This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Tsinghua University. Active years: 2007.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2007
Title: Innovations in Character Recognition by Changsong Liu
Introduction
Changsong Liu is a prominent inventor based in Beijing, China. He has made significant contributions to the field of character recognition technology. His innovative approach utilizes advanced filtering techniques to enhance the accuracy and efficiency of character recognition systems.
Latest Patents
Changsong Liu holds a patent for a "Method for character recognition based on Gabor filters." This method employs the Gabor filter's joint spatial/spatial-frequency localization to efficiently extract local structural features of characters. By focusing on stroke direction information, the method improves the system's ability to resist noise, background variations, brightness changes, and character deformation. The design incorporates a corrected Sigmoid function for adaptive processing of stroke direction information, ensuring optimal recognition performance. Additionally, a Gaussian filter array is utilized to enhance the discrimination ability of the extracted features.
Career Highlights
Changsong Liu is affiliated with Tsinghua University, where he continues to advance research in character recognition technologies. His work has garnered attention for its practical applications in various fields, including document processing and automated data entry.
Collaborations
Changsong Liu has collaborated with notable colleagues such as Liangrui Peng and Xuewen Wang. Their combined expertise has contributed to the development of innovative solutions in the realm of character recognition.
Conclusion
Changsong Liu's contributions to character recognition through his patented methods demonstrate the potential for innovation in this field. His work not only enhances recognition accuracy but also addresses challenges posed by varying image conditions.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
