Local feature-based recognition of partially occluded objects using neural network

Nanning Zheng; Yaoyong Li; Wiek P.M. Houwers

Local feature-based recognition of partially occluded objects using neural network

Nanning Zheng
, Yaoyong Li
, Wiek P.M. Houwers

College of Artificial Intelligence and Robotics

Xi'an Jiaotong University

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

A new method of recognizing partially occluded objects using neural networks is presented. The neural network constitutes of a simplified ART-2 and a two feed forward network, and its inputs are the local features of objects. The network is firstly trained using a set of local features of known objects, then it can be used to recognize unknown object(s). Our numerical experiments using this method show the encouraging results, especially for recognizing the occluded objects.

Original language	English
Pages	1301-1306
Number of pages	6
State	Published - 1995
Event	Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 1 (of 2) - Orlando, FL, USA Duration: 6 Nov 1995 → 10 Nov 1995

Conference

Conference	Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 1 (of 2)
City	Orlando, FL, USA
Period	6/11/95 → 10/11/95

Cite this

@conference{3396de698d2d4403be095cc6df70288f,

title = "Local feature-based recognition of partially occluded objects using neural network",

abstract = "A new method of recognizing partially occluded objects using neural networks is presented. The neural network constitutes of a simplified ART-2 and a two feed forward network, and its inputs are the local features of objects. The network is firstly trained using a set of local features of known objects, then it can be used to recognize unknown object(s). Our numerical experiments using this method show the encouraging results, especially for recognizing the occluded objects.",

author = "Nanning Zheng and Yaoyong Li and Houwers, \{Wiek P.M.\}",

year = "1995",

language = "英语",

pages = "1301--1306",

note = "Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 1 (of 2) ; Conference date: 06-11-1995 Through 10-11-1995",

}

Local feature-based recognition of partially occluded objects using neural network. / Zheng, Nanning; Li, Yaoyong; Houwers, Wiek P.M.
1995. 1301-1306 Paper presented at Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 1 (of 2), Orlando, FL, USA.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Local feature-based recognition of partially occluded objects using neural network

AU - Zheng, Nanning

AU - Li, Yaoyong

AU - Houwers, Wiek P.M.

PY - 1995

Y1 - 1995

N2 - A new method of recognizing partially occluded objects using neural networks is presented. The neural network constitutes of a simplified ART-2 and a two feed forward network, and its inputs are the local features of objects. The network is firstly trained using a set of local features of known objects, then it can be used to recognize unknown object(s). Our numerical experiments using this method show the encouraging results, especially for recognizing the occluded objects.

AB - A new method of recognizing partially occluded objects using neural networks is presented. The neural network constitutes of a simplified ART-2 and a two feed forward network, and its inputs are the local features of objects. The network is firstly trained using a set of local features of known objects, then it can be used to recognize unknown object(s). Our numerical experiments using this method show the encouraging results, especially for recognizing the occluded objects.

UR - https://www.scopus.com/pages/publications/0029486618

M3 - 论文

AN - SCOPUS:0029486618

SP - 1301

EP - 1306

T2 - Proceedings of the 1995 IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation. Part 1 (of 2)

Y2 - 6 November 1995 through 10 November 1995

ER -

Local feature-based recognition of partially occluded objects using neural network

Abstract

Conference

Other files and links

Fingerprint

Cite this