Related Experiment Video
Updated: May 29, 2026
05:58
Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment
Published on: March 11, 2021
Research on machine recognition of handprinted characters
1Electrotechnical Laboratory, Sakura-mura, Niihari-gun, Ibaraki 305, Japan; Nippon-Schlumberger, Schinjiku-Daiichiseimei, Tokyo, Japan.
IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 27, 2011
Summary
Researchers in Japan are achieving encouraging results in machine recognition of handprinted Chinese characters. Feature matching is emerging as a key method for this task, with ongoing comparative studies and discussions on future directions.
Area of Science:
- Computer Science
- Artificial Intelligence
- Pattern Recognition
Background:
- Machine recognition of handprinted Chinese characters is an active research area in Japan.
- Encouraging practical and academic results have been reported.
Purpose of the Study:
- To systematically describe and analyze machine recognition methods for handprinted Chinese characters.
- To provide a qualitative evaluation of different recognition methods through comparative studies.
Main Methods:
- Feature matching is identified as a mainstream approach.
- Experiments were conducted using the ETL8 database (881 Kanji, 71 hirakana, 160 variations).
- Comparative analysis of feature and structural matching techniques.
Main Results:
- Encouraging recognition rates have been achieved.
- A qualitative evaluation of various methods is presented.
- Merits and demerits of feature and structural matching are discussed.
Conclusions:
- Feature matching is a promising direction for handprinted Chinese character recognition.
- Comparative studies aid in understanding method performance.
- Future research directions are outlined based on current findings.
