Search

Road extraction from high-resolution remote sensing images is a critical task in image understanding and analysis, yet it poses significant challenges because of road occlusions caused by vegetation, buildings, and shadows. Deep convolutional neural networks have emerged as the leading approach for road extraction because of their exceptional feature representation capabilities. However, existing methods often yield incomplete and disjointed road extraction results. To address this issue, we propose CR-HR-RoadNet, a novel high-resolution road extraction network that incorporates local and global context reasoning. In this work, we introduce a road-adapted high-resolution network as the feature encoder, effectively preserving intricate details of narrow roads and spatial information. To capture multi-scale local context information and model the interplay between roads and background environments, we integrate multi-scale features with residual learning in a specialized multi-scale feature representation module. Moreover, to enable efficient long-range dependencies between different dimensions and reason the correlation between various road segments, we employ a lightweight coordinate attention module as a global context-aware algorithm. Extensive quantitative and qualitative experiments on three datasets demonstrate that CR-HR-RoadNet achieves superior extraction accuracy across various road datasets, delivering road extraction results with enhanced completeness and continuity. The proposed method holds promise for advancing road extraction in challenging remote sensing scenarios and contributes to the broader field of deep-learning-based image analysis for geospatial applications.

A high-quality built environment is important for human health and well-being. Assessing the quality of the urban built environment can provide planners and managers with decision-making for urban renewal to improve resident satisfaction. Many studies evaluate the built environment from the perspective of street scenes, but it is difficult for street-view data to cover every area of the built environment and its update frequency is low, which cannot meet the requirement of built-environment assessment under rapid urban development. Earth-observation data have the advantages of wide coverage, high update frequency, and good availability. This paper proposes an intelligent evaluation method for urban built environments based on scene understanding of high-resolution remote-sensing images. It contributes not only the assessment criteria for the built environment in remote-sensing images from the perspective of visual cognition but also an image-caption dataset applicable to urban-built-environment assessment. The results show that the proposed deep-learning-driven method can provide a feasible paradigm for representing high-resolution remote-sensing image scenes and large-scale urban-built-area assessment.

Building extraction using deep learning techniques has advantages but relies on a large number of clean labeled samples to train the model. Complex appearance and tilt shots often cause many offsets between building labels and true locations, and these noises have a considerable impact on building extraction. This paper proposes a new knowledge distillation-based building extraction method to reduce the impact of noise on the model and maintain the generalization of the model. The method can maximize the generalizable knowledge of large-scale noisy samples and the accurate supervision of small-scale clean samples. The proposed method comprises two similar teacher and student networks, where the teacher network is trained by large-scale noisy samples and the student network is trained by small-scale clean samples and guided by the knowledge of the teacher network. Experimental results show that the student network can not only alleviate the influence of noise labels but also obtain the capability of building extraction without incorrect labels in the teacher network and improve the performance of building extraction.

Accurate information on grassland above-ground biomass (AGB) is critical to better understanding the carbon cycle and conserve grassland resources. As a climate-sensitive key ecological function area, it is important to accurately estimate the grassland AGB of the Tibetan Plateau. Sentinel-2 (S2) images have advantages in reducing mixed pixels and the scale effect for remote sensing, while the data volume is correspondingly larger. In order to improve the estimation accuracy while reducing the data volume required for AGB estimation and improving the computational efficiency, this study used the Recursive Feature Elimination (RFE) algorithm to find the optimal feature set and compared the performance of the Cubist, Gradient Boosting Regression Tree (GBRT), random forest (RF) and eXtreme Gradient Boosting (XGBoost) algorithms for estimating AGB. In this study, ten S2 bands, ten S2-derived vegetation indexes, 218 pieces of AGB field survey data, four types of meteorological data and three types of topographic data were used as the alternative input features for the AGB estimation model. The impurity and permutation importance were used as the feature importance calculation method input to the RFE, and the Cubist, GBRT, RF and XGBoost algorithms were used to construct the AGB estimation models. The results showed that the RF algorithm based on the monthly average temperature (T), elevation, Normalized Difference Phenology Index (NDPI), Normalized Difference Infrared Index (NDII) and Palmer Drought Severity Index (PDSI) performed best (R2 = 0.8838, RMSE = 35.05 g/m2, LCCC = 2.44, RPPD = 0.91). The above findings suggest that the RF model based on the features related to temperature, altitude, humidity and leaf water content is beneficial to estimate the grassland AGB on the Tibetan Plateau.

China’s urbanization has dramatically accelerated in recent decades. Land for urban build-up has changed not only in large cities but also in small counties. Land cover mapping is one of the fundamental tasks in the field of remote sensing and has received great attention. However, most current mapping requires a significant manual effort for labeling or classification. It is of great practical value to use the existing low-resolution label data for the classification of higher resolution images. In this regard, this work proposes a method based on noise-label learning for fine-grained mapping of urban build-up land in a county in central China. Specifically, this work produces a build-up land map with a resolution of 10 m based on a land cover map with a resolution of 30 m. Experimental results show that the accuracy of the results is improved by 5.5% compared with that of the baseline method. This notion indicates that the time required to produce a fine land cover map can be significantly reduced using existing coarse-grained data.

The lack of pixel-level labeling limits the practicality of deep learning-based building semantic segmentation. Weakly supervised semantic segmentation based on image-level labeling results in incomplete object regions and missing boundary information. This paper proposes a weakly supervised semantic segmentation method for building detection. The proposed method takes the image-level label as supervision information in a classification network that combines superpixel pooling and multi-scale feature fusion structures. The main advantage of the proposed strategy is its ability to improve the intactness and boundary accuracy of a detected building. Our method achieves impressive results on two 2D semantic labeling datasets, which outperform some competing weakly supervised methods and are close to the result of the fully supervised method.