Human posture estimation method from a single image using genetic algorithm and fuzzy inference

Hiroshi Kitajima, Masafumi Hagiwara

Research output: Contribution to journalArticle

Abstract

In this paper, we propose a method for estimating the observation direction of a person by applying fuzzy inference to the features obtained when a genetic algorithm is used to recognizee human posture from a natural image. Posture recognition uses the constraints obtained by representing the human body as parts expressed as connected structures to reduce the search space of the genetic algorithm. First, the head, which is a relatively easy part of the body to detect, is detected by pattern matching. Then, based on the position of the head, the body, arms, and legs are detected successively. A complex matching model becomes unnecessary because each part is approximated by a combination of straight lines and ellipses. The generated image of each part constructed from straight lines and ellipses is compared to the input image and optimized to obtain some level of overlap by the genetic algorithm. The obtained features are input to a fuzzy inference neural network, and the observation direction of the person can be estimated. The fuzzy rules can be automatically extracted by using Kohonen's self-organizing algorithm and the minimum least squares algorithm. The method of this paper applies to many people, and its effectiveness is verified.

Original languageEnglish
Pages (from-to)52-61
Number of pages10
JournalSystems and Computers in Japan
Volume31
Issue number12
DOIs
Publication statusPublished - 2000 Nov 15

Fingerprint

Fuzzy Inference
Fuzzy inference
Genetic algorithms
Genetic Algorithm
Straight Line
Person
Model Matching
Pattern matching
Least Square Algorithm
Fuzzy rules
Pattern Matching
Self-organizing
Fuzzy Rules
Search Space
Overlap
Neural Networks
Neural networks
Human
Observation

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Hardware and Architecture
  • Information Systems
  • Theoretical Computer Science

Cite this

Human posture estimation method from a single image using genetic algorithm and fuzzy inference. / Kitajima, Hiroshi; Hagiwara, Masafumi.

In: Systems and Computers in Japan, Vol. 31, No. 12, 15.11.2000, p. 52-61.

Research output: Contribution to journalArticle

@article{34e18a603d0a414a89594046aa929fd2,
title = "Human posture estimation method from a single image using genetic algorithm and fuzzy inference",
abstract = "In this paper, we propose a method for estimating the observation direction of a person by applying fuzzy inference to the features obtained when a genetic algorithm is used to recognizee human posture from a natural image. Posture recognition uses the constraints obtained by representing the human body as parts expressed as connected structures to reduce the search space of the genetic algorithm. First, the head, which is a relatively easy part of the body to detect, is detected by pattern matching. Then, based on the position of the head, the body, arms, and legs are detected successively. A complex matching model becomes unnecessary because each part is approximated by a combination of straight lines and ellipses. The generated image of each part constructed from straight lines and ellipses is compared to the input image and optimized to obtain some level of overlap by the genetic algorithm. The obtained features are input to a fuzzy inference neural network, and the observation direction of the person can be estimated. The fuzzy rules can be automatically extracted by using Kohonen's self-organizing algorithm and the minimum least squares algorithm. The method of this paper applies to many people, and its effectiveness is verified.",
author = "Hiroshi Kitajima and Masafumi Hagiwara",
year = "2000",
month = "11",
day = "15",
doi = "10.1002/1520-684X(20001115)31:12<52::AID-SCJ6>3.0.CO;2-Z",
language = "English",
volume = "31",
pages = "52--61",
journal = "Systems and Computers in Japan",
issn = "0882-1666",
publisher = "John Wiley and Sons Inc.",
number = "12",

}

TY - JOUR

T1 - Human posture estimation method from a single image using genetic algorithm and fuzzy inference

AU - Kitajima, Hiroshi

AU - Hagiwara, Masafumi

PY - 2000/11/15

Y1 - 2000/11/15

N2 - In this paper, we propose a method for estimating the observation direction of a person by applying fuzzy inference to the features obtained when a genetic algorithm is used to recognizee human posture from a natural image. Posture recognition uses the constraints obtained by representing the human body as parts expressed as connected structures to reduce the search space of the genetic algorithm. First, the head, which is a relatively easy part of the body to detect, is detected by pattern matching. Then, based on the position of the head, the body, arms, and legs are detected successively. A complex matching model becomes unnecessary because each part is approximated by a combination of straight lines and ellipses. The generated image of each part constructed from straight lines and ellipses is compared to the input image and optimized to obtain some level of overlap by the genetic algorithm. The obtained features are input to a fuzzy inference neural network, and the observation direction of the person can be estimated. The fuzzy rules can be automatically extracted by using Kohonen's self-organizing algorithm and the minimum least squares algorithm. The method of this paper applies to many people, and its effectiveness is verified.

AB - In this paper, we propose a method for estimating the observation direction of a person by applying fuzzy inference to the features obtained when a genetic algorithm is used to recognizee human posture from a natural image. Posture recognition uses the constraints obtained by representing the human body as parts expressed as connected structures to reduce the search space of the genetic algorithm. First, the head, which is a relatively easy part of the body to detect, is detected by pattern matching. Then, based on the position of the head, the body, arms, and legs are detected successively. A complex matching model becomes unnecessary because each part is approximated by a combination of straight lines and ellipses. The generated image of each part constructed from straight lines and ellipses is compared to the input image and optimized to obtain some level of overlap by the genetic algorithm. The obtained features are input to a fuzzy inference neural network, and the observation direction of the person can be estimated. The fuzzy rules can be automatically extracted by using Kohonen's self-organizing algorithm and the minimum least squares algorithm. The method of this paper applies to many people, and its effectiveness is verified.

UR - http://www.scopus.com/inward/record.url?scp=0034323872&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034323872&partnerID=8YFLogxK

U2 - 10.1002/1520-684X(20001115)31:12<52::AID-SCJ6>3.0.CO;2-Z

DO - 10.1002/1520-684X(20001115)31:12<52::AID-SCJ6>3.0.CO;2-Z

M3 - Article

AN - SCOPUS:0034323872

VL - 31

SP - 52

EP - 61

JO - Systems and Computers in Japan

JF - Systems and Computers in Japan

SN - 0882-1666

IS - 12

ER -