SAR (Synthetic Aperture Radar) image segmentation method based on dictionary learning and sparse representation

Abstract

The invention discloses a SAR (Synthetic Aperture Radar) image segmentation technique based on dictionary learning and sparse representation, and mainly solves the problems that the existing feature extraction needs a lot of time and some defects exist in the distance measurement. The method comprises the following steps: (1) inputting an image to be segmented, and determining a segmentation class number k; (2) extracting a p*p window for each pixel point of the image to be segmented so as to obtain a test sample set, and randomly selecting a small amount of samples from the test sample set to obtain a training sample set; (3) extracting wavelet features of the training sample set; (4) dividing the training sample set by using a spectral clustering algorithm; (5) training a dictionary by using a K-SVD (Kernel Singular Value Decomposition) algorithm for each class of training samples; (6) solving sparse representation vectors of the test sample on the dictionary; (7) calculating a reconstructed error function of the test sample; and (8) calculating a test sample label according to the reconstructed error function to obtain the image segmentation result. The invention has the advantages of high segmentation speed and favorable effect; and the technique can be further used for automatic target identification of SAR images.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to SAR image segmentation, and is used for SAR image target recognition. Background technique [0002] Synthetic Aperture Radar (SAR) has all-weather and all-weather detection and reconnaissance capabilities. It uses pulse compression technology to obtain high distance resolution, and uses synthetic aperture principle to improve azimuth resolution, so it has unique advantages in the field of remote sensing compared with real aperture radar. The understanding and interpretation of SAR images belongs to the category of image processing, and also involves many disciplines such as signal processing, pattern recognition and machine learning. Due to the unique role of SAR, the understanding and interpretation of SAR images is receiving more and more attention in the fields of national defense and civilian use. SAR image segmentation, as one of the key links in the subsequent interpret...

AI Technical Summary

Problems solved by technology

However, most clustering-based SAR image segmentation methods need to extract image features. Although feature extraction is an offline process, it is very time-consuming because each pixel of the image needs to be extracted separately. In the clustering process, it is generally necessary to use some distance measure, such as Euclidean distance and manifold distance, to calculate the similarity between samples. Using a certain distance will be limited by the fixed defects of this distance measure method. For example, Euclidean distance is only good for spherical distribution data, while manifold distance is only good for data with manifold distribution. When the data distribution is unknown, if a certain distance measurement method is forced to be used, it may lead to the final SAR poor image segmentation

Images