Multipolar SAR ATR: Experiments with the GTRI dataset

Abstract

Multipolar radar data is a rich source of information. Even though multipolar synthetic aperture radar (SAR) images have been used extensively for remote sensing applications, the use of the same for airborne radar systems has been limited. With the release of the GTRI turn table multipolar data in public domain, the scopes to experiment with real multipolar data has increased. The current work reports the experiments done with the GTRI dataset to examine the effect of using multipolar data for SAR based target classification. The results indicate that multipolar data may give better classification performance and may also be useful to make the classifier less sensitive to minor variations between the training and test target images.

Separability indices and their use in radar signal based target recognition

“…The most important use of SIs is in predicting the classification performance possible for a given feature set, without running the actual classification exercises. This is important in predicting recognition performance from limited datasets [9]. The new SIs introduced in this work are deemed to be useful for different fields of machine learning and pattern recognition, especially in the field of radar signal processing.…”

Polarimetric Range Extended Target Detection via Adaptive Range Weighted Feature Extraction

“…The data used in this paper are constructed on the basis of the Georgia Technology Research Institute (GTRI) dataset [63,64]. In the GTRI raw data, HRRP data of T-72 target's full azimuth angles and different elevation angles are included.…”