Remote sensing images (RSIs) Spatiotemporal fusion (STF) make a significant contribution to acquisition RSIs sequence with simultaneously high temporal and spatial resolution, which broadens its application fields. However, the existing RSIs STF methods lack effective strategies for extracting global information and fusion features between different images. Conversely, the existing State-Of-The-Art (SOTA) methods generally require more than two RSIs from different satellites as reference, which increases the difficulty of data collection and limits the application in practice. To address these problems, this paper proposed an end-to-end CNN and Transformer combined RSIs STF model (CTSTFM) based on two reference RSIs. Specifically, the proposed CTSTFM consists of three basic modules: Multi Kernel Convolutional Transformer encoder (MKCE), Cross Fusion module (CFM) and Convolutional-based Compression decoder (CCD). The MKCE combines Multi Kernel Channel Attention block and Multi Kernel Spatial Attention block to extract shadow features and long-term interdependencies in reference RSIs. The CFM uses the unique Cross Exchange Transformer block and Combine Fusion Transformer block to enhance the feature fusion results. Due to the powerful Encoder and Fusion Module, in the CCD part we only use a simple design convolution module to save the consumption of computational resources. Experiments on two well-known open access datasets shows that CTSTFM achieves competitive results in both qualitative and quantitative comparisons compared to the SOTA methods. Meanwhile, we conduct experiment to analyze the image Tessellation Effects and its solution. The effectiveness of the proposed module will be demonstrated through ablation experiments

Urban land intensive use meets the requirements for the sustainable development of urban land and is an important part of urban sustainable development. The Yangtze River Economic Belt (YREB) spans the three major regions of China, which are the most active areas of China’s economy. The contradiction between humans and land is becoming more acute. There are also regional differences in land use patterns affecting the coordinated development of the YREB and the construction of an ecological civilization. Therefore, the scientific evaluation of urban land intensive use is a key area in the current research field of urban sustainable development. In this study, the YREB is chosen as the research object, and urban land intensive use is studied using nighttime light data and statistical data on the urban built-up area. An evaluation model based on urban nighttime light intensity and land urbanization is constructed with an allometric growth model. Considering that the impact of land urbanization on urban nighttime light has a possible lag effect, an evaluation model of land intensive use that considers the lag effect between urban nighttime light and the land urbanization level is proposed. Using urban agglomerations and some typical cities in the study area as research samples, the characteristics of urban nighttime light and land urbanization are analyzed to reveal the spatial and temporal characteristics of land development in the YREB. The results show that nighttime light remote-sensing data can better reflect the level of urban land use, the allometric growth model can better fit the intensity of urban light and the land urbanization level, and the allometric growth characteristics can reflect the land use characteristics of different cities and urban agglomerations. In regional experiments with typical cities and with urban agglomerations, compared to the original allometric growth model, the goodness of fit of the allometric growth model with the lag effect improves, on average, by 3.2% and 2%, respectively, with the highest increases being by 9.9% and 4.9%, respectively. The level of intensive land use in the YREB gradually decreases from east to west, and there are great differences among different cities in the provinces and urban agglomerations. The lower reaches of the Yangtze River have high land intensive use on the whole. In the middle reaches, multicenter cities have a greater efficiency of land use than the surrounding cities. In the upper reaches, only Chengdu and Chongqing have clear advantages in urban land intensive use. The results of this study can be helpful in providing an important reference for the sustainable development of land in the YREB and can provide a basis for future urban land optimization and sustainable development. Realizing the coordination and linkage between key cities and major cities is the key to enhancing the overall sustainable development ability of the core cities in the YREB.

Remote sensing images are primary data sources for land use classification. High spatial resolution images enable more accurate analysis and identification of land cover types. However, a higher spatial resolution also brings new challenges to the existing classification methods. In the low-level feature spaces of remote sensing images, it is difficult to improve classification performance by modifying classifiers. Probabilistic topic models can connect low-level features and high-level semantics of remote sensing images. Latent Dirichlet allocation (LDA) models are representatives of probabilistic topic models. However, at present, probabilistic topic models are mainly adopted for scene classification and image retrieval in remote sensing image analysis only. In this study, multiscale segmentation was employed to construct bag-of-words (BoW) representations of high-resolution images. The segmented patches were then utilized as “image documents.” A structural topic model was used with an LDA model to import spatial information from the image documents at two levels: topical prevalence and topical content in the form of covariates. In this way, latent topic features in image documents can be more accurately deduced. The proposed method showed more satisfactory classification performance than standard LDA models and demonstrated a certain degree of robustness against the changes in the segmentation scale.

