Recent advancements in Earth observation technologies have accelerated remote sensing (RS) data acquisition, yet cross-domain semantic segmentation remains challenged by domain shifts. Traditional unsupervised domain adaptation (UDA) methods often rely on computationally intensive and unstable generative adversarial networks (GANs). This study introduces elevation-aware domain adaptation (EADA), a multi-task framework that integrates elevation estimation (via digital surface models) with semantic segmentation to address distribution discrepancies. EADA employs a shared encoder and task-specific decoders, enhanced by a spatial attention-based feature fusion module. Experiments on Potsdam and Vaihingen datasets under cross-domain settings (e.g., Potsdam IRRG → Vaihingen IRRG) show that EADA achieves state-of-the-art performance, with a mean IoU of 54.62% and an F1-score of 65.47%, outperforming single-stage baselines. Elevation awareness significantly improves the segmentation of height-sensitive classes, such as buildings, while maintaining computational efficiency. Compared to multi-stage approaches, EADA’s end-to-end design reduces training complexity without sacrificing accuracy. These results demonstrate that incorporating elevation data effectively mitigates domain shifts in RS imagery. However, lower accuracy for elevation-insensitive classes suggests the need for further refinement to enhance overall generalizability.

Monitoring Suspended Particulate Matter concentrations (CSPM) in inland waters is crucial for water management. Traditional semi-analytical models, limited by using only the four general bands (Red, Green, Blue, and Near-Infrared) in complex inland waters, struggle with CSPM retrieval for Medium-Spatial-Resolution Sensors. In this study, we divided inland waters into three classes and developed a hybrid semi-analytical retrieval model using four general bands. Two IOPs deriving algorithms (MQAR and MQANIR) were proposed to obtain bbp(R) and bbp(NIR). The spectral angle weighting method was used to incorporate bbp(R) and bbp(NIR) to enhance accuracy in optically variable Class 2 waters. Our model outperformed state-of-the-art methods (RMSLE = 0.24 mg/L, MAPE = 34.52%) and was independently validated with the Sediment Bulletin (RMSLE = 0.35 mg/L, MAPE = 43.47%). Applied to nine Landsat-8 OLI images in Poyang Lake, the model proved versatile for MSRS equipped with R, G, B, and NIR bands, eliminating the dependence on specific watercolor bands and expanding its applicability in various scenarios.

In this paper, we will conclude the results of Bufeng-1 (BF-1) A/B data processing, calibration workflow, and validation of the calibrated sea surface winds, land surface soil moisture, and sea surface height measurements. Since 2019, the BF-1 mission has operated in-orbit for over 4 years. The Earth reflected delay Doppler maps (DDMs) are continuously collected to perform global sea surface and land observations. At the same time, the intermediate frequency (IF) raw data are also obtained for 12 seconds every pass in diagnostic mode. To begin with, a brief description of the spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) technique will be provided in the introduction. Next, we will present the overview of Chinese BF-1 mission and the data specifications used in our research. In the next section, the BF-1 mission-related spaceborne power calibration and validation are presented to show the support to power DDM observable production for sea surface and land surface applications. Then, the status of Chinese Beidou System (BDS) Equivalent Isotropic Radiated Power (EIRP) acquisition programme is then introduced. Furthermore, the latest sea surface height (SSH) measurements results including two modes (group delay and carrier phase) and wind speed derivation based on machine learning (ML) method will be spatial-temporal aligned and validated with auxiliary datasets including Denmark Technology University (DTU) mean sea surface (MSS) products and European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5 reanalysis. The previous published results of sea surface winds retrieval under Hurricane conditions and soil moisture retrieval are also reviewed for the BF-1 mission applications. Finally, the conclusion of BF-1 derived results will be discussed to draw out ongoing/future works.

