Abstract
The unavoidable noise often present in synthetic aperture radar (SAR) images, such as speckle noise, negatively impacts the subsequent processing of SAR images. Further, it is not easy to find an appropriate application for SAR images, given that the human visual system is sensitive to color and SAR images are gray. As a result, a noisy SAR image fusion method based on nonlocal matching and generative adversarial networks is presented in this paper. A nonlocal matching method is applied to processing source images into similar block groups in the pre-processing step. Then, adversarial networks are employed to generate a final noise-free fused SAR image block, where the generator aims to generate a noise-free SAR image block with color information, and the discriminator tries to increase the spatial resolution of the generated image block. This step ensures that the fused image block contains high resolution and color information at the same time. Finally, a fused image can be obtained by aggregating all the image blocks. By extensive comparative experiments on the SEN1–2 datasets and source images, it can be found that the proposed method not only has better fusion results but is also robust to image noise, indicating the superiority of the proposed noisy SAR image fusion method over the state-of-the-art methods.
Highlights
As one of the active microwave imaging radars, synthetic aperture radar (SAR) can work at any time and in any weather conditions
The training datasets were selected from SEN1–2 datasets [4], which contain more than twenty hundred thousand SAR-optical image pairs with the size of 256 ∗ 256 collected from across the globe and throughout all meteorological seasons
When nonlocal matching (NLM) is employed to process the source images, the size of the image block is set to 32 ∗ 32, and the maximum image block in each similar group is 20
Summary
As one of the active microwave imaging radars, synthetic aperture radar (SAR) can work at any time and in any weather conditions. The many advantages of SAR includes, among others, multi-polarization and variable angles, which allows SAR images to be widely used in geological surveys, military exercises, etc. [1,2]; due to its special coherent imaging mechanism, noise is inevitably generated in image acquisition, especially for speckle noise, resulting in serious inconvenience to the subsequent interpretation of the image processing; the effective suppression or removal of noise is one of the essential tasks required for SAR image pre-processing [3]. Multi-spectral sensors can obtain images with rich spectral information, such as color optical images [4]. Image fusion [5,6,7] is a powerful image processing tool for integrating complementary information from different sensors, by which a fused image with a more comprehensive and clearer description of the scene can be obtained. An increasing number of papers about image fusion are published every year—indicating the importance of image fusion—few papers are published regarding noisy SAR image fusion, despite the urgent need for an effective and practical SAR image fusion method
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
