English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Synthetic Aperture Radar (SAR) image registration is to establish reliable correspondences among the images of the same scene. It is a challenging problem to register the airborne SAR images for the instability of airborne SAR systems and the lack of appropriate geo-reference data. Besides, techniques for registering satellite-based SAR images relying on rigorous SAR geocoding cannot be directly applied to airborne SAR images. To address this problem, we present a coarse-to-fine registration method for airborne SAR images by combining SAR-FAST (Features from Accelerated Segment Test) feature detector and DSP-LATCH (Domain-Size Pooling of Learned Arrangements of Three patCH) feature descriptor, which only relies on the gray level intensity of SAR data. More precisely, we first apply SAR-FAST, which is an adapted version of FAST for analyzing SAR images, to detect corners with high accuracy and low computational complexity. To reduce the disturbance of speckle noise as well as to achieve efficient and discriminative feature description, we further propose an improved descriptor named DSP-LATCH to describe the features, which combines the Domain-size Pooling scheme of DSP-SIFT (Scale-Invariant Feature Transform) and the idea of comparing triplets of patches rather than individual pixel values of LATCH. Finally, we conduct a coarse-to-fine strategy for SAR image registration by employing binary feature matching and the Powell algorithm. Compared with the existing feature based SAR image registration methods, e.g., SIFT and its variants, our method yields more reliable matched feature points and achieves higher registration accuracy. The experimental results on different scenes of airborne SAR images demonstrate the superiority of the proposed method in terms of robustness and accuracy.

COARSE-TO-FINE ACCURATE REGISTRATION FOR AIRBORNE SAR IMAGES USING SAR-FAST AND DSP-LATCH