Machine learning has been increasingly used to retrieve chlorophyll-a (Chl-a) in optically variable waters. However, without the guidance of physical principles or expert knowledge, machine learning may produce biased mapping relationships, or waste considerable time searching for physically infeasible hyperparameter domains. In addition, most Chl-a retrieval models cannot evaluate retrieval uncertainty when ground observations are not available, and the retrieval uncertainty is crucial for understanding the model limitations and evaluating the reliability of retrieval results. In this study, we developed a novel knowledge-guided mixture density network to retrieve Chl-a in optically variable inland waters based on Sentinel-3 Ocean and Land Color Instrument (OLCI) imagery. The proposed method embedded prior knowledge derived from spectral shape classification into the mixture density network. Compared to another deterministic model, the knowledge-guided mixture density network outputted the conditional distribution of Chl-a given an input spectrum, enabling us to estimate the optimal retrieval and the associated uncertainty. The proposed method showed favorable correspondence with the field Chl-a, with root mean square error (RMSE) of 6.56 μg/L, and mean absolute percentage error (MAPE) of 43.64 %. Calibrated against Sentinel-3 OLCI spectrum, the proposed method also performed well when applied to field spectrum (RMSE = 4.58 μg/L, MAPE = 72.70 %), suggesting its effectiveness and good generalization. The proposed method provided the standard deviation of each estimated Chl-a, which enabled us to inspect the reliability of the estimated results and understand the model limitations. Overall, the proposed method improved the Chl-a retrieval in terms of model accuracy and uncertainty evaluation, providing a more comprehensive Chl-a observation of inland waters.

As an important soil nutrient, soil organic matter (SOM) plays an important role in plant growth and development and is an important indicator of soil fertility. Compared with the traditional laboratory chemical analysis methods, reflectance spectroscopy and hyperspectral remote sensing satellite data provide a new way of thinking for obtaining soil nutrient content in a wide range, non-destructively and efficiently. In this paper, we investigated the prediction effect of the two data based on different transformation forms, and also proposed a method based on the integration of laboratory measured spectral and hyperspectral remote sensing satellite image data, and the results showed that the integration of the two data can better combine the advantages of each other, and the best transformation form is the logarithmic transformation (LG), with the R 2 of 0.783 and the RMSE of 4.162, which can help the wide range of spatial mapping of soil organic matter content.

Unsupervised domain adaptation (UDA) is an approach to minimizing the domain gap. Generative methods are common approaches to minimizing the domain gap of aerial images, which improves the performance of the downstream tasks, for example, cross-domain semantic segmentation. For aerial images, the digital surface model (DSM) is usually available in both the source domain and the target domain. Depth information in DSM brings external information to generative models. However, little research utilizes it. In this letter, depth-assisted ResiDualGAN (DRDG) is proposed where depth supervised loss (DSL) and depth cycle consistency loss (DCCL) are used to bring depth information into the generative model. Experimental results show that DRDG reaches state-of-the-art accuracy between generative methods in cross-domain semantic segmentation tasks. Source code is available at https://github.com/miemieyanga/ResiDualGAN-DRDG .

Eutrophication is one of the major threats to the inland water ecosystem. Satellite remote sensing provides a promising way to monitor trophic state at large spatial scale in an efficient manner. Currently, most satellitebased trophic state evaluation approaches have focused on water quality parameters retrieval (e.g., transparency, chlorophyll-a), based on which trophic state was evaluated. However, the retrieval accuracies of individual parameter do not meet the demand for accurate trophic state evaluation, especially for the turbid inland waters. In this study, we proposed a novel hybrid model to estimate trophic state index (TSI) by integrating multiple spectral indices associated with different eutrophication level based on Sentinel-2 imagery. The TSI estimated by the proposed method agreed well with the in-situ TSI observations, with root mean square error (RMSE) of 6.93 and mean absolute percentage error (MAPE) of 13.77%. Compared with the independent observations from Ministry of Ecology and Environment, the estimated monthly TSI also showed good consistency (RMSE = 5.91,MAPE = 10.66%). Furthermore, the congruent performance of the proposed method in the 11 sample lakes (RMSE = 5.91, MAPE = 10.66%) and the 51 ungauged lakes (RMSE = 7.16, MAPE = 11.56%) indicated the favorable model generalization. The proposed method was then applied to assess the trophic state of 352 permanent lakes and reservoirs across China during the summers of 2016-2021. It showed that 10%, 60%, 28%, and 2% of the lakes/reservoirs are in oligotrophic, mesotrophic, light eutrophic, and middle eutrophic states respectively. Eutrophic waters are concentrated in the Middle-and-Lower Yangtze Plain, the Northeast Plain, and the Yunnan-Guizhou Plateau. Overall, this study improved the trophic state representativeness and revealed trophic state spatial distribution of Chinese inland waters, which has the significant meanings for aquatic environment protection and water resource management.

