Abstract
With the resolution increasing, the structure information becomes more and more abundant in Synthetic Aperture Radar (SAR) images. The speckle noise generated by the coherent imaging mechanism, has a great influence on the detection accuracy and detection difficulty accordingly in high-resolution SAR change detection. In this paper, a multivariate change detection framework based on non-subsampled contourlet transform (NSCT), deep belief networks (DBN), fuzzy c-means (FCM) clustering, and global-local spatial pyramid pooling (SPP) net is proposed. NSCT decomposes the image into multiple scales and DBN is used for extracting feature of the decomposed coefficient matrix. FCM converges the similarity matrix of the initial features by DBN into two classes as a pseudo-label for global-local SPP net training data. The global-local SPP net consists of a large-scale region of interest (ROI) SPP net and a small-scale change detection SPP net. The combination of ROI and the SPP net, as well as the fusion between different scales, weakens the interference of the unchanged information and effectively eliminates a large number of redundant information. The experimental results show that our proposed method can effectively remove speckle noise and improve the robustness of high-resolution SAR change detection.
Highlights
High-resolution Synthetic Aperture Radar (SAR) image change detection is an important part of earth observing technology, which aims to reveal changed areas resulted from the evolution of the earth surface and has been widely used in the field of environmental monitoring, disaster assessment and military applications [1]–[3] and so on in recent years
We propose a multivariate change detection framework, in which non-subsampled contourlet transform (NSCT) [32], deep belief networks (DBN) [33], fuzzy c-means (FCM) clustering [34], [35], global-local spatial pyramid pooling (SPP) net [36] and region of interest (ROI) detection network [37] are combined to solve high-resolution SAR images change detection
DATA OF EXPERIMENTS In our experiment, there are three datasets were used to verify the validity of our proposed model in high-resolution SAR image change detection, namely: Namibia data, Brazil data and one set of simulation data
Summary
High-resolution SAR image change detection is an important part of earth observing technology, which aims to reveal changed areas resulted from the evolution of the earth surface and has been widely used in the field of environmental monitoring, disaster assessment and military applications [1]–[3] and so on in recent years. L. Li et al.: High-Resolution SAR Change Detection Based on ROI and SPP Net filtering technology uses various variational methods to transform the spatial domain into a frequency domain, such as Gabor transform [15], wavelet transform [16] and contourlet transform [17]. We propose a multivariate change detection framework, in which NSCT [32], DBN [33], FCM clustering [34], [35], global-local SPP net [36] and ROI detection network [37] are combined to solve high-resolution SAR images change detection. FEATURE EXTRACTION In order to improve the information obtaining ability of SAR images, we construct a robust network structure based on DBN. A couple of multitemporal high-resolution SAR images are respectively extracted by two DBNs with the same structure, and the extracted feature matrices are denoted as F1 and F2
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
