Algorithm to Detect the Ship Wakes from SAR Imagery Based on the Wavelet Multiresolution Analysis

摘要

The characteristics of ship wakes in SAR images is analyzed, the model of speckle noise and its local statistic characteristics are also analyzed. The speckle noise which results from the coherent imaging mechanism of SAR images influences the image quality and blurs the image details. Since the speckle noise in SAR images can be approximated by additive noise, it is possible to suppress the speckle noise by reducing the amplitude of the detail images in wavelet subspace and preserve the edge information. A ship wake detection algorithm is developed in this paper, which combined the wavelet multiresolution analysis and adaptive threshold selection and edge detector with Radon transform to suppress the speckle noise and detect the ship wakes. The wavelet multiresolution analysis is used to suppress the speckle noise in SAR images and enhance the ship wake linear features, while Radon transform is used to detect the linear ship wakes in SAR images. The results of data processing show that the proposed method is more efficient and more accurate to detect the ship wakes in SAR images than that of applying Radon transform to SAR images directly.