Title :
Multi-feature extraction of ships from SAR images
Author :
Dandan Gu ; Xiaojian Xu
Author_Institution :
Sch. of Electron. & Inf. Eng., Beihang Univ., Beijing, China
Abstract :
Pose and structural features are widely used in automatic target recognition (ATR) systems. A procedure is proposed to extract such features of ships from synthetic aperture radar (SAR) images. The ship closed-contour image is first produced using a standard algorithm. A new two-step process is then presented for exact azimuth angle estimation. Modified criteria of centerline and minimum enclosing rectangle (MER) extraction are developed, and applied to calculate the ship dimension. Experimental results for numerically simulated SAR images demonstrate the effectiveness of the proposed technique.
Keywords :
feature extraction; radar imaging; synthetic aperture radar; SAR images; automatic target recognition systems; azimuth angle estimation; minimum enclosing rectangle extraction; ship closed-contour image; ships multifeature extraction; synthetic aperture radar images; Azimuth; Estimation; Feature extraction; Marine vehicles; Solid modeling; Synthetic aperture radar; Transforms; Hough transform; automatic ship recognition; feature extraction; synthetic aperture radar image;
Conference_Titel :
Image and Signal Processing (CISP), 2013 6th International Congress on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4799-2763-0
DOI :
10.1109/CISP.2013.6744037
