Figure Posture Recognition Using Texture Information

Tayayuki NAKATA; Yue BAO; Naofumi FUJIWARA

doi:10.11371/aiieej.32.0.25.0

Proceedings of the 32nd Annual Conference of the Institute of Image Electronics Engineers of Japan 2004

Session ID : 32-13

DOI https://doi.org/10.11371/aiieej.32.0.25.0

Conference information

Host: The Institute of Image Electronics Engineers of Japan

General Session

Figure Posture Recognition Using Texture Information

*Tayayuki NAKATA, Yue BAO, Naofumi FUJIWARA

Author information

Keywords: Log-Polar transform , Affine invariant recognition , Pattern matching , Computer vision

CONFERENCE PROCEEDINGS RESTRICTED ACCESS

Details

Abstract

Log-Polar transform is one of the models of human vision system, and is especially used well in face recognition. But, it is the detection method only for scaling and rotation and is mainly used in 2D environment in image-processing field. In this paper, a new Log-Polar transform algorithm is proposed that detects not only scaling and rotation of a 2D image but an affine transform parameter. By using this algorithm, it is possible to recognize a 2D image that rotates within 3D environment.

Corresponding author

Register with J-STAGE for free!