实验室两项研究成果获微软亚洲研究院高校关系部好评

2008-4-16

实验室在微软亚洲研究院高校合作基金资助下完成的两项科研成果获得微软亚洲研究院好评：

Achievement 1. Building Compact MQDF Classifier for Large Character Set Recognition by Subspace Distribution Sharing

Achievement 2. Calligraphic Beautification of Handwritten Chinese Characters

龙腾及朱星华同学分别是上述两个成果的主要完成者之一。

我们参与的这个主题研究项目的页面是：
http://research.microsoft.com/ur/asia/research/MCEdefault_ASIA.aspx

微软亚洲研究院高校关系部在其网页上对上述成果进行了示范宣传,网址是：

http://research.microsoft.com/ur/asia/research/MCE_lianwen.aspx

原文如下：

Online Cursive Script Handwritten Chinese Character/Word Recognition
Lianwen JIN, HCII Lab of Souch China University of Technology

Under the sponsored research fund by MSRA, a research team from the Human-Computer Intelligent Interface Lab (http://www.hcii-lab.net) of Souch China University of Technology (SCUT) has explored several new modular technologies for high performance handwritten Chinese character recognition and calligraphic beautification of handwritten Chinese characters. Two significant achievements have been accomplished by SCUT team.

Achievement 1. Building Compact MQDF Classifier for Large Character Set Recognition by Subspace Distribution Sharing

The SCUT research team has proposed a method for building compact and high accuracy MQDF classifier. Quadratic classifier with modified quadratic discriminant function (MQDF) has been successfully applied to recognition of handwritten characters to achieve very high performance. By using a small set of prototypes clustered from the original subspaces to represent the uncompressed sub-vectors, the storage of the MQDF parameters is greatly compressed. By seeking for the optimal tradeoff curves between parameter size and recognition accuracy, some sets of parameter settings are discovered to form the optimal compact dictionary for MQDF parameters. The fast recognition speed (1.8ms/char for PC and 64ms/char for Pocket PC) and compact dictionary size (2.06MB) make the high accuracy MQDF classifier become practical for memory limited hand-held devices such as PDAs, mobile phones and Pocket PCs.

Achievement 2. Calligraphic Beautification of Handwritten Chinese Characters

SCUT team focus on a patternized approach in attempting beautification of handwritten Chinese characters through feature corresponding and fusing. Simulated Analogous Reasoning Process is employed in the proposed system to adopt some certain styles onto users’ input characters. The system is made applicable by accommodating to common variations in handwritings. The verification-based stroke corresponding algorithm allows inputs that contain connected or falsely-separated strokes. Also, the tri-unit stroke model which handles the connection parts between successive strokes ensures a natural simulation and beautification for the connection parts expressly, and thus better preserves reflections of user individualities. The proposed system is proved to be effective and feasible on transfiguring handwritings and preserving originality of users’ in series of experiments.