Classification results of ISAR sea targets based on their two features

This paper focuses on a classification of ISAR low resolution imaging of sea targets (cargo vessels) using the most popular classifiers. First, a method of the raw, passive data collection is described. Second, two feature vectors related to a size of the object (its width and length) are extracted, which then are applied for separating the objects into classes. Finally, the accuracy analysis of different ISAR image classifiers is presented. It is shown that the algorithms (ex. decision tree, SVM) may effectively distinguish different types of the cargo vessels.