Presentation 2005/11/12
FPGA Implementation of Elastic Graph Matching for Face/Object Recognition and Its Performance Evaluation
Teppei NAKANO, Takashi MORIE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The elastic graph matching (EGM) is based on the dynamic link architecture (DLA) which is a conceptual neural network models. In the EGM, the same graph with plural nodes is defined on both input and memorized image planes, and the pixel on each node is used as an evaluation point. The graph is distorted to find the best trade-off between better matching in the feature space and less distortion of the evaluation point graph. Therefore, distortion-invariant recognition can be achieved. We have proposed the digital architecture of the EGM. In this paper, we present an FPGA implementation of the architecture and its performance evaluation results.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) face recognition / object recognition / elastic graph matching / digital image processor LSI
Paper # NC2005-67
Date of Issue

Conference Information
Committee NC
Conference Date 2005/11/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) FPGA Implementation of Elastic Graph Matching for Face/Object Recognition and Its Performance Evaluation
Sub Title (in English)
Keyword(1) face recognition
Keyword(2) object recognition
Keyword(3) elastic graph matching
Keyword(4) digital image processor LSI
1st Author's Name Teppei NAKANO
1st Author's Affiliation Kyushu Institute of Technology()
2nd Author's Name Takashi MORIE
2nd Author's Affiliation Kyushu Institute of Technology
Date 2005/11/12
Paper # NC2005-67
Volume (vol) vol.105
Number (no) 419
Page pp.pp.-
#Pages 6
Date of Issue