The performance of a semantic segmentation model for remote sensing (RS) images pre-trained on an annotated dataset greatly decreases when testing on another unannotated dataset because of the domain gap. Adversarial generative methods, e.g., DualGAN, are utilized for unpaired image-to-image translation to minimize the pixel-level domain gap, which is one of the common approaches for unsupervised domain adaptation (UDA). However, the existing image translation methods face two problems when performing RS image translation: (1) ignoring the scale discrepancy between two RS datasets, which greatly affects the accuracy performance of scale-invariant objects; (2) ignoring the characteristic of real-to-real translation of RS images, which brings an unstable factor for the training of the models. In this paper, ResiDualGAN is proposed for RS image translation, where an in-network resizer module is used for addressing the scale discrepancy of RS datasets and a residual connection is used for strengthening the stability of real-to-real images translation and improving the performance in cross-domain semantic segmentation tasks. Combined with an output space adaptation method, the proposed method greatly improves the accuracy performance on common benchmarks, which demonstrates the superiority and reliability of ResiDualGAN. At the end of the paper, a thorough discussion is conducted to provide a reasonable explanation for the improvement of ResiDualGAN. Our source code is also available.

As a pilot mission for the Chinese global navigation satellite system reflectometry (GNSS-R) constellation, the BuFeng-1 (BF-1) twin satellites A/B were launched on June 5th, 2019. Using three-month of sample data, this study presents the first evaluation of this mission for soil moisture (SM) estimations. The results show that the surface reflectivity (SR) derived from BF-1 bistatic radar cross section (BRCS) correlates well with the SM active passive (SMAP) SM (mean $R =0.85$ ), and the resulting BF-1 SM shows good correlation with the SMAP SM [ $R =0.94$ and root mean square deviation (RMSD) = 0.029] and the in situ measurements ( $R =0.77$ and RMSD = 0.049). The evaluation provides supportive information for the production and release of the new version of BF-1 Level 1 data and for the development of new algorithms to retrieve SM using data of this mission.

Eutrophication is a severe environmental pollution problem for inland waters and poses significant threats to the water safety. Monitoring trophic state of inland waters using optical remote sensing generally requires the inversion of water quality parameters, such as chlorophyll-a, secchi depth, etc. However, the accurate inversion of these individual indicators remains challenging, while the associated retrieval errors can propagate and degrade the evaluation of trophic state. Hence, we proposed a novel monitoring method by developing a Trophic State Index (TSI) based on optical remote-sensing parameters, i.e., Forel-Ule index (FUI) and non-water absorption coefficient at 674 nm (referred to as at-w(674)) retrieved from Sentinel-3 Ocean and Land Color Instrument (OLCI) imagery. The estimated TSI showed favorable correspondence with observed water quality data, including coefficient of determination (r2 = 0.91), root mean squared error (RMSE = 5.54), and mean absolute percentage error (MAPE = 10.69%). Using the Sentinel-3 OLCI data, the proposed method also had very good performance in the field spectrum (MAPE = 5.25 % , RMSE = 3.36). The monthly trophic state evaluation also showed congruence (MAPE = 12.51 % , RMSE = 6.41) with surface water quality monthly report (SWQMR) from the Ministry of Environment and Ecology of the People's Republic of China. The monthly TSI showed favorable agreement for 23 ungauged lakes (RMSE = 7.26, MAPE = 12.78%), indicating potential utility for regional lake water quality monitoring. The proposed method was then applied to 47 other large (>50 km2) water bodies in the Middle-and-Lower watershed of Yangtze River and the Huaihe watershed to evaluate the spatial and temporal variation of trophic state from 2016 to 2020. The TSI results revealed several lakes, such as Lake Honghu and Lake Luoma, with rapidly deteriorating water quality during the study period, while other lakes show relative improvement (e.g., Xiashan Reservoir), indicating unbalanced environmental pressure over the region. Overall, this study showed promising performance and potential for satellite-based monitoring of regional aquatic environments.

