Determining the number of classes for segmentation in SAR sea ice imagery

Leen Kiat Soh, Costas Tsatsoulis

Research output: Contribution to conferencePaper

Abstract

In this paper, we describe a segmentation technique for SAR sea ice imagery that determines the number of classes in the image without a priori knowledge of the characteristics of the image. Image segmentation is important to sea ice research such as classification, and floe and lead analyses. In SAR sea ice imagery, however, backscatter characteristics vary for different seasons, temperatures, wind activity, and geographical locations, etc. As a result, image processing techniques that pre-determine the number of classes could generate segmentation that contains erroneous merging of classes and/or unnecessary separation of a class leading to unrecoverable mistakes during the classification phase. We have designed an image segmentation technique that combines image processing and machine learning methodologies. It computes spatial and textural statistics from the image and determine the number of classes by conceptually clustering these statistics. We have also tested this technique on a large database of sea ice imagery, and it has shown successes in determining the number of classes without human intervention.

Original languageEnglish (US)
Pages1565-1567
Number of pages3
StatePublished - Jan 1 1996
EventProceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4) - Lincoln, NE, USA
Duration: May 28 1996May 31 1996

Other

OtherProceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4)
CityLincoln, NE, USA
Period5/28/965/31/96

Fingerprint

Sea ice
segmentation
sea ice
synthetic aperture radar
imagery
Image segmentation
Image processing
image processing
Statistics
Merging
Learning systems
Lead
backscatter
methodology
Temperature
temperature

ASJC Scopus subject areas

  • Computer Science Applications
  • Earth and Planetary Sciences(all)

Cite this

Soh, L. K., & Tsatsoulis, C. (1996). Determining the number of classes for segmentation in SAR sea ice imagery. 1565-1567. Paper presented at Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4), Lincoln, NE, USA, .

Determining the number of classes for segmentation in SAR sea ice imagery. / Soh, Leen Kiat; Tsatsoulis, Costas.

1996. 1565-1567 Paper presented at Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4), Lincoln, NE, USA, .

Research output: Contribution to conferencePaper

Soh, LK & Tsatsoulis, C 1996, 'Determining the number of classes for segmentation in SAR sea ice imagery' Paper presented at Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4), Lincoln, NE, USA, 5/28/96 - 5/31/96, pp. 1565-1567.
Soh LK, Tsatsoulis C. Determining the number of classes for segmentation in SAR sea ice imagery. 1996. Paper presented at Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4), Lincoln, NE, USA, .
Soh, Leen Kiat ; Tsatsoulis, Costas. / Determining the number of classes for segmentation in SAR sea ice imagery. Paper presented at Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4), Lincoln, NE, USA, .3 p.
@conference{0e98c508f8d64fc1bd6dcc5c43008bc4,
title = "Determining the number of classes for segmentation in SAR sea ice imagery",
abstract = "In this paper, we describe a segmentation technique for SAR sea ice imagery that determines the number of classes in the image without a priori knowledge of the characteristics of the image. Image segmentation is important to sea ice research such as classification, and floe and lead analyses. In SAR sea ice imagery, however, backscatter characteristics vary for different seasons, temperatures, wind activity, and geographical locations, etc. As a result, image processing techniques that pre-determine the number of classes could generate segmentation that contains erroneous merging of classes and/or unnecessary separation of a class leading to unrecoverable mistakes during the classification phase. We have designed an image segmentation technique that combines image processing and machine learning methodologies. It computes spatial and textural statistics from the image and determine the number of classes by conceptually clustering these statistics. We have also tested this technique on a large database of sea ice imagery, and it has shown successes in determining the number of classes without human intervention.",
author = "Soh, {Leen Kiat} and Costas Tsatsoulis",
year = "1996",
month = "1",
day = "1",
language = "English (US)",
pages = "1565--1567",
note = "Proceedings of the 1996 International Geoscience and Remote Sensing Symposium. Part 3 (of 4) ; Conference date: 28-05-1996 Through 31-05-1996",

}

TY - CONF

T1 - Determining the number of classes for segmentation in SAR sea ice imagery

AU - Soh, Leen Kiat

AU - Tsatsoulis, Costas

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In this paper, we describe a segmentation technique for SAR sea ice imagery that determines the number of classes in the image without a priori knowledge of the characteristics of the image. Image segmentation is important to sea ice research such as classification, and floe and lead analyses. In SAR sea ice imagery, however, backscatter characteristics vary for different seasons, temperatures, wind activity, and geographical locations, etc. As a result, image processing techniques that pre-determine the number of classes could generate segmentation that contains erroneous merging of classes and/or unnecessary separation of a class leading to unrecoverable mistakes during the classification phase. We have designed an image segmentation technique that combines image processing and machine learning methodologies. It computes spatial and textural statistics from the image and determine the number of classes by conceptually clustering these statistics. We have also tested this technique on a large database of sea ice imagery, and it has shown successes in determining the number of classes without human intervention.

AB - In this paper, we describe a segmentation technique for SAR sea ice imagery that determines the number of classes in the image without a priori knowledge of the characteristics of the image. Image segmentation is important to sea ice research such as classification, and floe and lead analyses. In SAR sea ice imagery, however, backscatter characteristics vary for different seasons, temperatures, wind activity, and geographical locations, etc. As a result, image processing techniques that pre-determine the number of classes could generate segmentation that contains erroneous merging of classes and/or unnecessary separation of a class leading to unrecoverable mistakes during the classification phase. We have designed an image segmentation technique that combines image processing and machine learning methodologies. It computes spatial and textural statistics from the image and determine the number of classes by conceptually clustering these statistics. We have also tested this technique on a large database of sea ice imagery, and it has shown successes in determining the number of classes without human intervention.

UR - http://www.scopus.com/inward/record.url?scp=0029706431&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029706431&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029706431

SP - 1565

EP - 1567

ER -