Title :
A handwritten character recognition system using directional element feature and asymmetric Mahalanobis distance
Author :
Kato, Nei ; Suzuki, Masato ; Omachi, Shin Ichiro ; Aso, Hirotomo ; Nemoto, Yoshiaki
Author_Institution :
Graduate Sch. of Inf. Sci., Tohoku Univ., Sendai, Japan
fDate :
3/1/1999 12:00:00 AM
Abstract :
This paper presents a precise system for handwritten Chinese and Japanese character recognition. Before extracting directional element feature (DEF) from each character image, transformation based on partial inclination detection (TPID) is used to reduce undesired effects of degraded images. In the recognition process, city block distance with deviation (CBDD) and asymmetric Mahalanobis distance (AMD) are proposed for rough classification and fine classification. With this recognition system, the experimental result of the database ETL9B reaches to 99.42%
Keywords :
handwritten character recognition; image classification; AMD; CBDD; DEF; ETL9B database; TPID; asymmetric Mahalanobis distance; city block distance; degraded images; deviation; directional element feature; directional element feature extraction; fine classification; handwritten Chinese character recognition; handwritten Japanese character recognition; image transformation; partial inclination detection; rough classification; Character recognition; Cities and towns; Degradation; Feature extraction; Handwriting recognition; Image converters; Image recognition; Nonlinear distortion; Pattern matching; Pattern recognition;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
