Robust Head Tracking Based on Hybrid Color Histogram and Random Walk Kalman Filter

Abstract

This paper examines a new robust color scheme and an adaptive object tracking technique. There are several popular color schemes used in face tracking which include Normalized RGB, Hue, Saturation, and Hybrid type of colors. Hybrid color schemes provide improved results compared to any single color scheme technique. Extensive experiments show the new robust Hybrid color scheme produced superior results in various lighting conditions. in conjunction with the robust hybrid color scheme to track head movements a supporting algorithm was needed to approximate the random path of the head movement. Kalman filter is a famous estimation technique in many areas to predict the route of moving object. We tested and developed a random-walk Kalman filter to track unpredictable and fast moving objects. The random-walk Kalman filter tolerates for tracking of quick random movements made by a person, which was not accommodated by linear tracking techniques.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 12, 2002
Accession Number
ADP014025

Entities

People

  • Dong C. Lin
  • Gwang-myung Kim
  • Jung H. Kim
  • Sung H. Yoon

Organizations

  • North Carolina Agricultural and Technical State University

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Automated Speech Recognition
  • Computer Science
  • Computer Vision
  • Computers
  • Equations
  • Estimators
  • Filters
  • Histograms
  • Image Processing
  • Image Recognition
  • Kalman Filters
  • Measurement
  • Object Recognition
  • Optimal Estimators
  • Random Walk
  • Recognition

Fields of Study

  • Computer science

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Computer Vision.
  • Vision Science/Vision Psychology/Cognitive Neuroscience.