Search

The use of remote sensing images for land cover classification is an important and challenging pixel-level classification task. However, the different distribution of the same land cover categories across different datasets, the accuracy of the classification is significantly reduced when a classification model trained on one dataset is used directly on another dataset. To address the issue, numerous unsupervised domain adaptation (UDA) methods have been proposed. However, the existing UDA methods focus mainly on natural images and are not suited to remote sensing images with large variations in spectral information and texture features. Therefore, we develop a new full-level domain adaptation network (FLDA-NET) applicable for cross-domain land cover classification of remote sensing images. It aligns the source domain and target domain through a two-stage process encompassing image-level, feature-level, and output-level. In stage I, we align at image-level by converting the source domain image to the target domain image style. In stage II, at the feature-level we align the entropy of the two domain features. At the output-level we do not use simple global alignment, but category-level alignment. Furthermore, a self-training strategy based on superpixel segmentation and softmax probability is proposed to further enhance the model's performance on the target domain. Extensive experiments on our proposed FLDA-NET are performed on the Potsdam and Vaihingen datasets and compared with other advanced UDA methods. The outcome demonstrates that this approach greatly improves the ability of cross-domain land cover classification in remote sensing images.

As the global climate changes and droughts become more frequent, understanding the characteristics and propagation dynamics of drought is critical for monitoring and early warning. This study utilized the Standardized Precipitation Evapotranspiration Index (SPEI), Vegetation Condition Index (VCI), and Groundwater Drought Index (GDI) to identify meteorological drought (MD), agricultural drought (AD), and groundwater drought (GD), respectively. Sen’s slope method and Mann–Kendall trend analysis were used to examine drought trends. The Pearson correlation coefficient (PCC) and theory of run were utilized to identify the propagation times between different types of droughts. Cross-wavelet transform (XWT) and wavelet coherence (WTC) were applied to investigate the linkages among the three types of droughts. The results showed that, from 2004 to 2022, the average durations of MD, AD, and GD in Henan Province were 4.55, 8.70, and 29.03 months, respectively. MD and AD were gradually alleviated, while GD was exacerbated. The average propagation times for the different types of droughts were as follows: 6.1 months (MD-AD), 4.4 months (MD-GD), and 16.3 months (AD-GD). Drought propagation exhibited significant seasonality, being shorter in summer and autumn than in winter and spring, and there were close relationships among MD, AD, and GD. This study revealed the characteristics and propagation dynamics of different types of droughts in Henan Province, providing scientific references for alleviating regional droughts and promoting the sustainable development of agriculture and food production.

Vapor pressure deficit (VPD) shows significant spatial and temporal variability in the context of global climate change, which is important for studying the implications of climate change on the structure and function of ecosystems to analyze the effects of VPD on vegetation dynamics. Spatial patterns of vegetation sensitivity to VPD have been recently investigated, however, the feedback of different vegetation types to VPD may vary depending on physiological characteristics, it is unclear how different types influence the sensitivity to VPD. In this study, the ERA5-Land reanalysis time-series dataset was used to analyze the spatial and temporal trends of VPD under different vegetation types. It was found that VPD showed an increasing trend in vegetated areas over the past 20 years with large spatial heterogeneity, generally increasing with drying conditions. On this basis, the spatial patterns of vegetation sensitivity to VPD and temporal trends in sensitivity were evaluated over the past 20 years in China using the enhanced vegetation index (EVI) and near-infrared reflectance of vegetation (NIRv) which can describe vegetation dynamics. The results show that the sensitivities under the two indices have high spatial consistency, with northeastern and central China showing positive sensitivities and southern China showing negative sensitivities, respectively. The positive sensitivities are relatively high for Deciduous Broadleaf Forests (DBF), Deciduous Needleleaf Forests (DNF), Grasslands (GL), and Croplands (CL) types, while the negative sensitivities are larger for Shrublands (SL) and Savannas (SA) types. Under different climatic zones, the sensitivity of CL and GL are independent of climatic zones (both showing positive), while the sensitivity of SL is negative in the Humid zone and positive in the Semi-Arid zone. Temporally, the sensitivity showed a slow increasing trend over the last 20 years. In terms of vegetation types, sensitivities of Evergreen Broadleaf Forests (EBF), DBF, GL and CL types showed a significant increasing trend (p

