Identification of Oceanic Eddies in Satellite Images

Authors:
Armando Manuel Fernandes
Affiliations:
CENTRIA --- Centre for Artificial Intelligence, Universidade Nova Lisboa, Caparica, Portugal 2829-516
Venue:
ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing, Part II
Year:
2008

Citing 4
Cited 1

Topographic-based feature labeling for infrared oceanographic images

Pattern Recognition Letters
Singular Features in Sea Surface Temperature Data

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 1
Automatic water eddy detection in SST maps using random ellipse fitting and vectorial fields for image segmentation

DS'06 Proceedings of the 9th international conference on Discovery Science
A high-speed algorithm for elliptical object detection

IEEE Transactions on Image Processing

Automatic identification of oceanic eddies in infrared satellite images

Computers & Geosciences

Quantified Score

Hi-index	0.00

Visualization

Abstract

Ellipse detection has been presented in scientific literature as a potential method for finding eddies in satellite images because eddies may be roughly approximated by ellipses. In the present work, therefore, we describe a method to identify eddies based on ellipse center detection. The centers are determined using ellipse fitting. The proposed method allows finding several eddies per satellite image, an ability not demonstrated by previously reported eddy detectors based on ellipse recognition in binary images. The low values of the temperature gradients off the Iberian Peninsula complicate the application of the detection method because they raise difficulties in the creation of binary images with proper eddy outline. Most algorithms previously reported in the scientific literature are not applicable to the present case.