M
Misako Suwa
Researcher at Fujitsu
Publications - 50
Citations - 497
Misako Suwa is an academic researcher from Fujitsu. The author has contributed to research in topics: Character (mathematics) & Intelligent word recognition. The author has an hindex of 11, co-authored 50 publications receiving 494 citations.
Papers
More filters
Proceedings ArticleDOI
Separation of overlapped color planes for document images
TL;DR: This paper proposed a simple but effective linear projection based method for separating overlapped color planes, which assumes that the number of foreground colors is low, and overlapped areas contain mixed colors instead of opaque covering.
Patent
Image processing program, image processing method, and image processor
TL;DR: In this article, a storage part for storing standard image information as a standard shape, and a control part for correcting the deviation of the shape of an input image is provided for easily acquiring a correction image acquired by precisely correcting the position deviation of input image information.
Patent
Shaded character recovery device and method thereof, shaded character recognition device and method thereof
Lanlan Chang,Sun Jun,Misako Suwa,Akihiro Minagawa,Hao Yu,Satoshi Naoi,Etsunobu Hotta,Yuan He +7 more
TL;DR: A shaded character recovery method is proposed in this paper, which consists of calculating the size of an image block according to the stroke width of target characters comprising shaded characters to be recovered, marking a source area in a target area comprising an input image of the target characters, generating reference image blocks aiming at the pixels in the source area, wherein each pixel in the image is used as the center of a reference image block, and extracting the boundary of the foreground of the input image to be filled.
Patent
Apparatus and method for recognizing character string
Hongliang Bai,Danian Zheng,Son Shun,Misako Suwa,Hiroaki Takebe,Yoshinobu Hotta,Hao Yu,Satoshi Naoi +7 more
TL;DR: In this paper, a method for recognizing a character string consisting of various types of characters is proposed, which consists of dividing a character image into a plurality of segments, performing OCR recognition on the plurality of segmented characters, acquiring statistical information of candidate characters and/or statistical information formed from the candidate characters, and merging the statistical information with reliability of OCR Recognition on candidate characters to decide a candidate character string.
Proceedings ArticleDOI
Accuracy improvement for handwritten Japanese word recognition by combination of character and word recognizer
TL;DR: Experimental results show that the combination approach can achieve better recognition performance than the other two methods.