Unsupervised domain adaptation (UDA) is an approach to minimizing domain gap. Generative methods are common approaches to minimizing the domain gap of aerial images which improves the performance of the downstream tasks, e.g., cross-domain semantic segmentation. For aerial images, the digital surface model (DSM) is usually available in both the source domain and the target domain. Depth information in DSM brings external information to generative models. However, little research utilizes it. In this paper, depth-assisted ResiDualGAN (DRDG) is proposed where depth supervised loss (DSL), and depth cycle consistency loss (DCCL) are used to bring depth information into the generative model. Experimental results show that DRDG reaches state-of-the-art accuracy between generative methods in cross-domain semantic segmentation tasks.

A new land surface clustering algorithm is developed to retrieve soil moisture (SM) using the Global Navigation Satellite System reflectometry (GNSS-R) technique. Data from the BuFeng-1 (BF-1) twin satellites A/B, a pilot mission for the Chinese GNSS-R constellation, is used for SM retrieval. The core concept of the algorithm is to cluster global land areas into different types according to the land properties and calculate the SM type by type, based on the linear relationship between equivalent specular reflectivity and SM. The global comparison between the results and SM product from the Soil Moisture Active Passive mission shows the correlation coefficient (R) is 0.82, and unbiased root mean square error (ubRMSE) is 0.070 cm3·cm−3. The results also show good agreement compared with in situ SM measurements with the mean ubRMSE of 0.036 cm3·cm−3. This study proves that the global SM can be retrieved successfully from the BF-1 mission with the land surface clustering algorithm. By taking full advantage of the similarity of land surface physical properties in different regions, the algorithm provides a practical approach for global SM retrieval using spaceborne GNSS-R data.

Semantic segmentation is a critical problem for many remote sensing (RS) image applications. Benefiting from large-scale pixel-level labeled data and the continuous evolution of deep neural network architectures, the performance of semantic segmentation approaches has been constantly improved. However, deploying a well-trained model on unseen and diverse testing environments remains a major challenge: a large gap between data distributions in train and test domains results in severe performance loss, while manual dense labeling is costly and not scalable. To this end, we proposed an unsupervised domain adaptation framework for RS image semantic segmentation that is both practical and effective. The framework is supported by the consistency principle, including the cycle consistency in the input space and self-supervised consistency in the training stage. Specifically, we introduce cycle-consistent generative adversarial networks to reduce the discrepancy between source and target distributions by translating one into the other. The translated source data then drive a pipeline of supervised semantic segmentation model training. We enforce consistency of model predictions across target image transformations in order to provide self-supervision for the unlabeled target data. Experiments and extensive ablation studies demonstrate the effectiveness of the proposed approach on two challenging benchmarks, on which we achieve up to 9.95% and 7.53% improvements in accuracy over the state-of-the-art methods, respectively.

Suspended sediment concentration (SSC) is an important indicator of water quality that affects the biological processes of river ecosystems and the evolution of floodplains and river channels. The in situ SSC measurements are costly, laborious and spatially discontinuous, while the spaceborne SSC overcome these drawbacks and becomes an effective supplement for in situ observation. However, the spaceborne SSC observations of rivers are more challenging than those of lakes and reservoirs due to their narrow widths and the broad range of SSCs, among other factors. We developed a novel SSC retrieval method that is suitable for the rivers. Water was classified as clear or turbid based on the Forel–Ule index, and optimal SSC models were constructed based on the spectral responses to SSCs in cases of different turbidity. The estimated SSC had a strong correspondence with in situ measurements, with a root mean squared error (RMSE) of 24.87 mg/L and a mean relative error (MRE) of 51.91%. Satellite-derived SSC showed good consistency with SSCs obtained from gauging stations (r2 > 0.79). We studied the spatiotemporal variation in SSC in the Yangtze main stream from 2017 to 2021. It increased considerably from May to October each year, with the peak generally occurring in July or August (ca. 200–300 mg/L in a normal year and 800–1000 mg/L in a flood year), while it remained stable and decreased to around 50 mg/L from November to April of the following year. It was high in the east and low in the west, with local maxima in Chongqing (ca. 80–150 mg/L) and in the lower Dongting Lake reaches (ca. 80–100 mg/L) and a local minima in the downstream of the Three Gorges Dam (ca. 1–20 mg/L). Case studies in the Yibin reach and Three Gorges Reservoir determined that local variation in SSCs is due to special hydrodynamic conditions and anthropogenic activities. The procedure applied to process Sentinel-2 imagery and the novel SSC retrieval method we developed supplement the deficiencies in river SSC retrieval.

