Rotated Kanji Character Recognition
نویسندگان
چکیده
We proposed a rotated character recognition method using eigen-subspace characterized multiple projection and simple projection. Then the multiple projections showed a higher accuracy at low dimensions than a simple projection for alphanumeric 62 categories. This time, we applied it for the first class of Japanese Industrial Standard (JIS) Kanji set which includes 2,965 categories. As the result, very high recognition accuracy over 99.8% was achieved especially by the multiple projections of the input rotated images.
منابع مشابه
Evaluation of the SVM based Multi-Fonts Kanji Character Recognition Method for Early-Modern Japanese Printed Books
The national diet library in Japan provides a web based digital archive for early-modern printed books by image. To make better use of the digital archive, the book images should be converted to text data. In this paper, we evaluate the SVM based multi-fonts Kanji character recognition method for early-modern Japanese printed books. Using several sets of Kanji characters clipped from different ...
متن کاملSubstroke Approach to HMM-Based On-line Kanji Handwriting Recognition
A new method is proposed for on-line handwriting recognition of Kanji characters. The method employs substroke HMMs as minimum units to constitute Japanese Kanji characters and utilizes the direction of pen motion. The main motivation is to fully utilize the continuous speech recognition algorithm by relating sentence speech to Kanji character , phonemes to substrokes, and grammar to Kanji stru...
متن کاملNeural Networks that Learn to Discriminate Similar Kanji Characters
A neural network is applied to the problem of recognizing Kanji characters. Using a b a c k propagation network learning algorithm. a threelayered. feed-forward network is trained to recognize similar handwritten Kanji characters. In addition. two new methods are utilized to make training effective. The recognition accuracy was higher than that of conventional methods. An analysis of connection...
متن کاملModeling of Pen-Coordinate Information in SCPR-based HMM for On-line Recognition of Handwritten Japanese Characters
This paper describes stochastic modeling of pencoordinate information in HMMs with structured character pattern representation (SCPR) for on-line recognition of handwritten Japanese characters. SCPR allows HMMs for Kanji character patterns to share common subpatterns. Although SCPR-based HMMs have been successfully applied to Kanji character recognition, the pen-coordinate feature has not been ...
متن کاملSurvey of Pattern Recognition Approaches in Japanese Character Recognition
Optical Character Recognition (OCR) in Japanese, both handwritten and printed, is difficult to perform, owing to several reasons. Firstly, the Japanese language is comprised of over 3000 characters which can be classified as syllabic characters, or Kana, and ideographic characters, called Kanji. Secondly, Japanese text does not have delimiters like spaces, separating different words. Thirdly, s...
متن کامل