the probability output with Sigmoid function and act as the posterior probability of observation of HMM. Finally HMMs are employed as final recognizer to recognize the unknown stroke. Based on this algorithm, a tutoring system is designed to identify the sketches of the flowchart diagrams. Experiment results show the hybrid algorithm has a good learning and recognition ability, which is benefited from combining the SVM's classification ability of static properties with the HMM's recognition ability of dynamic properties.

Citation Download Citation

Zhenming Yuan and Hong Pan "Hybrid SVM-HMM based recognition algorithm for pen-based tutoring system", Proc. SPIE 6790, MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications, 679058 (14 November 2007); https://doi.org/10.1117/12.774813

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Detection and tracking algorithms

Binary data

Independent component analysis

Computer programming

Connectors

Image segmentation

Visualization

Show All Keywords

Keywords/Phrases

Search In:

Publication Years