Polarimetric synthetic aperture radar (PolSAR) has unique advantages in building extraction due to its sensitivity to building structures and all-time/all-weather imaging capabilities. However, the structure of buildings is complex, and buildings are easily confused with other objects in polarimetric SAR images. The speckle noise of SAR images will affect the accuracy of building extraction. This paper proposes a novel building extraction approach from PolSAR images with statistical texture and polarization features by using a convolutional neural network and superpixel. A feature space that is sensitive to building, including G0 statistical texture and PualiRGB features, is constructed and used as CNN input. Considering that the building boundary of the CNN classification result is inaccurate due to speckle noise, the simple linear iterative cluster (SLIC) superpixel is utilized to constrain the building extraction result. Finally, the effectiveness of the proposed method has been verified by experimenting with PolSAR images from three different sensors, including ESAR, GF-3, and RADARSAT-2. Experiment results show that compared with the other five PolSAR building extraction methods including threshold, SVM, RVCNN, and PFDCNN, our method without superpixel constraint, the F1-score of this method is the highest, reaching 84.22%, 91.24%, and 87.49%, respectively. The false alarm rate of this method is at least 10% lower and the F1 index is at least 6% higher when the building extraction accuracy is comparable. Further, the discussion and method parameter analysis results show that increasing the use of G0 statistical texture parameters can improve building extraction accuracy and reduce false alarms, and the introduction of superpixel constraints can further reduce false alarms.
The soil freeze/thaw (F/T) cycles play an important role in the climate system and human activities. However, the harsh environment in the Qinghai-Tibet Plateau (QTP) poses great challenges for both in-situ observation and remote-sensing monitoring of the soil F/T process. In this article, the time series of scattering and coherence of the high-resolution Sentinel-1 C-band synthetic aperture radar (SAR) is analyzed to identify the soil F/T state. The time series of scattering, including intensity and decomposition parameters, and coherence, are analyzed based on three typical landcover types (i.e., desert, grassland, and meadow) in the QTP. They are given the mathematical description by second-order and fourth-order Fourier functions, respectively. Based on Fourier functions, the initial F/T time points of the soil are detected in each pixel to draw the F/T map of the entire study area. The experiment results are cross-validated with the initial F/T time points of the soil calculated from the MODIS land surface temperatures, showing that the differences in days are less than one revisit cycle of Sentinel-1 (i.e., 12 days). Furthermore, the possible impacts of environmental factors acquired from the Wudaoliang meteorological station, including air temperature, ground surface temperature, snow depth, and precipitation, on scattering and coherence are discussed. This study explores that Sentinel-1 has great potential for soil F/T monitoring in the QTP, which can indicate F/T states of the surface soil as well as F/T information of the deeper soil with a high spatial-temporal resolution.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.