Recently, generating dense maps in real-time has become a hot research topic in the mobile robotics community, since dense maps can provide more informative and continuous features compared with sparse maps. Implicit depth representation (e.g., the depth code) derived from deep neural networks has been employed in the visual-only or visual-inertial simultaneous localization and mapping (SLAM) systems, which achieve promising performances on both camera motion and local dense geometry estimations from monocular images. However, the existing visual-inertial SLAM systems combined with depth codes are either built on a filter-based SLAM framework, which can only update poses and maps in a relatively small local time window, or based on a loosely-coupled framework, while the prior geometric constraints from the depth estimation network have not been employed for boosting the state estimation. To well address these drawbacks, we propose DiT-SLAM, a novel real-time Dense visual-inertial SLAM with implicit depth representation and Tightly-coupled graph optimization. Most importantly, the poses, sparse maps, and low-dimensional depth codes are optimized with the tightly-coupled graph by considering the visual, inertial, and depth residuals simultaneously. Meanwhile, we propose a light-weight monocular depth estimation and completion network, which is combined with attention mechanisms and the conditional variational auto-encoder (CVAE) to predict the uncertainty-aware dense depth maps from more low-dimensional codes. Furthermore, a robust point sampling strategy introducing the spatial distribution of 2D feature points is also proposed to provide geometric constraints in the tightly-coupled optimization, especially for textureless or featureless cases in indoor environments. We evaluate our system on open benchmarks. The proposed methods achieve better performances on both the dense depth estimation and the trajectory estimation compared to the baseline and other systems.

Soil nutrients, including soil available potassium (SAK), soil available phosphorous (SAP), and soil organic matter (SOM), play an important role in farmland soil productivity, food security, and agricultural management. Spectroscopic analysis has proven to be a rapid, nondestructive, and effective technique for predicting soil properties in general and potassium, phosphorous, and organic matter in particular. However, the successful estimation of soil nutrient content by visible and near-infrared (Vis-NIR) reflectance spectroscopy depends on proper calibration methods (including preprocessing transformation methods and multivariate methods for regression analysis) and the selection of appropriate variable selection techniques. In this study, raw spectrum and 13 preprocessing transformations combined with 2 variable selection methods (competitive adaptive reweighted sampling (CARS) and the successive projections algorithm (SPA)) and 2 regression algorithms (support vector machine (SVM) and partial least squares regression (PLSR)), for a total of 56 calibration methods, were investigated for modeling and predicting the above three soil nutrients using hyperspectral Vis-NIR data (400–2450 nm). The results show that first-order derivatives based on logarithmic and inverse transformations (FD-LGRs) can provide better predictions of soil available potassium and phosphorous, and the best form of soil organic matter transformation is SG+MSC. CARS was superior to the SPA in selecting effective variables, and the PLSR model outperformed the SVM models. The best estimation accuracies (R2, RMSE) for soil available potassium, phosphorous, and organic matter were 0.7532, 32.3090 mg/kg; 0.7440, 6.6910 mg/kg; and 0.9009, 3.2103 g/kg, respectively, and their corresponding calibration methods were (FD-LGR)/SPA/PLSR, (FD-LGR)/SPA/PLSR, and SG+MSC/CARS/SVM, respectively. Overall, for the prediction of the soil nutrient content, organic matter was superior to available phosphorous, followed by available potassium. It was concluded that the application of hyperspectral images (Vis-NIR data) was an efficient method for mapping and monitoring soil nutrients at the regional scale, thus contributing to the development of precision agriculture.

