A neural network approach for remote detection of marine eddies

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

A neural network approach for remote detection of marine eddies. / Castellani, Marco.

OCEANS 2006 - Asia Pacific. 2007. 4393861.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Castellani, M 2007, A neural network approach for remote detection of marine eddies. in OCEANS 2006 - Asia Pacific., 4393861, OCEANS 2006 - Asia Pacific, Singapore, 16/05/07. https://doi.org/10.1109/OCEANSAP.2006.4393861

APA

Vancouver

Author

Bibtex

@inproceedings{3f3c3ac2e9944299b76ac4e5d53d1c07,
title = "A neural network approach for remote detection of marine eddies",
abstract = "This paper presents a machine learning approach for detection of Mediterranean water eddies from sea surface temperature maps of the Atlantic Ocean. Two methods based on texture analysis of the satellite imagery are evaluated. Given a map point, the first method extracts information on the surrounding thermal gradient and arranges it as a numerical vector of gradient angles. The second method uses Laws' algorithm to create a vector of numerical measures of structural features. In both the cases, a neural network is trained to recognise those numerical patterns that reveal the presence of eddy structures. Both the algorithms achieve high recognition accuracy and fast and robust learning results. Particularly important are the very low rates of false detections obtained, since eddies occupy only a small portion of the ocean area. Compared to Laws' method, the gradient-based algorithm gives comparable recognition accuracies with a lower design effort and at reduced computational costs. The simple and modular structure of the gradient-based method also compares favorably to the complexity other algorithms for identification of marine phenomena published in the literature. Given the competitive accuracy results obtained, the gradient-based approach may be preferable to the currently employed techniques since it is simpler and more easily reconfigurable.",
author = "Marco Castellani",
year = "2007",
doi = "10.1109/OCEANSAP.2006.4393861",
language = "English",
isbn = "1424401380",
booktitle = "OCEANS 2006 - Asia Pacific",
note = "OCEANS 2006 - Asia Pacific ; Conference date: 16-05-2007 Through 19-05-2007",

}

RIS

TY - GEN

T1 - A neural network approach for remote detection of marine eddies

AU - Castellani, Marco

PY - 2007

Y1 - 2007

N2 - This paper presents a machine learning approach for detection of Mediterranean water eddies from sea surface temperature maps of the Atlantic Ocean. Two methods based on texture analysis of the satellite imagery are evaluated. Given a map point, the first method extracts information on the surrounding thermal gradient and arranges it as a numerical vector of gradient angles. The second method uses Laws' algorithm to create a vector of numerical measures of structural features. In both the cases, a neural network is trained to recognise those numerical patterns that reveal the presence of eddy structures. Both the algorithms achieve high recognition accuracy and fast and robust learning results. Particularly important are the very low rates of false detections obtained, since eddies occupy only a small portion of the ocean area. Compared to Laws' method, the gradient-based algorithm gives comparable recognition accuracies with a lower design effort and at reduced computational costs. The simple and modular structure of the gradient-based method also compares favorably to the complexity other algorithms for identification of marine phenomena published in the literature. Given the competitive accuracy results obtained, the gradient-based approach may be preferable to the currently employed techniques since it is simpler and more easily reconfigurable.

AB - This paper presents a machine learning approach for detection of Mediterranean water eddies from sea surface temperature maps of the Atlantic Ocean. Two methods based on texture analysis of the satellite imagery are evaluated. Given a map point, the first method extracts information on the surrounding thermal gradient and arranges it as a numerical vector of gradient angles. The second method uses Laws' algorithm to create a vector of numerical measures of structural features. In both the cases, a neural network is trained to recognise those numerical patterns that reveal the presence of eddy structures. Both the algorithms achieve high recognition accuracy and fast and robust learning results. Particularly important are the very low rates of false detections obtained, since eddies occupy only a small portion of the ocean area. Compared to Laws' method, the gradient-based algorithm gives comparable recognition accuracies with a lower design effort and at reduced computational costs. The simple and modular structure of the gradient-based method also compares favorably to the complexity other algorithms for identification of marine phenomena published in the literature. Given the competitive accuracy results obtained, the gradient-based approach may be preferable to the currently employed techniques since it is simpler and more easily reconfigurable.

UR - http://www.scopus.com/inward/record.url?scp=50249124425&partnerID=8YFLogxK

U2 - 10.1109/OCEANSAP.2006.4393861

DO - 10.1109/OCEANSAP.2006.4393861

M3 - Conference contribution

AN - SCOPUS:50249124425

SN - 1424401380

SN - 9781424401383

BT - OCEANS 2006 - Asia Pacific

T2 - OCEANS 2006 - Asia Pacific

Y2 - 16 May 2007 through 19 May 2007

ER -