Surface albedo plays a pivotal role in the Earth’s energy balance and climate. This study conducted an analysis of the spatial distribution patterns and temporal evolution of albedo, normalized difference vegetation index (NDVI), normalized difference snow index snow cover (NSC), and land surface temperature (LST) within the Qilian Mountains (QLMs) from 2001 to 2022. This study evaluated the spatiotemporal correlations of albedo with NSC, NDVI, and LST at various temporal scales. Additionally, the study quantified the driving forces and relative contributions of topographic and natural factors to the albedo variation of the QLMs using geographic detectors. The findings revealed the following insights: (1) Approximately 22.8% of the QLMs exhibited significant changes in albedo. The annual average albedo and NSC exhibited a minor decline with rates of −0.00037 and −0.05083 (Sen’s slope), respectively. Conversely, LST displayed a marginal increase at a rate of 0.00564, while NDVI experienced a notable increase at a rate of 0.00178. (2) The seasonal fluctuations of NSC, LST, and vegetation collectively influenced the overall albedo changes in the Qilian Mountains. Notably, the highly similar trends and significant correlations between albedo and NSC, whether in intra-annual monthly variations, multi-year monthly anomalies, or regional multi-year mean trends, indicate that the changes in snow albedo reflected by NSC played a major role. Additionally, the area proportion and corresponding average elevation of PSI (permanent snow and ice regions) slightly increased, potentially suggesting a slow upward shift of the high mountain snowline in the QLMs. (3) NDVI, land cover type (LCT), and the Digital Elevation Model (DEM, which means elevation) played key roles in shaping the spatial pattern of albedo. Additionally, the spatial distribution of albedo was most significantly influenced by the interaction between slope and NDVI.

Land-cover mapping plays a crucial role in resource detection, ecological environmental protection, and sustainable development planning. The existing large-scale land-cover products with coarse spatial resolution have a wide range of categories, but they suffer from low mapping accuracy. Conversely, land-cover products with fine spatial resolution tend to lack diversity in the types of land cover they encompass. Currently, there is a lack of large-scale land-cover products simultaneously possessing fine-grained classifications and high accuracy. Therefore, we propose a mapping framework for fine-grained land-cover classification. Firstly, we propose an iterative method for developing fine-grained classification systems, establishing a classification system suitable for Sentinel-2 data based on the target area. This system comprises 23 fine-grained land-cover types and achieves the most stable mapping results. Secondly, to address the challenges in large-scale scenes, such as varying scales of target features, imbalanced sample quantities, and the weak connectivity of slender features, we propose an improved network based on Swin-UNet. This network incorporates a pyramid pooling module and a weighted combination loss function based on class balance. Additionally, we independently trained models for roads and water. Guided by the natural spatial relationships, we used a voting algorithm to integrate predictions from these independent models with the full classification model. Based on this framework, we created the 2017 Beijing–Tianjin–Hebei regional fine-grained land-cover product JJJLC-10. Through validation using 4254 sample datasets, the results indicate that JJJLC-10 achieves an overall accuracy of 80.3% in the I-level validation system (covering seven land-cover types) and 72.2% in the II-level validation system (covering 23 land-cover types), with kappa coefficients of 0.7602 and 0.706, respectively. In comparison with widely used land-cover products, JJJLC-10 excels in accurately depicting the spatial distribution of various land-cover types and exhibits significant advantages in terms of classification quantity and accuracy.