近年来视觉惯性SLAM是机器人及计算机视觉领域中的研究热点问题,在无人驾驶、增强现实、三维重建等领域有着广泛的应用.简述了视觉惯性SLAM的关键技术,并从统一的贝叶斯框架出发介绍了后端的两种形式——滤波和优化.对几个代表性视觉惯性SLAM系统进行深入对比分析,讨论了近年研究热点与发展趋势,并做出总结与展望.

Recently, deep learning has become the most innovative trend for a variety of high-spatial-resolution remote sensing imaging applications. However, large-scale land cover classification via traditional convolutional neural networks (CNNs) with sliding windows is computationally expensive and produces coarse results. Additionally, although such supervised learning approaches have performed well, collecting and annotating datasets for every task are extremely laborious, especially for those fully supervised cases where the pixel-level ground-truth labels are dense. In this work, we propose a new object-oriented deep learning framework that leverages residual networks with different depths to learn adjacent feature representations by embedding a multibranch architecture in the deep learning pipeline. The idea is to exploit limited training data at different neighboring scales to make a tradeoff between weak semantics and strong feature representations for operational land cover mapping tasks. We draw from established geographic object-based image analysis (GEOBIA) as an auxiliary module to reduce the computational burden of spatial reasoning and optimize the classification boundaries. We evaluated the proposed approach on two subdecimeter-resolution datasets involving both urban and rural landscapes. It presented better classification accuracy (88.9%) compared to traditional object-based deep learning methods and achieves an excellent inference time (11.3 s/ha).

This paper presents a novel change detection method for very-high-resolution images based on deep learning. In the method, an ensemble CNN change detection framework is proposed. Different from other deep learning change detection methods, samples of changed and unchanged regions of two very-high-resolution images acquired at different times are fed into two CNN. The discriminative deep metric learning based on dissimilarity degree is used to adjust discriminative distance metric of two CNN output layers quantitatively, under which the distance of unchanged samples becomes smaller and that of changed samples becomes higher, respectively. During its training procedure, cost module function based on dissimilarity degree of samples is used to train the ensemble CNN and high-level and abstract features of changed and unchanged pair of samples are driven to learn by the proposed framework. After training, the discriminative distance of unchanged samples becomes smaller and that of changed samples becomes larger. The proposed method justifies the changed and unchanged area of original images and change detection results can be obtained. Experiments on real datasets and theoretical analysis validate the effectiveness and superiority of the proposed method.

Spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) provides a new opportunity for land observation. This study is the first to compare and evaluate the performance of the only two spaceborne GNSS-R satellite missions whose data are publicly available, i.e., the UK’s TechdemoSat-1 (TDS-1) and the US’s Cyclone Global Navigation Satellite System (CYGNSS), for sensitivity analysis with SMAP SM on a daily basis and soil moisture (SM) estimates on a monthly basis over Mainland China. For daily sensitivity analysis, the two data were matched up and compared for the period (i.e., May 2017 through April 2018) when they coexisted (R = 0.561 vs. R = 0.613). For monthly SM estimates, a back-propagation artificial neural network (BP-ANN) was used to construct a model using data from more than two years. The model was subsequently used to derive long-term and continuous SM maps over Mainland China. The results showed that TDS-1 and CYGNSS agree and correlate very well with the SMAP SM in Mainland China (R = 0.676, MAE = 0.052 m3m−3, and ubRMSE = 0.060 m3m−3 for TDS-1; R = 0.798, MAE = 0.040 m3m−3, and ubRMSE = 0.062 m3m−3 for CYGNSS). The retrieved results were further validated using monthly in situ SM data from dense sites across Mainland China. It was found that the SM derived from the TDS-1/CYGNSS also correlated well with in situ SM (R = 0.687, MAE = 0.066 m3m−3, and ubRMSE = 0.056 m3m−3 for TDS-1; R = 0.724, MAE = 0.052 m3m−3, and ubRMSE = 0.053 m3m−3 for CYGNSS). The results in this study suggested that TDS-1/CYGNSS and the upcoming spaceborne GNSS-R mission could be new and powerful data sources to produce SM data set at a large scale and with relatively high precision.