Recognition of building group patterns is of great significance for understanding and modeling the urban space. However, many current methods cannot fully utilize spatial information and have trouble efficiently dealing with topographic data with high complexity. The design of intelligent computational models that can act directly on topographic data to extract spatial features is critical. To this end, we propose a novel deep neural network based on graph convolutions to automatically identify building group patterns with arbitrary forms. The method first models buildings by a general graph, and then the neural network simultaneously learns the structural information as well as vertex attributes to classify building objects. We apply this method to real building data, and the experimental results show that the proposed method can effectively capture spatial information to make more accurate predictions than traditional methods.

In recent years, main cities in China have been suffering from hazy weather, which is gaining great attention among the public, government managers and researchers in different areas. Many studies have been conducted on the topic of urban air quality to reveal different aspects of the air quality problem in China. This paper focuses on the visualization problem of the big air quality monitoring data of all main cities on a nationwide scale. To achieve the intuitive visualization of this data set, this study develops two novel visualization tools for multi-granularity time series visualization (timezoom.js) and a dynamic symbol declutter map mashup layer for thematic mapping (symadpative.js). With the two invented tools, we develops an interactive web map visualization application of urban air quality data of all main cities in China. This application shows us significant air pollution findings at the nationwide scale. These results give us clues for further studies on air pollutant characteristics, forecasting and control in China. As the tools are invented for general visualization purposes of geo-referenced time series data, they can be applied to other environmental monitoring data (temperature, precipitation, etc.) through some configurations.

Line simplification is an important component of map generalization. In recent years, algorithms for line simplification have been widely researched, and most of them are based on vector data. However, with the increasing development of computer vision, analysing and processing information from unstructured image data is both meaningful and challenging. Therefore, in this paper, we present a new line simplification approach based on image processing (BIP), which is specifically designed for raster data. First, the key corner points on a multi-scale image feature are detected and treated as candidate points. Then, to capture the essence of the shape within a given boundary using the fewest possible segments, the minimum-perimeter polygon (MPP) is calculated and the points of the MPP are defined as the approximate feature points. Finally, the points after simplification are selected from the candidate points by comparing the distances between the candidate points and the approximate feature points. An empirical example was used to test the applicability of the proposed method. The results showed that (1) when the key corner points are detected based on a multi-scale image feature, the local features of the line can be extracted and retained and the positional accuracy of the proposed method can be maintained well; and (2) by defining the visibility constraint of geographical features, this method is especially suitable for simplifying water areas as it is aligned with people’s visual habits.

Street patterns reflect the distribution characteristics of a street network and affect the urban structure and human behavior. The recognition of street patterns has been a topic of interest for decades. In this study, a linear tessellation model is proposed to identify the spatial patterns in street networks. The street segments are broken into consecutive linear units with equal length. We define five focal operations using neighborhood analysis to extract the geometric and topological characteristics of each linear unit for the purpose of grid-pattern recognition. These are then classified by Support Vector Machine, and the result is optimized based on Gestalt principles. The experimental results demonstrate that our method is effective for mining grid patterns in a street network.

Unmanned autonomous Air-to-Air Refueling (AAR) capability is the key guarantee to support the distant-field, high-intensity and durable operations of the penetration counterair combat system. In the future, the long-range unmanned reconnaissance and attack platform can reach the maximum flight range requirement through AAR. At present, large transport aircraft platforms in China are still equipped with probe-and-drogue systems, and the refueling mode is gradually changing from manned to unmanned autonomous operation. The docking process is the riskiest and most important part, and there are strict safety, precision, and efficiency requirements for refueling operation, especially during close-distance docking and formation maintenance phases. In this paper, five issues that need to be solved to achieve autonomous AAR docking are summarized. On this basis, five key technology development needs are proposed to solve these engineering issues. Finally, some prospects are given.

The "food desert" problem has been treated under a national strategy in the United States and other countries. At present, there is little research on the phenomenon of "food desert" in China. This study takes Shanghai as the research area and proposes a multiscale analysis method using a linear tessellation model that splits the street network into homogeneous linear units. Firstly, the network kernel density estimation using a linear tessellation model is used to measure the travel-mode-based food accessibility. Considering the actual travel constraints, the GPS trajectory data of four travel modes (walking, bicycle, metro and taxi) are applied to calculate the speed of each linear unit. Secondly, the “food desert” phenomenon in Shanghai are identified combing with the results of the network K-function. Finally, the resident income conditions in different modes are fitted based on the housing price data and the spatial distribution of four "food desert" patterns are detected by the overlay analysis of food accessibility and resident income conditions. The experimental results show that fifty percent of Shanghai is characterized by low food accessibility, and half of these areas are disadvantaged and low-income areas in suburbs, which are the locations experiencing the “food desert” phenomenon. Comparing the results of the proposed method and that of the traditional planar method, the identification results for all modes based on the traditional planar method underestimate the severity of the “food desert”, especially for the bicycle and taxi modes. This study also provides corresponding decision-making reference for the alleviation and resolution of “food desert” issues. Moreover, the proposed method provides a new research perspective for urban research under the street network space.