New Fourier descriptor based on areas (AFD) and its applications in object recognition

Zhongfei Zhang; Renjie Jing; Weikang Gu

New Fourier descriptor based on areas (AFD) and its applications in object recognition

Zhongfei Zhang
, Renjie Jing
, Weikang Gu

School of Computing

Binghamton University

Zhejiang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

The area Fourier descriptor (AFD), based on areas of the object's silhouettes, is presented, and the concept of multidimensional vector space is introduced. To speed up search procedures in the model base for matching, the authors apply a novel coding method to generate a curve-corner code (CCC) which can be used for any planar silhouettes. Taking the coordinate vectors in the multidimensional space introduced by the AFD as the main features, they developed an algorithm which can not only tolerate any translations, rotations, scale changing, and affine distortion, but also requires less computation. Because of the limited terms of the AFD, the algorithm is insensitive to truncation errors and can also be extended to recognize 3-D general cylindrical objects.

Original language	English
Title of host publication	Proc 1988 IEEE Int Conf Syst Man Cybern
Editors	Anon
Pages	502-505
Number of pages	4
State	Published - 1988
Event	Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics - Beijing/Shenyang, China Duration: Aug 8 1988 → Aug 12 1988

Publication series

Name	Proc 1988 IEEE Int Conf Syst Man Cybern

Conference

Conference	Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics
City	Beijing/Shenyang, China
Period	08/8/88 → 08/12/88

Cite this

@inproceedings{0669ba9ef0a3429999b2b4536c4cfd8d,

title = "New Fourier descriptor based on areas (AFD) and its applications in object recognition",

abstract = "The area Fourier descriptor (AFD), based on areas of the object's silhouettes, is presented, and the concept of multidimensional vector space is introduced. To speed up search procedures in the model base for matching, the authors apply a novel coding method to generate a curve-corner code (CCC) which can be used for any planar silhouettes. Taking the coordinate vectors in the multidimensional space introduced by the AFD as the main features, they developed an algorithm which can not only tolerate any translations, rotations, scale changing, and affine distortion, but also requires less computation. Because of the limited terms of the AFD, the algorithm is insensitive to truncation errors and can also be extended to recognize 3-D general cylindrical objects.",

author = "Zhongfei Zhang and Renjie Jing and Weikang Gu",

year = "1988",

language = "English",

isbn = "7800030393",

series = "Proc 1988 IEEE Int Conf Syst Man Cybern",

pages = "502--505",

editor = "Anon",

booktitle = "Proc 1988 IEEE Int Conf Syst Man Cybern",

note = "Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics ; Conference date: 08-08-1988 Through 12-08-1988",

}

TY - GEN

T1 - New Fourier descriptor based on areas (AFD) and its applications in object recognition

AU - Zhang, Zhongfei

AU - Jing, Renjie

AU - Gu, Weikang

PY - 1988

Y1 - 1988

N2 - The area Fourier descriptor (AFD), based on areas of the object's silhouettes, is presented, and the concept of multidimensional vector space is introduced. To speed up search procedures in the model base for matching, the authors apply a novel coding method to generate a curve-corner code (CCC) which can be used for any planar silhouettes. Taking the coordinate vectors in the multidimensional space introduced by the AFD as the main features, they developed an algorithm which can not only tolerate any translations, rotations, scale changing, and affine distortion, but also requires less computation. Because of the limited terms of the AFD, the algorithm is insensitive to truncation errors and can also be extended to recognize 3-D general cylindrical objects.

AB - The area Fourier descriptor (AFD), based on areas of the object's silhouettes, is presented, and the concept of multidimensional vector space is introduced. To speed up search procedures in the model base for matching, the authors apply a novel coding method to generate a curve-corner code (CCC) which can be used for any planar silhouettes. Taking the coordinate vectors in the multidimensional space introduced by the AFD as the main features, they developed an algorithm which can not only tolerate any translations, rotations, scale changing, and affine distortion, but also requires less computation. Because of the limited terms of the AFD, the algorithm is insensitive to truncation errors and can also be extended to recognize 3-D general cylindrical objects.

UR - https://www.scopus.com/pages/publications/0024170106

M3 - Conference contribution

SN - 7800030393

T3 - Proc 1988 IEEE Int Conf Syst Man Cybern

SP - 502

EP - 505

BT - Proc 1988 IEEE Int Conf Syst Man Cybern

A2 - Anon, null

T2 - Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics

Y2 - 8 August 1988 through 12 August 1988

ER -

New Fourier descriptor based on areas (AFD) and its applications in object recognition

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this