Article
Geochemistry & Geophysics
Xiang Ding, Jian Kang, Zhe Zhang, Yan Huang, Jialin Liu, Naoto Yokoya
Summary: In this article, a novel convolutional sparse coding method called CoComCSC is proposed for interferometric phase restoration in SAR images. CoComCSC effectively reduces noise and preserves phase details, demonstrating superior performance compared to state-of-the-art methods. The study also indicates CoComCSC's potential for generating high-resolution DEM products.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Remote Sensing
Bowen Chi, Huifu Zhuang, Hongdong Fan, Yang Yu, Lei Peng
Summary: In this paper, an adaptive patch-based Goldstein filter (AP-Goldstein filter) is proposed, which adapts patch sizes based on pseudo-variation coefficient and controls noise suppression parameter alpha through pseudo-coherence. The proposed method effectively suppresses phase noise of interferograms while maintaining edge detailed information and improving filtering automation.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Environmental Sciences
Boyu Liu, Lingda Wu, Xiaorui Song, Hongxing Hao, Ling Zou, Yu Lu
Summary: Synthetic Aperture Radar Interferometry (InSAR) is a rapidly developing remote sensing technique, mainly used in terrain mapping and monitoring. However, the collected phase information in InSAR data inevitably contains noise, making it difficult to obtain the absolute phase from the wrapped phase. This study proposed a deep learning framework (PUnet) for phase unwrapping from InSAR data, which demonstrated high accuracy and robustness in obtaining absolute phases compared to traditional algorithms under various noise levels.
FRONTIERS IN ENVIRONMENTAL SCIENCE
(2023)
Article
Multidisciplinary Sciences
Josef Kellndorfer, Oliver Cartus, Marco Lavalle, Christophe Magnard, Pietro Milillo, Shadi Oveisgharan, Batu Osmanoglu, Paul A. Rosen, Urs Wegmuller
Summary: This dataset is the first of its kind to provide spatial representation of multi-seasonal C-band Synthetic Aperture Radar (SAR) interferometric repeat-pass coherence and backscatter signatures globally. It contains detailed information on how decorrelation affects interferometric measurements of surface displacement, making it valuable for various mapping applications.
Article
Geochemistry & Geophysics
Phan Viet Hoa Vu, Arnaud Breloy, Frederic Brigui, Yajing Yan, Guillaume Ginolhac
Summary: Phase linking (PL) is a prominent methodology for estimating coherence and phase difference in InSAR. However, the accuracy of the covariance matrix estimation step affects the performance of PL algorithms. In this study, we propose alternative statistical models and derive a unified algorithm for PL, which is validated using simulations and real Sentinel-1 data.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geography, Physical
Fengming Hu, Feng Wang, Yexian Ren, Feng Xu, Xiaolan Qiu, Chibiao Ding, Yaqiu Jin
Summary: This paper investigates the error sources of the array-InSAR interferograms and proposes a hybrid 3D phase unwrapping approach for 3D reconstruction. A hypothesis test is developed to identify the phase ambiguity and three indicators are proposed to identify reliable arcs. The L-1 norm approach is used to detect unwrapping errors and a two-tie network strategy is employed for global optimization. Experimental results demonstrate that the proposed algorithm can eliminate errors and provide a viable solution for rapid 3D SAR imaging.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Arturo Villarroya-Carpio, Juan M. Lopez-Sanchez, Marcus E. Engdahl
Summary: This study explores the use of Sentinel-1 interferometric coherence data as a tool for crop monitoring. By analyzing time series of Sentinel-1 and 2 images acquired during 2017, it was found that coherence can serve as a good measure for monitoring the crop growing season, showing strong correlations with the NDVI.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Geochemistry & Geophysics
Antonio Pepe, Pietro Mastro, Cathleen E. Jones
Summary: This article introduces an innovative space-time adaptive multilooking technique for sequence of differential synthetic aperture radar interferograms, based on directional statistics theory. The approach proposes a new method for the selection of statistically homogenous pixels, exclusively analyzing the multitemporal sequence of full resolution DInSAR interferograms, and demonstrates its effectiveness and simplicity.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Molly S. Zebker, Jingyi Chen, Marc A. Hesse
Summary: This study presents a method to characterize both surface deformation and tropospheric noise from interferogram subsets. By choosing different subsets of interferograms that use a common-reference SAR scene, tropospheric noise and deformation signals can be quantified. The results show that there is no detectable deformation signal in Oman, while the observed interference phase in Hawaii is mostly associated with tropospheric noise.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Maxwell Nogueira Peixoto, Gerhard Krieger, Alberto Moreira, Christian Waldschmidt, Michelangelo Villano
Summary: This paper proposes a method to enhance bistatic SAR interferometer with CubeSats, which can form additional interferograms with small baselines to resolve phase unwrapping errors in digital elevation models (DEMs). Despite the lower quality of CubeSat images, they can be used to detect and resolve phase unwrapping errors without impacting the resolution or accuracy of DEMs. This concept provides a cost-effective solution for generating highly accurate and robust DEMs, and paves the way for distributed SAR interferometric concepts based on CubeSats.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Jun Su Kim, Konstantinos Papathanassiou
