The annotations used during the training process are crucial for the inference results of remote sensing images (RSIs) based on a deep learning framework. Unlabeled RSIs can be obtained relatively easily. However, pixel-level annotation is a process that necessitates a high level of expertise and experience. Consequently, the use of small sample training methods has attracted widespread attention as they help alleviate reliance on large amounts of high-quality labeled data and current deep learning methods. Moreover, research on small sample learning is still in its infancy owing to the unique challenges faced when completing semantic segmentation tasks with RSI. To better understand and stimulate future research that utilizes semantic segmentation tasks with small data, we summarized the supervised learning methods and challenges they face. We also reviewed the supervised approaches with data that are currently popular to help elucidate how to efficiently utilize a limited number of samples to address issues with semantic segmentation in RSI. The main methods discussed are self-supervised learning, semi-supervised learning, weakly supervised learning and few-shot methods. The solution of cross-domain challenges has also been discussed. Furthermore, multi-modal methods, prior knowledge constrained methods, and future research required to help optimize deep learning models for various downstream tasks in relation to RSI have been identified.

Given the increasing frequency and severity of floods caused by climate change, there is a pressing requirement for creative ways to improve public comprehension and control of hydrological phenomena. Contemporary technology provides unparalleled possibilities to transform this domain. This project investigates the possibilities for merging gaming engines and digital twins to enhance flood education, data visualization, and interactive monitoring. This study proposes the utilization of immersive digital twins to enhance the comprehension of hydrological and hydraulic systems. The suggested method utilizes game engines to generate dynamic and interactive models that connect raw data to practical insights, enabling a more profound understanding of flood dynamics. This study underscores the wide-ranging usefulness of digital twins in various watersheds by focusing on the development of advanced monitoring systems, the benefits of improved data visualization, and educational outreach. The incorporation of real-time data via IoT technology considerably improves the significance and precision of these virtual models. This novel approach seeks to refashion flood management approaches by cultivating well-informed stakeholders and advocating for effective environmental education, ultimately leading to more resilient and prepared communities. An immersive digital twin of the real world can assist decision-makers technically, psychologically, and mentally by making complex phenomena easier to understand and visualize, thanks to real-time data and simulations that keep the information up-to-date, consequently leading to a more precise and intuitive decision-making process.

Spatiotemporally seamless precipitable water vapor (PWV) maps with high accuracy, spatiotemporal resolution, and continuity are of significance for climatical research. The current frameworks for the PWV maps have predominantly concentrated on fusing PWV derived from ERA5 reanalysis (ERA-PWV) and MODIS near-infrared data (MOD-NIR-PWV), falling short in accuracy. In this study, PWV derived from Global Navigation Satellite Systems (GNSS-PWV) is introduced to produce spatiotemporally PWV maps with improved accuracy. The fusion framework involves two main steps: 1) spatial fusion for producing an initial PWV map with spatiotemporal continuity through spherical cap harmonic analysis; 2) temporal fusion for producing a refined PWV map with high accuracy through residual correction. GNSS-PWV over 188 stations, 0.25°×0.25° ERA-PWV, and 0.05°×0.05° MOD-NIR-PWV over China from 2013 to 2018 are used to produce the daily 0.05°×0.05° PWV maps. The performance is evaluated using out-of-sample data, containing 18 GNSS-PWV and 90 ERA-PWV, and independent reference data, containing radiosonde-derived PWV over 72 stations. When compared to the out-of-sample GNSS-PWV and ERA-PWV, the PWV maps exhibit mean biases of -1.04 and -0.55 mm and RMS of 1.75 and 0.97 mm, respectively. These are equivalent to 8.7% and 32.1% reductions in bias and 25.5% and 49.5% reductions in RMSE relative to MOD-NIR-PWV. When radiosonde-derived PWV is used as the reference, the PWV maps have the mean bias and RMSE of -0.53 and 2.21 mm, respectively, which outperforms ERA-PWV (-0.75 and 2.69 mm). These results indicate the effectiveness of the novel fusion framework in producing seamless PWV maps.

