Road networks play a significant role in modern city management. It is necessary to continually extract current road structure, as it changes rapidly with the development of the city. Due to the success of semantic segmentation based on deep learning in the application of computer vision, extracting road networks from VHR (Very High Resolution) imagery becomes a method of updating geographic databases. The major shortcoming of deep learning methods for road networks extraction is that they need a massive amount of high quality pixel-wise training datasets, which is hard to obtain. Meanwhile, a large amount of different types of VGI (volunteer geographic information) data including road centerline has been accumulated in the past few decades. However, most road centerlines in VGI data lack precise width information and, therefore, cannot be directly applied to conventional supervised deep learning models. In this paper, we propose a novel weakly supervised method to extract road networks from VHR images using only the OSM (OpenStreetMap) road centerline as training data instead of high quality pixel-wise road width label. Large amounts of paired Google Earth images and OSM data are used to validate the approach. The results show that the proposed method can extract road networks from the VHR images both accurately and effectively without using pixel-wise road training data.

Geographic meshing system is an essential technology in digital earth framework and has essential applications in the integration and organization of heterogeneous spatial data, along with corresponding coding method. But current meshing and coding methods show unsatisfying locality and performance. To make an improvement, an adaptive Hilbert-Geohash meshing and coding method called AHG is proposed, which could represent both the location and the approximate size of the coded object directly by the meshing hierarchy and the corresponding coding length. This unique feature helps to accelerate the spatial range query and neighbor query. By simple string operations, many candidate objects that do not meet the query criteria can be quickly filtered out without precise spatial calculation. In addition, AHG code can also support spatial size query by finding objects whose size is within a certain interval quickly, without calculating the precise size of each object, which brings great convenience to spatial size statistics of massive spatial dataset. Demonstrated by experiments over different types of spatial dataset in a common PC environment, AHG shows favorable stability and scalability besides its capability in accelerating spatial query. The method is now applied successfully in several spatial query tools in a high-performance geographic information system called HiGIS.

Sparse view 3D reconstruction has attracted increasing attention with the development of neural implicit 3D representation. Existing methods usually only make use of 2D views, requiring a dense set of input views for accurate 3D reconstruction. In this paper, we show that accurate 3D reconstruction can be achieved by incorporating geometric priors into neural implicit 3D reconstruction. Our method adopts the signed distance function as the 3D representation, and learns a generalizable 3D surface reconstruction model from sparse views. Specifically, we build a more effective and sparse feature volume from the input views by using corresponding depth maps, which can be provided by depth sensors or directly predicted from the input views. We recover better geometric details by imposing both depth and surface normal constraints in addition to the color loss when training the neural implicit 3D representation. Experiments demonstrate that our method both outperforms state-of-the-art approaches, and achieves good generalizability.

Building tile-pyramids is an effective way for publishing and accessing the map visualization service of large-scale geospatial data in the web. But it is a time-consuming task in Geog ...

The processing and analysis of trajectories are the core of many location-based applications and services, while trajectory similarity is an essential concept regularly used. To address the time-consuming problem of similarity query, an efficient algorithm based on Fréchet distance called Ordered Coverage Judge (OCJ) is proposed, which could realize the filtering query with a given Fréchet distance threshold on large-scale trajectory datasets. The OCJ algorithm can obtain the result set quickly by a two-step operation containing morphological characteristic filtering and ordered coverage judgment. The algorithm is expedient to be implemented in parallel for further increases of speed. Demonstrated by experiments over real trajectory data in a multi-core hardware environment, the new algorithm shows favorable stability and scalability besides its higher efficiency in comparison with traditional serial algorithms and other Fréchet distance algorithms.

Measuring the similarity between a pair of trajectories is the basis of many spatiotemporal clustering methods and has wide applications in trajectory pattern mining. However, most measures of trajectory similarity in the literature are based on precise models that ignore the inherent uncertainty in trajectory data recorded by sensors. Traditional computing or mining approaches that assume the preciseness and exactness of trajectories therefore risk underperforming or returning incorrect results. To address the problem, we propose an amended ellipse model, which takes both interpolation error and positioning error into account by making use of the motion features of the trajectory to compute the ellipse’s shape parameters. A specialized similarity measure method considering uncertainty called the Uncertain Trajectory Similarity Measure (UTSM) based on the model is also proposed. We validate the approach experimentally on both synthetic and real-world data and show that UTSM is not only more robust to noise and outliers, but also more tolerant of different sample frequencies and asynchronous sampling of trajectories.