Jiechen Wang

Multiscale Grid Method for Detection and Reconstruction of Building Roofs from Airborne LiDAR Data

This study proposes a multiscale grid method to detect and reconstruct building roofs from airborne LiDAR data. The method interpolates unorganized LiDAR point cloud into two sets of grids with different spatial scales. In the large-scale grid, building seed regions are obtained, including detection of initial building seed regions and refinement of building seed regions. In the small-scale grid, to detect the detailed features of building roofs with complicated top structures, a high-resolution depth image is generated by a new iterative morphological interpolation using gradually increasing scales, and then segmented by using a full λ-schedule algorithm. Based on the building seed regions, detailed roof features are detected for each building and 3-D building roof models are then reconstructed according to the elevation of these features. Experiments are analyzed from several aspects: the correctness and completeness, the elevation accuracy of building roof models, and the influence of elevation to 3-D roof reconstruction. The experimental results demonstrate promising correctness, completeness, and elevation accuracy, with a satisfactory 3-D building roof models. The strategy of hierarchical spatial scale (from large scale to small scale) obtains the complementary advantage between technical applicability in a large urban environment and high quality in 3-D reconstruction of building roofs with fine details.

Integration of LiDAR Data and Orthophoto for Automatic Extraction of Parking Lot Structure

To overcome the challenges of parking lot structure extraction using optical remote sensing images, this study proposes an automatic method for the extraction of parking lot structure by integrating LiDAR data and orthophoto, which consists of three steps. The first step is to extract vehicles from LiDAR data and then to identify the corresponding central axes for each vehicle. In the second step, orientations of the identified vehicle central axes are used as principle orientation constraints for parking lines extraction from orthophoto. The third step is the determination of parking lot structure with vehicle central axes and parking lines, in which parking lot parameters are calculated and an adaptive growth method is used for parking lot structure determination. In this method, vehicle central axes identified from LiDAR data and parking lines extracted from orthophoto are integrated for the extraction of parking lot structures. The main novelty of this study lies in two new algorithms: an algorithm on parking lines extraction with principal orientation constraints and an algorithm on parking lot structure determination based on parameter solution and adaptive growth. The experiment shows that the proposed method can effectively extract parking lot structure with high correctness, high completeness, and good geometric accuracy.

Parallel scanline algorithm for rapid rasterization of vector geographic data

With the expansion of complex geographic calculations and the increase of spatial data types involved in the spatial analysis of large areas, the need becomes urgent for fast rasterization of massive multi-source geographic vector data. A parallel scanline algorithm is proposed for rapid rasterization. It provides a systematic solution to solve the complicated situation in parallel processing (cross-processor boundaries, common boundaries, and tiny polygons), thus ensuring the accuracy of the parallel scanline algorithm. The relationship of parallel speedup with the number of processors, the data partition pattern, and the raster grid size is discussed. Massive vector geographic data (approximately 0.7 million polygons) used in the experiment were effectively processed, thereby dramatically reducing the processing time and getting good speedup.

Rangeland Grasshoppers in Relation to Soils in the Qinghai Lake Region, China

The relationship between rangeland grasshopper density and soil type as well as topsoil moisture content was analyzed with in situ soil data collected in the Qinghai Lake region of China. Grasshoppers were confined mainly to the areas with light chestnut soil or chestnut soil, and very few were found in areas with subalpine meadow soil. Grasshoppers were almost absent from other types of soil, such as aeolian soil. In addition, analysis of 14 soil samples collected in the study area revealed that a soil moisture content between 18 and 32 g kg−1 coincided spatially with a higher density of grasshoppers, with the grasshopper density averaging 15 head m−2. In areas with a soil moisture content above 42 g kg−1 or below 10 g kg−1, grasshopper density dropped to less than 5 head m−2. These indicated that for the study area, soils with very high or very low moisture contents were not conducive to the survival of grasshoppers.

Network-constrained and category-based point pattern analysis for Suguo retail stores in Nanjing, China

The distribution of many geographical objects and events is affected by the road network; thus, network-constrained point pattern analysis methods are helpful to understand their space structures and distribution patterns. In this study, network kernel density estimation and network K-function are used to study retail service hot-spot areas and the spatial clustering patterns of a local retail giant (Suguo), respectively, in Nanjing city. Stores and roads are categorized to investigate the influence of weighting different categories of point events and network on the analysis. In addition, the competitive relation between Suguo and foreign-brand retail chains was revealed. The comprehensive analysis results derived from the combination of the first-order and second-order properties can be further used to examine the reasonability of the existing store distribution and optimize the locational choice of new stores.

