A spatiotemporal neural network for recognizing partially occluded objects

Pau Choo Chung, E. Liang Chen, Jia Bin Wu

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

In this paper, a spatiotemporal neural network for partially occluded object recognition is presented. The system consists of two major components: a feature extraction process and a spatiotemporal modular neural network. The former is made up of a sequence of preprocessing techniques including thresholding, boundary extraction, Gaussian filtering, and a splitand-merge algorithm to generate features that will represent the objects to be recognized. These acquired features are invariant to rotation, translation, and scaling and can serve as input to the spatiotemporal network that utilizes the concept of tap delay to account for spatial correlation between consecutive input features. A shape perceiver is designed into the network to extract continued parts of an object as well as to enable the inclusion of each object's unique characteristics into the system. Traditional neural network approaches for recognizing partially occluded objects have encountered significant problems because of the incomplete boundaries of the objects. In our approach, by creatively installing tap delays, the system can escape this limitation. Experimental results show that the proposed system can produce satisfactory results in efficiently and effectively recognizing partially occluded objects. Furthermore, intrinsic to this system is the ease by which it can be realized through parallel implementation, thus minimizing the tremendous time spent in matching object contours stored in a model database, as is the case in conventional recognition systems.

Original languageEnglish
Pages (from-to)1991-2000
Number of pages10
JournalIEEE Transactions on Signal Processing
Volume46
Issue number7
DOIs
Publication statusPublished - 1998 Dec 1

Fingerprint

Neural networks
Object recognition
Feature extraction

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

@article{ef18cb81628643539957b523ec3948b5,
title = "A spatiotemporal neural network for recognizing partially occluded objects",
abstract = "In this paper, a spatiotemporal neural network for partially occluded object recognition is presented. The system consists of two major components: a feature extraction process and a spatiotemporal modular neural network. The former is made up of a sequence of preprocessing techniques including thresholding, boundary extraction, Gaussian filtering, and a splitand-merge algorithm to generate features that will represent the objects to be recognized. These acquired features are invariant to rotation, translation, and scaling and can serve as input to the spatiotemporal network that utilizes the concept of tap delay to account for spatial correlation between consecutive input features. A shape perceiver is designed into the network to extract continued parts of an object as well as to enable the inclusion of each object's unique characteristics into the system. Traditional neural network approaches for recognizing partially occluded objects have encountered significant problems because of the incomplete boundaries of the objects. In our approach, by creatively installing tap delays, the system can escape this limitation. Experimental results show that the proposed system can produce satisfactory results in efficiently and effectively recognizing partially occluded objects. Furthermore, intrinsic to this system is the ease by which it can be realized through parallel implementation, thus minimizing the tremendous time spent in matching object contours stored in a model database, as is the case in conventional recognition systems.",
author = "Chung, {Pau Choo} and Chen, {E. Liang} and Wu, {Jia Bin}",
year = "1998",
month = "12",
day = "1",
doi = "10.1109/78.700970",
language = "English",
volume = "46",
pages = "1991--2000",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

A spatiotemporal neural network for recognizing partially occluded objects. / Chung, Pau Choo; Chen, E. Liang; Wu, Jia Bin.

In: IEEE Transactions on Signal Processing, Vol. 46, No. 7, 01.12.1998, p. 1991-2000.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A spatiotemporal neural network for recognizing partially occluded objects

AU - Chung, Pau Choo

AU - Chen, E. Liang

AU - Wu, Jia Bin

PY - 1998/12/1

Y1 - 1998/12/1

N2 - In this paper, a spatiotemporal neural network for partially occluded object recognition is presented. The system consists of two major components: a feature extraction process and a spatiotemporal modular neural network. The former is made up of a sequence of preprocessing techniques including thresholding, boundary extraction, Gaussian filtering, and a splitand-merge algorithm to generate features that will represent the objects to be recognized. These acquired features are invariant to rotation, translation, and scaling and can serve as input to the spatiotemporal network that utilizes the concept of tap delay to account for spatial correlation between consecutive input features. A shape perceiver is designed into the network to extract continued parts of an object as well as to enable the inclusion of each object's unique characteristics into the system. Traditional neural network approaches for recognizing partially occluded objects have encountered significant problems because of the incomplete boundaries of the objects. In our approach, by creatively installing tap delays, the system can escape this limitation. Experimental results show that the proposed system can produce satisfactory results in efficiently and effectively recognizing partially occluded objects. Furthermore, intrinsic to this system is the ease by which it can be realized through parallel implementation, thus minimizing the tremendous time spent in matching object contours stored in a model database, as is the case in conventional recognition systems.

AB - In this paper, a spatiotemporal neural network for partially occluded object recognition is presented. The system consists of two major components: a feature extraction process and a spatiotemporal modular neural network. The former is made up of a sequence of preprocessing techniques including thresholding, boundary extraction, Gaussian filtering, and a splitand-merge algorithm to generate features that will represent the objects to be recognized. These acquired features are invariant to rotation, translation, and scaling and can serve as input to the spatiotemporal network that utilizes the concept of tap delay to account for spatial correlation between consecutive input features. A shape perceiver is designed into the network to extract continued parts of an object as well as to enable the inclusion of each object's unique characteristics into the system. Traditional neural network approaches for recognizing partially occluded objects have encountered significant problems because of the incomplete boundaries of the objects. In our approach, by creatively installing tap delays, the system can escape this limitation. Experimental results show that the proposed system can produce satisfactory results in efficiently and effectively recognizing partially occluded objects. Furthermore, intrinsic to this system is the ease by which it can be realized through parallel implementation, thus minimizing the tremendous time spent in matching object contours stored in a model database, as is the case in conventional recognition systems.

UR - http://www.scopus.com/inward/record.url?scp=0032121788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032121788&partnerID=8YFLogxK

U2 - 10.1109/78.700970

DO - 10.1109/78.700970

M3 - Article

AN - SCOPUS:0032121788

VL - 46

SP - 1991

EP - 2000

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 7

ER -