Hirobumi Nishida
Hirobumi Nishida is a researcher in structural pattern recognition whose work links character recognition with shape and image retrieval. In 1991, he filed a Ricoh patent for classifying character line patterns, and in 1995 he served as an associate professor at the University of Aizu and chair of IAPR Technical Committee 8 while helping develop a fast Ricoh OCR package. He published a structural model of curve deformation in 1996 and a paper on partly visible shape retrieval in 2000. IAPR named him a Fellow in 2000 for contributions to structural pattern recognition and service; his 2002 Pattern Recognition paper examined structural feature indexing for partially visible shapes. In 2007, he organized an SICE session on image-based measurement as a Ricoh representative, and a Ricoh patent application he filed in 2009 for predicting optimal processing was granted in 2012.
Timeline
-
1991–1994
A character-line patent
In 1991, he files a Ricoh patent application for classifying character line patterns in a dictionary; it is granted in 1994.
-
1995
Associate professor and IAPR chair
In 1995, he is an associate professor in the University of Aizu's Image Processing Laboratory and chair of IAPR Technical Committee 8 on Application in Industry. His earlier work at Ricoh's Research and Development Center helps an OCR group develop a fast package called “filtering in character recognition.”
-
1996
Modeling curve deformation
In 1996, he publishes “A Structural Model of Curve Deformation by Discontinuous Transformations” in CVGIP: Graphical Model and Image Processing.
-
2000
Partly visible shape retrieval
In 2000, he publishes “Retrieval of Partially Visible Shapes through Structural Feature Indexing” in the joint proceedings of IAPR's SSPR and SPR workshops.
-
2000
An IAPR fellowship
IAPR names him a Fellow in 2000 for contributions to structural pattern recognition and service to the association.
-
2002
Structural feature indexing
In 2002, he publishes “Structural feature indexing for retrieval of partially visible shapes” in Pattern Recognition, pages 55–67.
-
2007
Image-based measurement session
At SICE's 2007 Annual Conference, he co-organizes an Advanced Image-Based Measurement session under the Technical Committee on Pattern Measurement, with Ricoh Co., Ltd. as his affiliation. The session covers noncontact, nondestructive extraction of three-dimensional or spatial information from images.
-
2009–2012
Predicting optimal data processing
In 2009, he files a Ricoh patent for predicting optimal processing from input features and a user's processing history; the patent is granted in 2012.
Sources and credits
Photo: We have no published photo of Hirobumi Nishida yet. Send us a free-licensed photo.
- US5317649A - Method for classifying line patterns of characters in dictionary - Google Patents
- Image Processing Laboratory
- CVGIP: Graphical Model and Image Processing, Volume 58
- Hirobumi Nishida
- Chronological List of IAPR Fellows
- Pattern Recognition, Volume 35
- SICE Annual Conference 2007: List of Organized Sessions
- US8321367B2 - Data processing apparatus, method, and computer program product for user objective prediction - Google Patents