Zhixin Li

Accurate building geometry information is crucial for urban planning in constrained spaces, fueling the growing demand for large-scale, high-precision 3D city modeling. Traditional methods like oblique photogrammetry and LiDAR prove time consuming and expensive for low-cost 3D reconstruction of expansive urban scenes. Addressing this challenge, our study proposes a novel approach to leveraging single-view remote sensing images. By integrating shadow information with deep learning networks, our method measures building height and employs a semantic segmentation technique for single-image high-rise building reconstruction. In addition, we have designed complex shadow measurement algorithms and building contour correction algorithms to improve the accuracy of building models in conjunction with our previous research. We evaluate the method’s precision, time efficiency, and applicability across various data sources, scenarios, and scales. The results demonstrate the rapid and accurate acquisition of 3D building data with maintained geometric accuracy (mean error below 5 m). This approach offers an economical and effective solution for large-scale urban modeling, bridging the gap in cost-efficient 3D reconstruction techniques.

Sea ice extraction and segmentation of remote sensing images is the basis for sea ice monitoring. Traditional image segmentation methods rely on manual sampling and require complex feature extraction. Deep-learning-based semantic segmentation methods have the advantages of high efficiency, intelligence, and automation. Sea ice segmentation using deep learning methods faces the following problems: in terms of datasets, the high cost of sea ice image label production leads to fewer datasets for sea ice segmentation; in terms of image quality, remote sensing image noise and severe weather conditions affect image quality, which affects the accuracy of sea ice extraction. To address the quantity and quality of the dataset, this study used multiple data augmentation methods for data expansion. To improve the semantic segmentation accuracy, the SC-U²-Net network was constructed using multiscale inflation convolution and a multilayer convolutional block attention module (CBAM) attention mechanism for the U²-Net network. The experiments showed that (1) data augmentation solved the problem of an insufficient number of training samples to a certain extent and improved the accuracy of image segmentation; (2) this study designed a multilevel Gaussian noise data augmentation scheme to improve the network’s ability to resist noise interference and achieve a more accurate segmentation of images with different degrees of noise pollution; (3) the inclusion of a multiscale inflation perceptron and multilayer CBAM attention mechanism improved the ability of U²-Net network feature extraction and enhanced the model accuracy and generalization ability.