Enhanced Deep Learning SAR Despeckling Networks Based on SAR Assessing Metrics

Abstract

The proposal of deep learning (DL) solutions for synthetic aperture radar (SAR) image despeckling has recently widespread. Such solutions have been mainly designed from a DL perspective by leveraging the training and validation stage on the use of typical norm-based cost functions. For going beyond the DL perspective, in this letter, we propose an SAR-based validation stage by using SAR assessing metrics in the design and hyperparameter selection of neural networks. In the first phase, SAR assessing metrics may be used only as validation metrics to highlight critical issues that cannot be spotted with standard image-processing quality metrics. In a second phase, the same SAR assessing metrics may be used directly for enhancing the DL solution by addressing specific issues that arose during the previous SAR-based validation stage. To this aim, three different DL SAR despeckling solutions and four different SAR assessing metrics have been considered. The outcome of this analysis shows the importance of including SAR knowledge in the training and validation stages of the design of a DL solution for SAR image despeckling.