SAR image statistical distribution and DBN-based SAR image classification method

Abstract

The invention discloses an SAR image statistical distribution and DBN-based SAR image classification method, and mainly solving the problem that traditional deep confidence networks DBN are easy to cause bad area consistency and incomplete edge information when being used for SAR image classification. The method comprises the following steps of: preprocessing a to-be-classified SAR image so as toobtain a DBN input matrix; designing a DBN formed by three limited Boltzmann machines; pre-training the designed network by using the input matrix so as to obtain a trained DBN; randomly selecting a set of partial pixels with category labels from labels of SAR image category mark sheet, and carrying out micro-adjustment on the trained DBN by using a counter-propagation algorithm; and carrying outpixel-by-pixel classification on the to-be-classified image by using the micro-adjusted DBN so as to obtain a classification result, and coloring and outputting the classification result. The method has the advantages of being excellent in classification result, good in area consistency and complete in edge information, and can be applied to terrain classification and target recognition of SAR images.

Description

Technical field [0001] The invention belongs to the field of image processing, specifically a SAR image classification method, which can be applied to SAR image classification and target recognition. Background technique [0002] Synthetic aperture radar SAR is an airborne radar or spaceborne radar that can produce high-resolution images. It has the characteristics of all-weather and all-weather, and is widely used in remote sensing and map surveying and other fields. A synthetic aperture radar image, that is, a SAR image is a radar signal that illuminates the ground. It reflects a landform image with a difference in light and dark tones, which is very similar to a black and white photo. For the interpretation of SAR images, it is much more difficult than optical images. The goal of SAR image classification is to label the SAR images obtained by using a well-designed classification algorithm for different types of ground features. . [0003] The statistical model of SAR images pl...

AI Technical Summary

Problems solved by technology

Among them, the model based on Markov random field MRF is mainly used to study the spatial consistency of neighborhood pixels. The statistical relationship between resolutions only studies the prior probability problem in the case of a single resolution, so it is not suitable for the classification of multi-resolution SAR images; and based on the Bag-of-Words model proposed by Jie Feng et al., in When it is used for the classification of SAR images, it is necessary to manually extract the underlying features, and the process is cumbersome

Images