Summary: This letter presents an alternative approach for estimating the vertical wavenumber over sloped terrain using range corregistration shifts, which simplifies the calculation effort without compromising the estimation performance. The proposed approach is demonstrated on ALOS PALSAR data and compared against the conventional methodology.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Geochemistry & Geophysics
S. Zwieback
Summary: In this paper, the author proposes three regularization methods (Hadamard, spectral, and Hadamard-spectral regularization) to estimate magnitudes more reliably in multilooked InSAR stack. The experimental results show that these regularizers can improve the accuracy of phase history, especially in low long-term coherence scenarios.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Roghayeh Zamani, Hossein Aghababaei
Summary: Synthetic aperture radar (SAR) interferometry has great potential in monitoring Earth's surface and detecting deformations. The use of multiple interferograms can improve parameter estimation accuracy. This paper proposes a method that utilizes contextual spatial information to enhance parameter estimation accuracy.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Environmental Sciences
Pasquale Imperatore, Eugenio Sansosti
Summary: This paper investigates the parallel processing of SAR image geometric coregistration using shared-memory architectures, proposing an efficient and scalable processor with thread-level parallelism. Experimental results demonstrate the potential operational use of the developed parallel processor in large-scale interferometric SAR data processing.
Article
Environmental Sciences
Ruonan Zhao, Zhabko Andrey Viktorovich, Junfeng Li, Chuang Chen, Meinan Zheng
Summary: This paper presents a strategy for extracting three-dimensional mining deformation using single-geometry SAR data. The methodology includes modeling the relationship between horizontal displacement and subsidence gradient and proposing a solution strategy to improve stability. The proposed method allows for the reconstruction of 3D displacements in mining areas using various types of SAR data. The effectiveness of the method is validated through simulation and in-site data, showcasing its applicability in mining deformation monitoring.
Article
Geography, Physical
Xin-Yi Tong, Gui-Song Xia, Xiao Xiang Zhu
Summary: High-resolution satellite images are valuable for land cover classification, but their application in detailed mapping at large scale is limited. To address this, we present a large-scale land cover dataset called Five-Billion-Pixels, with over 5 billion labeled pixels from 150 high-resolution Gaofen-2 satellite images. We also propose a deep-learning-based unsupervised domain adaptation approach for large-scale land cover mapping. Experimental results show promising performance even with entirely unlabeled images.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Geography, Physical
Wei Huang, Yilei Shi, Zhitong Xiong, Qi Wang, Xiao Xiang Zhu
Summary: RS image scene classification has gained attention for its applications. Conventional supervised approaches require labeled data, but with more RS images available, utilizing unlabeled data becomes urgent. This paper proposes a SSDA method called BSCA for RS cross-domain scene classification, using unsupervised and supervised alignment strategies to reduce domain shift.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Rodrigo Capobianco Guido, Tulay Adali, Emil Bjoernson, Laure Blanc-Feraud, Ulisses Braga-Neto, Behnaz Ghoraani, Christian Jutten, Alle-Jan Van der Veen, Hong Vicky Zhao, Xiaoxing Zhu
Summary: The IEEE Signal Processing Society has provided 75 years of service to the signal processing community, making significant contributions to technological advancement.
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Article
Geography, Physical
Jianhua Guo, Qingsong Xu, Yue Zeng, Zhiheng Liu, Xiao Xiang Zhu
Summary: Urban tree canopy maps are crucial for providing urban ecosystem services. This study developed a semi-supervised deep learning method to robustly segment urban trees from high-resolution remote sensing images in order to better serve Brazil's urban ecosystem. The results showed that the urban tree canopy coverage in Brazil ranges from 5% to 35%, with an average coverage of approximately 18.68%. These canopy maps quantified the nationwide urban tree canopy inequality problem in Brazil. It is expected that these maps will encourage research on Brazilian urban ecosystem services, support urban development, and improve inhabitants' quality of life to achieve the Agenda for Sustainable Development goals.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Jakob Gawlikowski, Sudipan Saha, Julia Niebling, Xiao Xiang Zhu
Summary: This paper proposes a method to incorporate source-wise out-of-distribution (OOD) detection into the fusion process of SAR and optical satellite data, aiming to improve the robustness to different types of OOD data and maintain the classification performance. The method adjusts the weights of extracted information based on the in-distribution probabilities, and shows significant improvement in handling individual data source failures or cloud coverage.
EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING
(2023)
Article
Geochemistry & Geophysics
Yi Wang, Nassim Ait Ali Braham, Zhitong Xiong, Chenying Liu, Conrad M. Albrecht, Xiao Xiang Zhu
Summary: This article introduces an unlabeled dataset SSL4EO-S12 for self-supervised pretraining of Earth observation satellite imagery. The authors demonstrate the effectiveness of SSL4EO-S12 in representative methods and multiple applications, and compare it with existing datasets.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2023)
Article
Remote Sensing
Xiangyu Zhao, Jingliang Hu, Lichao Mou, Zhitong Xiong, Xiao Xiang Zhu
Summary: This paper presents a deep transfer model with multiple sub-networks optimized by supervised loss and unsupervised loss. The model improves overall accuracy and average accuracy in climate zone classification. The proposed deep transfer network demonstrates outstanding performance.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Geochemistry & Geophysics
Fahong Zhang, Yilei Shi, Zhitong Xiong, Wei Huang, Xiao Xiang Zhu
Summary: This article proposes a self-training-based unsupervised domain adaptation method to tackle the domain shift problem in semantic segmentation. By exploiting feature-level relation among neighboring pixels, the method can regularize the prediction of the adapted model and outperform other UDA methods in public benchmarks.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Qian Song, Conrad M. M. Albrecht, Zhitong Xiong, Xiao Xiang Zhu
Summary: We propose a tree-level biomass estimation model using LiDAR data. Our model correlates tree height with biomass and demonstrates the Gaussian process regression model as a viable alternative to traditional biomass-height-diameter models. The model is validated with a dataset of 8342 samples covering seven global biomes. The study confirms a low relative error of below 1% for our model at stand-level (or plot-level).
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Remote Sensing
Qingyu Li, Lichao Mou, Yuansheng Hua, Yilei Shi, Sining Chen, Yao Sun, Xiao Xiang Zhu
Summary: Three-dimensional (3D) building structures play a vital role in understanding urban dynamics. Monocular remote sensing imagery is a cost-effective data source for large-scale building height retrieval. However, existing methods fail to consider the information of neighboring pixels belonging to the same building. Therefore, this study proposes a novel representation called 3D centripetal shifts, which incorporates both planar and vertical structures of buildings, and presents a robust solution named 3DCentripetalNet for building height retrieval.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Engineering, Electrical & Electronic
Lukas Kondmann, Sudipan Saha, Xiao Xiang Zhu
Summary: In this article, we explore the combination of unsupervised and supervised methods in a semisupervised framework to improve change detection performance. By using the unsupervised SiROC model to generate pseudolabels and selecting the most confident ones for pretraining different student models, we achieve robust results across various scenarios. The results show that pseudo-label pretraining produces significant performance gains, even when more labeled data is available. Moreover, the confidence selection of SiROC is effective and the performance gains are generalizable to unseen scenes.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Yaxiong Chen, Jinghao Huang, Lichao Mou, Pu Jin, Shengwu Xiong, Xiao Xiang Zhu
Summary: This article proposes a novel deep saliency smoothing hashing (DSSH) algorithm to learn effective hash codes for drone image retrieval by leveraging saliency capture mechanism, distribution smoothing term, global features, and local fine-grained features. Extensive experiments demonstrate that the DSSH algorithm can further improve retrieval performance compared with other deep hashing algorithms.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Qingsong Xu, Yilei Shi, Xin Yuan, Xiao Xiang Zhu
Summary: In this paper, a practical universal domain adaptation (UniDA) approach is proposed for remote sensing image scene classification, which requires no prior knowledge on the label sets. The proposed UniDA method distinguishes the shared and private label sets in each domain to promote adaptation and successfully recognize unknown samples. Empirical results demonstrate the effectiveness and practicality of the proposed model for remote sensing image scene classification, regardless of the availability of source data.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Remote Sensing
Qingyu Li, Sebastian Krapf, Yilei Shi, Xiao Xiang Zhu
Summary: Promoting solar technology can provide affordable, reliable, and modern energy for all people while reducing energy-related emissions and pollutants, contributing to sustainable development goals. Aerial imagery offers a cost-effective approach for large-scale rooftop solar potential analysis compared to other data sources.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Remote Sensing
Konrad Heidler, Lichao Mou, Di Hu, Pu Jin, Guangyao Li, Chuang Gan, Ji-Rong Wen, Xiao Xiang Zhu
Summary: Many deep learning approaches rely on pretrained backbone networks from large datasets, but the lack of such datasets in remote sensing is a challenge. We propose a label-free, self-supervised approach using imagery and audio correspondence to pretrain deep neural networks in remote sensing. We introduce the SoundingEarth dataset containing aerial imagery and crowd-sourced audio samples, and use it to pretrain ResNet models. Our approach outperforms existing pretraining strategies for remote sensing imagery. The dataset, code, and pretrained model weights are available at https://github.com/khdlr/SoundingEarth.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)