Hypothesis test on quality measures for synthetic aperture radar images

In order to measure the quality of Synthetic Aperture Radar (SAR) images, we propose an evaluation based on hypothesis testing consisting of a parametric and non-parametric scenario. Parametric hypothesis test includes the u test, chi-square test, and F test, etc. Non-parametric hypothesis test is composed of K-S test, and the Pearson test, etc. The measurement objects of SAR images are divided into point targets and distributed targets, where point targets criteria comprise spacial resolution, equivalent resolution, peak side-lobe ratio (PSLR), and integration side-lobe ratio (ISLR), etc. Distributed targets criteria are mean value, variance, equivalent number of looks (ENL) and radiometric resolution, etc. In experiments of real SAR images indoors and overseas, the measure ENL is selected as an example to test the image quality by the parametric hypothesis u test. The result shows that the presented method demonstrates the hypothesis test is feasible and applicable for SAR images quality evaluation.