标题: A tile-map-based method for the typification of artificial polygonal water areas considering the legibility

作者: Shen, YL (Shen, Yilang); Ai, TH (Ai, Tinghua); Li, JZ (Li, Jingzhong); Wang, L (Wang, Lu); Li, WD (Li, Wende)

来源出版物: COMPUTERS & GEOSCIENCES 卷: 143 文献号: 104552 DOI: 10.1016/j.cageo.2020.104552 出版年: OCT 2020

摘要: A tile map in image format is one of the most important tools that people can use to acquire multiscale geographic information on the Internet. Traditional methods of typification in map generalization are used to handle traditional vector-based buildings and linear drainages such as rivers and ditches. In this paper, a new raster-tile-based method called artificial pond typification (APT) is developed for the typification of artificial water areas while maintaining the original distribution characteristics and reducing the number of water areas. First, combining the second-order neighborhoods of superpixels and median filtering, the water areas are grouped at different levels of detail. Then, three different types of superpixel generation algorithms, including simple linear iterative clustering (SLIC), linear spectral clustering (LSC) and superpixel extraction via energy-driven sampling (SEEDS), are applied to generate new typified positions. Finally, different strategies, such as maximum area and global feature strategies, are designed to reconstruct the shapes of water areas. To test the proposed APT method, the map tiles from the Baidu map in China are used as the raw experimental data. The experimental results show that the proposed APT method can be effectively used for the typification of artificial ponds at different levels of detail while gradually removing the details of boundaries as the scale changes. In addition, different typification strategies provide various cartographic alternatives in different cases.

入藏号: WOS:000571449700001

语言: English

文献类型: Article

作者关键词: Water area typification; Tile map; Superpixel segmentation; Map generalization; Raster

地址: [Shen, Yilang; Ai, Tinghua; Li, Jingzhong; Wang, Lu; Li, Wende] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China.

通讯作者地址: Ai, TH (通讯作者)，Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China.

电子邮件地址: tinghuaai@whu.edu.cn

影响因子：2.991