Structure tensor-based SIFT algorithm for SAR image registration

• Published in: IET image processing Vol. 14; no. 5; pp. 929 - 938
• Main Authors: S V, Divya, Paul, Sourabh, Pati, Umesh Chandra
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 17.04.2020
• Subjects: correct matches; correct matching rate; feature extraction; feature matching method; geophysical image processing; image matching; image registration; input SAR images; radar imaging; remote sensing; remote sensing image registration; Research Article; SAR image pairs; SAR image registration; scale layers; structure tensor‐based SIFT algorithm; synthetic aperture radar; synthetic aperture radar images; transforms; widely used feature extraction; structure tensor-based SIFT algorithm; correct matching rate; image registration; input SAR images; transforms; geophysical image processing; remote sensing; SAR image registration; widely used feature extraction; image matching; synthetic aperture radar images; remote sensing image registration; synthetic aperture radar; scale layers; feature extraction; radar imaging; feature matching method; SAR image pairs; correct matches
• ISSN: 1751-9659; 1751-9667; 1751-9667
• DOI: 10.1049/iet-ipr.2019.0568

The scale-invariant feature transform (SIFT) algorithm is the most widely used feature extraction as well as a feature matching method in remote sensing image registration. However, the performance of this algorithm is affected by the influence of speckle noise in synthetic aperture radar (SAR) images. It reduces the number of correct matches as well as the correct matching rate in SAR image registration. Moreover, SAR image registration is considered to be a challenging task as the images generally have significant geometric as well as intensity variations. To address these problems, a structure tensor-based SIFT algorithm is proposed to register the SAR images. At first, the tensor diffusion technique is used to construct the scale layers. Then, the features are extracted in the scale layers. Finally, feature matching is performed between the input SAR images and correct matches are identified. The proposed method can increase the number of correct matches as well as position accuracy in registration. Experiments have been conducted on five SAR image pairs to verify the effectiveness of the method.