An introduction to ENVI tools for Synthetic Aperture Radar (SAR) image despeckling and quantitative comparison of denoising filters
Mohammad R. Khosravi, Omid Akbarzadeh, Seyed Reza Salari, Sadegh Samadi, Habib Rostami · 2017 IEEE International Conference on Power, Control, Signals and Instrumentation Engineering (ICPCSI) · 2017
Presence of speckle noise in satellite and air-borne SAR images makes the images low quality in terms of textural features and spatial resolution which are required for processing issues such as image classification and clustering. Already, there are many adaptive filters in order to noise removal in SAR images. ENVI software is a fully applicable tool for this purpose which has a good library including several filters in the classes of adaptive, order-statistics and non-linear filters. In this study, the toolbox of ENVI is reviewed, analyzed and then numerically evaluated based on several single-band and multi-band (Pol-SAR) images achieved from SAR sensors such as TerraSAR-X. In order to the evaluation, two metrics including equivalent number of looks (ENL) and edge preservation index (EPI) are used which show the ability of the filters in preserving the joint spatial/textural features based on general information and edges' quality, respectively. It is notable that both metrics are classified into blind approaches in image quality assessment (IQA).