Patents by Inventor Xinchun Huang

Xinchun Huang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Method and system for recognizing a handwritten character
    Patent number: 8396295
    Abstract: The present invention discloses a method for recognizing a handwritten character, which includes the following steps of: obtaining a coarse classification template and a fine classification template; receiving a handwritten character input signal from a user, gathering a discrete coordinate sequence of trajectory points of the inputted character, and pre-processing the discrete coordinate sequence; extracting eigenvalues and calculating a multi-dimensional eigenvector of the inputted character; matching the inputted character with the coarse classification template to select a plurality of the most similar candidate character classes; and matching the eigen-transformed inputted character with sample centers of the candidate character classes selected from the fine classification template, and determining the most similar character classes among the candidate character classes. The present invention further discloses a system for recognizing a handwritten character.
    Type: Grant
    Filed: February 25, 2009
    Date of Patent: March 12, 2013
    Assignee: Guangdong Guobi Technology Co., Ltd
    Inventors: Jing-lian Gao, Xinchun Huang, Binghui Chen, Anjin Hu, Muyu Cai, Huaxing Lu, Zhipin Liu, Zhiai Wang, Fang Guo, Jingping Li, Honghui Wang, Chuntao Tan, Zhengwei Wu
  • METHOD AND SYSTEM FOR RECOGNIZING A HANDWRITTEN CHARACTER
    Publication number: 20110311141
    Abstract: The present invention discloses a method for recognizing a handwritten character, which includes the following steps of: obtaining a coarse classification template and a fine classification template; receiving a handwritten character input signal from a user, gathering a discrete coordinate sequence of trajectory points of the inputted character, and pre-processing the discrete coordinate sequence; extracting eigenvalues and calculating a multi-dimensional eigenvector of the inputted character; matching the inputted character with the coarse classification template to select a plurality of the most similar candidate character classes; and matching the eigen-transformed inputted character with sample centers of the candidate character classes selected from the fine classification template, and determining the most similar character classes among the candidate character classes. The present invention further discloses a system for recognizing a handwritten character.
    Type: Application
    Filed: February 25, 2009
    Publication date: December 22, 2011
    Applicant: GUANGDONG GUOBI TECHNOLOGY CO., LTD.
    Inventors: Jinglian Gao, Xinchun Huang, Binghui Chen, Anjin Hu, Muyu Cai, Huaxing Lu, Zhipin Liu, Zhiai Wang, Fang Guo, Jingping Li, Honghui Wang, Chuntao Tan, Zhengwei Wu