This study addresses the challenges related to the distribution of elderly care facilities in the Yangtze River Delta (YRD) region, which is experiencing a rapidly aging population. With over 176 million people aged 65 and above in China as of 2019 and the elderly population in the YRD continuing to grow, the study analyzes the spatial distribution, evolution, and supply–demand balance of elderly care facilities. Using GIS technologies, multi-source data analysis, and spatial autocorrelation techniques, the research identifies key regional patterns. Shanghai exhibits a clear hierarchical distribution of facilities, Jiangsu shows a “south strong, north weak” trend, while Zhejiang and Anhui demonstrate the opposite. The study also highlights a shift towards smaller, community-based care facilities, reflecting the growing demand for more localized services. It uncovers significant spatial mismatches and low coordination between supply and demand, particularly in rural and urban fringe areas, indicating the need for better regional coordination and more balanced resource distribution. To address these challenges, the study recommends (a) establishing cross-regional elderly care resource-sharing mechanisms; (b) promoting the development of small, community-based facilities; (c) integrating urban and rural services; and (d) leveraging technology for smart elderly care, including the use of big data and AI to optimize service delivery. These strategies aim to improve the equity and accessibility of elderly care services, ensuring that underserved areas receive better support. The findings provide a comprehensive framework for elderly care policies, offering valuable insights for other rapidly urbanizing regions and countries facing similar demographic challenges.

Traditional methods for 3D reconstruction of unmanned aerial vehicle (UAV) images often rely on classical multi-view 3D reconstruction techniques. This classical approach involves a sequential process encompassing feature extraction, matching, depth fusion, point cloud integration, and mesh creation. However, these steps, particularly those that feature extraction and matching, are intricate and time-consuming. Furthermore, as the number of steps increases, a corresponding amplification of cumulative error occurs, leading to its continual augmentation. Additionally, these methods typically utilize explicit representation, which can result in issues such as model discontinuity and missing data during the reconstruction process. To effectively address the challenges associated with heightened temporal expenditures, the absence of key elements, and the fragmented models inherent in three-dimensional reconstruction using Unmanned Aerial Vehicle (UAV) imagery, an alternative approach is introduced—the neural radiance field. This novel method leverages neural networks to intricately fit spatial information within the scene, thereby streamlining the reconstruction steps and rectifying model deficiencies. The neural radiance field method employs a fully connected neural network to meticulously model object surfaces and directly generate the 3D object model. This methodology simplifies the intricacies found in conventional 3D reconstruction processes. Implicitly encapsulating scene characteristics, the neural radiance field allows for iterative refinement of neural network parameters via the utilization of volume rendering techniques. Experimental results substantiate the efficacy of this approach, demonstrating its ability to complete scene reconstruction within a mere 5 min timeframe, thereby reducing reconstruction time by 90% while markedly enhancing reconstruction quality.

It is of great convenience to map daily evapotranspiration (ET) by remote sensing for agricultural water management without computing each surface energy component. This study used the operational simplified surface energy balance (SSEBop) and the remote sensing-based Penman–Monteith and Priestly–Taylor (RSPMPT) models to compute continuous daily ET over irrigated fields with the MODIS and CMADS data. The estimations were validated with eddy covariance (EC) measurements. Overall, the performance of RSPMPT with locally calibrated parameters was slightly better than that of SSEBop, with higher NSE (0.84 vs. 0.78) and R² (0.86 vs. 0.81), lower RMSE (0.78 mm·d⁻¹ vs. 0.90 mm·d⁻¹), although it had higher bias (0.03 mm·d⁻¹ vs. 0.01 mm·d⁻¹) and PBias (1.41% vs. 0.59%). Due to the consideration of land surface temperature, the SSEBop was more sensitive to ET’s change caused by irrigation before sowing in March and had a lower PBias (6.7% vs. 39.8%) than RSPMPT. On cloudy days, the SSEBop is more likely to overestimate ET than the RSPMPT. To conclude, driven by MODIS and CMADS data, the two simple models can be easily applied to map daily ET over cropland. The SSEBop is more practical in the absence of measured data to optimize the RSPMPT model parameters.