Automatic Selection of Relevant Attributes for Multi-Sensor Remote Sensing Analysis: A Case Study on Sea Ice Classification

It is of considerable benefit to combine information obtained from different satellite sensors achieve advanced and improved characterization sea ice conditions. However, it also true that not all the relevant. may be redundant, corrupted, or unnecessary for given task, hence decreasing performance algorithms. Therefore, crucial select an optimal set image attributes which provides relevant content enhance efficiency accuracy interpretation retrieval geophysical parameters. Comprehensive studies have been focused on analysis features sensors, especially synthetic aperture radar. outcomes these are mostly data application-dependent can, therefore, rarely generalized. In this article, we employ a feature selection method based graph Laplacians, fully automatic easy implement. The proposed approach assesses global local level using two metrics selects regions according their physical characteristics observation recent study, investigate effectiveness classification, multi-sensor combinations. Experiments show advantage applying sets demonstrate selected by our result in high classification accuracies. We automatically considers varying technical, sensor-specific, environmental, conditions employing flexible adaptive as pre-processing step.