Abstract
Given the imaging characteristics of synthetic aperture radar (SAR) images and the inherent speckle noise in them, scale-invariant feature transform based algorithms are unable to perform satisfactorily. To improve registration efficiency between SAR images, we propose a robust and efficient registration method with three main contributions. First, considering sudden dark patches appearing in SAR images, we propose the ratio of exponentially weighted average blocks to suppress the sudden dark patches and better adapt to different test images. This new operator called blocks of the ratio of exponentially weighted averages (ROEWA-B) divides the processing windows of ROEWA into blocks, which can not only reduce speckle noise but also retain more edge details compared to ROEWA when sudden dark patches appear. Second, for outlier removal, we present an approach using the minimum moment map to remove erroneous keypoints. Finally, based on the gradient location orientation histogram descriptor, we propose a novel multiscale circle descriptor, which combines scale change information to give weights to feature points at different scales. Experimental results for various thresholds and evaluations demonstrate the advantage and robustness of our method in registration.
Highlights
KNOWN as a high resolution radar device, synthetic aperture radar (SAR) is a surveillance system with the ability of surface penetration for Earth observation, which can be installed on aircraft, satellite, spaceship and other flight platforms to observe the Earth in real time [1].SAR has unique superiorities in applications like disaster, environment and marine monitoring [2]
We present a novel multiscale circle descriptor based on the gradient location and orientation histogram (GLOH) [32] for accuracy and robustness in SAR image registration
It can be observed that the ratio of exponentially weighted averages (ROEWA)-B method reduces speckle noise and does not produce extremely high values for areas representing sudden dark patches, while retaining more clear edge details compared to the gradient by ratio (GR) method
Summary
KNOWN as a high resolution radar device, synthetic aperture radar (SAR) is a surveillance system with the ability of surface penetration for Earth observation, which can be installed on aircraft, satellite, spaceship and other flight platforms to observe the Earth in real time [1]. Some of the important issues in automatic SAR image matching are edge extraction with complex local distortion and multiplicative speckle noise The former results in imprecise detection of keypoints that affects transformation estimation; the latter generates multiple false feature points and has high computational cost. The main contributions of our work are: 1) Keypoint detection: we propose a new gradient computation method called blocks of the ratio of exponentially weighted averages (ROEWA-B) This new method suppresses large values of gradient magnitude when sudden dark patches appear in a SAR image and retains more edge details. Considering the constant false alarm rate property of the ROEWA detector, Dellinger et al [7] proposed the gradient by ratio (GR) method that is robust to speckle noise and dedicated to SAR images. Where gm, and go, are the magnitude and direction of gradient, respectively
Sign-in/Register to view highlights & summary 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 callMore From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
