Ship target detection and segmentation method based on multi-fractal analysis

Abstract

A novel automatic ship target detection and segmentation approach based on modified multi-fractal method in synthetic aperture radar (SAR) image is proposed in this study. Firstly, singular power spectrum (SPS) is developed to two-dimensional SPS (2D-SPS), and applied to feature extraction and SAR target detection. Secondly, 2D Holder exponent and 2D multi-fractal spectrum are jointly used to ship target segmentation in SAR image. The algorithm steps and detection flow is here conceptually assessed, analytically derived, numerically verified, and also tested on actual SAR images. Results indicate that the proposed method has significant advantages in ship target detection and segmentation in SAR image.

Keywords

• ships
• feature extraction
• object detection
• synthetic aperture radar
• radar imaging
• image segmentation
• fractals
• segmentation method
• multifractal analysis
• novel automatic ship target detection
• segmentation approach
• modified multifractal method
• synthetic aperture radar image
• singular power spectrum
• two-dimensional sps
• 2d-sps
• 2d holder exponent
• 2d multifractal spectrum
• target segmentation
• sar image
• detection flow
• actual sar images