Title :
Detection of Moroccan coastal upwelling in SST images using the Expectation-Maximization
Author :
Tamim, Ayoub ; Minaoui, Khalid ; Daoudi, Khalid ; Atillah, Abderrahman ; Aboutajdine, Driss
Author_Institution :
Mohammed V-Agdal Univ., Rabat, Morocco
Abstract :
This paper proposes an unsupervised algorithm for automatic detection and segmentation of upwelling region in Moroccan Atlantic coast using the Sea Surface Temperature (SST) satellite images. This has been done by exploring the Expectation-Maximization algorithm. The good number of clusters that best reproduces the shape of upwelling areas is selected by using the two popular Davies-Bouldin and Dunn indices. Area opening technique is developed that is used to remove and discarded the residuals noise in offshore waters not belonging to the upwelling region. The complete system has been validated by an oceanographer using a database of 30 SST images of the year 2007, demonstrating its capability and robustness for precise detection of Moroccan coastal upwelling.
Keywords :
expectation-maximisation algorithm; geophysical image processing; image denoising; image segmentation; object detection; ocean temperature; oceanographic regions; oceanographic techniques; remote sensing; Davies-Bouldin-Dunn indices; Moroccan Atlantic coast; Moroccan coastal upwelling detection; SST images; SST satellite images; area opening technique; expectation-maximization algorithm; oceanographer; offshore waters; residual noise removal; sea surface temperature satellite images; unsupervised algorithm; upwelling region segmentation; Clouds; Clustering algorithms; Image segmentation; Indexes; Ocean temperature; Satellites; Sea measurements; Area opening; Davies-Bouldin index; Dunn index; Expectation-Maximisation; Sea Surface Temperature; Upwelling;
Conference_Titel :
Intelligent Systems and Computer Vision (ISCV), 2015
Conference_Location :
Fez
Print_ISBN :
978-1-4799-7510-5
DOI :
10.1109/ISACV.2015.7106195