The scale and severity of forest disturbances across the globe are increasing due to climate change and human activities. Remote sensing analysis using time series data is a powerful approach for detecting large-scale forest disturbances and describing detailed forest dynamics. Various large-scale forest disturbance detection algorithms have been proposed, but most of them are only suitable for detecting high-magnitude forest disturbances (e.g., fire, harvest). Conversely, more continuous, subtle, and gradual lower-magnitude forest disturbances (e.g., thinning, pests, and diseases) have been subject to less focus. Deep learning (DL) can distinguish subtle differences in information within time series data, offering new opportunities to capture forest disturbances in a complete and detailed way. This study proposes an approach for analyzing forest dynamics across large areas and long time periods by combining DL time series classification and prior knowledge constraint. The approach consists of two stages: (1) an improved self-attention model used for time series classification to identify sequences with forest disturbance characteristics; (2) developed skip-disturbance recovery index (S-DRI) characterizing the temporal context, using prior knowledge constraint to identify forest disturbance years in time series with disturbance characteristics. In this study, the year of forest disturbances in five study areas located in the United States, Canada, and Poland from 2001 to 2020 was mapped. A total of 3082 manually interpreted test data with different disturbance causal agents (such as fire, harvest, conversion, hurricane, and pests) were sampled from five research areas for validation. Our approach was also evaluated against two forest disturbance benchmark datasets derived from LandTrendr and the Global Forest Change (GFC) dataset. The results demonstrate that our approach achieved an overall accuracy of 87.8%, surpassing the accuracy of LandTrendr (84.6%) and the Global Forest Change dataset (81.4%). Furthermore, our approach demonstrated lower omission rates (ranging from 10.0% to 67.4%) in detecting subtle to severe causal agents of forest disturbance, in comparison to LandTrendr (with a range of 18.0% to 81.6%) and GFC (with a range of 15.0% to 88.8%). This study, which involved mapping large-scale and long-term forest disturbance in multiple regions, revealed that our approach can be applied to new areas without a requirement for complex parameter adjustments. These results demonstrate the potential of our approach in generating comprehensive and detailed forest disturbance data, thus providing a new and effective method in this domain.

Imaging with different spectrums often leads to significant nonlinear differences in the intensity, which brings a great challenge to the automatic matching of multispectral images. Considering that there are often limitations to only using intensity information, this article studies multispectral image matching based on the structure consistency. First, an extended log-Gabor filter is constructed to build phase congruency maps, which encodes the structure information and provides rich and robust features. Then, a nonlinear diffusion-based algorithm is developed to detect the salient feature points on the phase congruency map, which are expected to be illumination and contrast invariant. Finally, the structure descriptors are built according to the orientation of the maximum log-Gabor filter responses, and the image matching is achieved by computing the correspondence. Extensive experiments are carried out on various multispectral image datasets. The results show that the proposed method holds a stable performance over the nonlinear intensity variance across spectrums, and outperforms the comparison methods in terms of the number of correct matches and the matching precision.

Timely and accurate agricultural information is essential for food security assessment and agricultural management. Synthetic aperture radar (SAR) systems are increasingly available in crop mapping, as they provide all-weather imagery. In particular, the Sentinel-1 sensor provides dense time-series data, thus offering a unique opportunity for crop mapping. However, in most studies, the Sentinel-1 complex backscatter coefficient was used directly which limits the potential of the Sentinel-1 in crop mapping. Meanwhile, most of the existing methods may not be tailored for the task of crop classification in time-series polarimetric SAR data. To solve the above problem, we present a novel deep learning strategy in this research. To be specific, we collected Sentinel-1 time-series data in two study areas. The Sentinel-1 image covariance matrix is used as an input to maintain the integrity of polarimetric information. Then, a depthwise separable convolution recurrent neural network (DSCRNN) architecture is proposed to characterize crop types from multiple perspectives and achieve better classification results. The experimental results indicate that the proposed method achieves better accuracy in complex agricultural areas than other classical methods. Additionally, the variable importance provided by the random forest (RF) illustrated that the covariance vector has a far greater influence than the backscatter coefficient. Consequently, the strategy proposed in this research is effective and promising for crop mapping.