Spatial Distribution Characteristics of Healthcare Facilities in Nanjing: Network Point Pattern Analysis and Correlation Analysis

The spatial distribution of urban service facilities is largely constrained by the road network. In this study, network point pattern analysis and correlation analysis were used to analyze the relationship between road network and healthcare facility distribution. The weighted network kernel density estimation method proposed in this study identifies significant differences between the outside and inside areas of the Ming city wall. The results of network K-function analysis show that private hospitals are more evenly distributed than public hospitals, and pharmacy stores tend to cluster around hospitals along the road network. After computing the correlation analysis between different categorized hospitals and street centrality, we find that the distribution of these hospitals correlates highly with the street centralities, and that the correlations are higher with private and small hospitals than with public and large hospitals. The comprehensive analysis results could help examine the reasonability of existing urban healthcare facility distribution and optimize the location of new healthcare facilities.

Generation of Pixel-Level SAR Image Time Series Using a Locally Adaptive Matching Technique

Synthetic Aperture Radar (SAR) image time series play an important role in many applications. To construct pixel-level SAR image time series, we propose a locally adaptive image matching technique for the high-precision geometric registration of SAR images. The basic idea is to adapt the local characteristics of ground objects during the process of image registration. Then, by analyzing the spatial distribution of the error of each matched pair in the previous iteration, local areas are divided based on the local clustering of pairs with large errors. A new polynomial is then used to satisfy the local geometric constraint. Based on this proposed matching technique, we introduce a pixel-level SAR image time series modeling method. The experimental results show that the average geometric error of corresponding pixels in this algorithm is 0.073 pixels, while that of the NEST software is 0.242 pixels. The Pearson correlation coefficients of 20 pixels’ time series are above 0.85, indicating that the series bears high curve similarity and geometric precision, which suggests the proposed technique can provide high-quality SAR image time series.

Novel parallel algorithm for constructing Delaunay triangulation based on a twofold-divide-and-conquer scheme

To increase the efficiency when processing large data sets, a novel parallel algorithm is proposed for constructing the Delaunay triangulation of a planar point set based on a twofold-divide-and-conquer scheme. This algorithm automatically divides the planar point set into several non-overlapping subsets along the x-axis and y-axis directions alternately, according to the number of points and their spatial distribution. Next, the Guibas–Stolfi divide-and-conquer algorithm is applied to construct Delaunay sub-triangulations in each subset. Finally, the sub-triangulations are merged based on the binary tree. All three sequential steps are processed using multitasking parallel technology. Our results show that the proposed parallel algorithm is efficient for constructing the Delaunay triangulation with a good speed-up.

GIS-based method of delimitating trade area for retail chains

Delimitating trade area with accuracy is a major concern for retail and service companies who want to adapt their marketing strategy to be competitive in todays highly competitive chain industry. A collection of methods have been proposed to delimitating trade area, but either some are too simple or the necessary data are not available. In this article, a geographic information system-based method for precisely delimitating trade area is proposed. It is based on the fact that the social and economic activities are actualized along the street networks. The trade area can be delineated in terms of time distance with high precision taking the practical traffic situations into account. In addition, centered on this method of delimitating trade area, a wide array of functions can be extended to support both operational day-to-day and long-term strategic decision making for retailers.

Grid interpolation algorithm based on nearest neighbor fast search

The nearest neighbor search algorithm is one of the major factors that influence the efficiency of grid interpolation. This paper introduces a KD-tree that is a two-dimensional index structure for use in grid interpolation. It also proposes an improved J-nearest neighbor search strategy based on “priority queue” and “neighbor lag” concepts. In the strategy, two types of J-nearest neighbor search algorithms can be used; these algorithms correspond to the consideration of a fixed number of points and a fixed radius. By using the KD-tree and proposed strategy, interpolation can be performed with methods such as Inverse Distance Weighting and Kriging. Experimental results show that the proposed algorithms has high operating efficiency, especially when the data amount is enormous, and high practical value for increasing the efficiency of grid interpolation.