Change Detection Method of Polarization SAR Image Based on Contourwave Integrated Dbn

Abstract

A polarimetric SAR image change detection method based on contour wave integrated DBN mainly solves the problem of low accuracy in polarimetric SAR image change detection in existing methods. The specific steps of the present invention are as follows: (1) input two polarization coherence matrices of two polarization SAR images of the same area at different phases; (2) filter out coherent noise; (3) normalize the filtered polarization Transform the coherence matrix; (4) Extract the feature matrix F 1 and F 2 The diagonal elements of each element in; (5) For the characteristic matrix E 1 and E 2 Perform non-subsampling contourlet transformation; (6) perform scale cascade fusion on the feature matrix; (7) construct a sample set; (8) construct a contourlet integration depth confidence network DBN change detection model; (9) train the contourlet integration depth Belief network DBN change detection model; (10) Output the change detection results. The present invention has the advantages of clear outline and high accuracy in detecting changes in polarized SAR images.

Description

Technical field [0001] The present invention belongs to the field of image processing technology, and further relates to a polarization synthetic aperture radar SAR (Synthetic Aperture Radar) image change detection based on the contour wave integrated deep belief network DBN model (Deep Belief Network) in the technical field of remote sensing image change detection technology. method. The present invention can be used to detect changes in water, ships, ports, etc. at corresponding positions in two polarized SAR images acquired in different time phases, and can be widely used in civil and military applications, such as land use, military monitoring, and urban planning , Post-disaster reconstruction, etc. Background technique [0002] Polarized SAR image change detection refers to the use of two polarized SAR images of the same area in different time phases to detect and analyze the changes on the ground. Compared with ordinary optical remote sensing technology, the polarization ...

AI Technical Summary

Problems solved by technology

Since this method only considers the local features of the two images, although it can obtain better change detection results, the disadvantage of this method is that the local filtering method used in this method makes the polarimetric SAR image still have residual spots. Noise, which affects the estimation of the equivalent visual number, so there is still the problem of low accuracy of change detection

Although this method extracts the unique polarization state of the polarimetric SAR image, it can better detect the polarimetric SAR image with obvious characteristics of the polarization state, but the disadvantage of this method is that the method uses the received power The ratio method is relatively simple to construct detection feature quantities, and cannot detect change areas that are not sensitive to power changes. At the same time, this method does not fully consider the scale feature information of polarimetric SAR images, resulting in fuzzy edges of change detection results and low accuracy.

Images