The Yellow River Basin (YRB) has been facing severe water shortages; hence, the long-term dynamic monitoring of its surface water area (SWA) is essential for the efficient utilization of its water resources and sustainable socioeconomic development. In order to detect the changing trajectory of the SWA of the YRB and its influencing factors, we used available Landsat images from 1986 through to 2019 and a water and vegetation indices-based method to analyze the spatial–temporal variability of four types of SWAs (permanent, seasonal, maximum and average extents), and their relationship with precipitation (Pre), temperature (Temp), leaf area index (LAI) and surface soil moisture (SM).The multi-year average permanent surface water area (SWA) and seasonal SWA accounted for 46.48% and 53.52% in the Yellow River Basin (YRB), respectively. The permanent and seasonal water bodies were dominantly distributed in the upper reaches, accounting for 70.22% and 48.79% of these types, respectively. The rate of increase of the permanent SWA was 49.82 km²/a, of which the lower reaches contributed the most (34.34%), and the rate of decrease of the seasonal SWA was 79.18 km²/a, of which the contribution of the source region was the highest (25.99%). The seasonal SWA only exhibited decreasing trends in 13 sub-basins, accounting for 15% of all of the sub-basins, which indicates that the decrease in the seasonal SWA was dominantly caused by the change in the SWA in the main river channel region. The conversions from seasonal water to non-water bodies, and from seasonal to permanent water bodies were the dominant trends from 1986 to 2019 in the YRB. The SWA was positively correlated with precipitation, and was negatively correlated with the temperature. Because the permanent and seasonal water bodies were dominantly distributed in the river channel region and sub-basins, respectively, the change in the permanent SWA was significantly affected by the regulation of the major reservoirs, whereas the change in the seasonal SWA was more closely related to climate change. The increase in the soil moisture was helpful in the formation of the permanent water bodies. The increased evapotranspiration induced by vegetation greening played a significant positive role in the SWA increase via the local cooling and humidifying effects, which offset the accelerated water surface evaporation caused by the atmospheric warming.

The deep learning method has achieved great success in hyperspectral image classification, but the lack of labeled training samples still restricts the development and application of deep learning methods. In order to deal with the problem of small samples in hyperspectral image classification, a novel small sample classification method based on rotation-invariant uniform local binary pattern (RULBP) features and a graph-based masked autoencoder is proposed in this paper. Firstly, the RULBP features of hyperspectral images are extracted, and then the k-nearest neighbor method is utilized to construct the graph. Furthermore, self-supervised learning is conducted on the constructed graph so that the model can learn to extract features more suitable for small sample classification. Since the self-supervised training mainly adopts the masked autoencoder method, only unlabeled samples are needed to complete the training. After training, only a small number of samples are used to fine-tune the graph convolutional network, so as to complete the classification of all nodes in the graph. A large number of classification experiments on three commonly used hyperspectral image datasets show that the proposed method could achieve higher classification accuracy with fewer labeled samples.

As one of the three major black soil regions in the world, northeastern China has an important strategic position there. Since the 20th century, the local environment has undergone great changes under the influence of the natural economy, and it is particularly important to quantitatively assess the degree of change. However, there have been few long-term quantitative studies of environmental spatial-temporal variances in the three northeastern provinces. Therefore, in this study, four typical remote sensing indices of the normalized difference vegetation index (NDVI), land surface temperature (LST), normalized differential building–soil index (NDBSI) and wetness (WET) were employed to construct the remote sensing ecological index (RSEI) using a principal component analysis (PCA) method based on the Google Earth Engine (GEE) platform in northeastern China. The spatiotemporal variations in the eco-environmental quality were detected using linear slope and M–K test, and the direct and interactive effects of different influencing factors on the RSEI changes during 2000–2020 were explored based on geographic detection. The results show that the interannual variations in the RSEI show a fluctuating upward trend, with an increase percentage of 12.45% in the last two decades, indicating that the ecological quality of northeast China has gradually improved. Furthermore, that the western and eastern Heilongjiang provinces and western Jilin provinces contributed substantially to the improvement of environmental quality, while the environmental quality of Jilin provinces and central Liaoning provinces decreased to varying degrees. Compared with 2000, the area with a fair environmental quality grade had the greatest change, and had decreased by 60.69%. This was followed by the area with an excellent quality grade, which increased by 117%. Land-use type had the greatest impact on environmental changes in northeastern China, but the impact degree gradually decreased, while the impact of socioeconomic factors such as the gross production of agriculture, forestry, animal husbandry and fishery and population density on environmental quality gradually increased. The major reason for the decline of environmental quality in central Jilin and central Liaoning is that urbanization development had occupied a large amount of cropland. This shows that taking into account the virtuous cycle of an ecological environment while promoting urban and rural development may be an important task for northeastern China in the future.