Automatic identification approach for sea surface bubbles detection

Abstract

In this work a novel system for bubbles detection on sea surface images is presented. This application is basic to verify radiometer satellite systems which are used to the study of the floor humidity and the sea salinity. 160 images of 8 kinds of salinity have been processed, 20 per class. Two main steps have been implemented: image pre-processing and enhancing in order to improve the bubbles features, and segmentation and bubbles detection. A combination system has been performed with Support Vector Machines (SVM) in order to detect the sea salinity, showing a recognition rate of 95.43%.