A Technique for Feature Labeling in Infrared Oceanographic Images

Abstract

Thermal infrared images of the ocean obtained from satellite sensors are used for the study of ocean dynamics. Mesoscale ocean information from satellite data depends to a large extent on the correct interpretation of infrared oceanographic images. The difficulty of the image analysis and understanding problem for oceanographic images is due in part to a lack of precise mathematical descriptions coupled with the time varying nature of these features and the complication that the view of the ocean surface is typically obscured by clouds. This paper describes a technique that utilizes a non-linear probabilistic relaxation method for the oceanographic feature labeling problem. A unified mathematical framework that helps solve the problem is presented. This paper highlights the advantages of using contextual information in the feature labeling algorithm. This technique makes use of an efficient edge detection algorithm based on cluster shade texture measure. This algorithm is more suitable for labeling the mesoscale features. The paper presents results of experiments conducted at Remote Sensing Branch, NORDA on the NOAA AVHRR imagery data.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1989
Accession Number
ADA229909

Entities

People

  • Matthew Lybanon
  • N. Krishnakumar
  • Ron Holyer
  • S. Iyengar

Organizations

  • Louisiana State University

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Space

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Change Detection
  • Classification
  • Computer Science
  • Detection
  • Detectors
  • Dynamics
  • Expert Systems
  • Gulf Stream
  • Image Processing
  • Inference Engines
  • Infrared Images
  • Oceanography
  • Operating Systems
  • Pattern Recognition
  • Remote Sensing

Readers

  • Neural Network Machine Learning.
  • Oceanography.
  • Systems Analysis and Design

Technology Areas

  • Space