基于语义优先的土地利用图斑综合的研究——以大连旅顺口区北海街道为例
作者简介:杨 俊(1978-),男,湖北孝昌人,博士,副教授,主要从事区域地表过程、城市人居环境与地理信息系统应用。E-mail:lnu_hsrc@126.com
收稿日期: 2012-12-05
要求修回日期: 2013-05-20
网络出版日期: 2013-08-20
基金资助
国家自然科学基金项目(40971299,41171137) 和教育部人文社科项目(09YJC790135)资助
Generalization of Land Use Patch Based on Semantic Priority—A Case of Beihai Sub-district of Lushun Port
Received date: 2012-12-05
Request revised date: 2013-05-20
Online published: 2013-08-20
Copyright
土地利用图斑综合是实现其多尺度表达的关键技术,主要包括相邻图斑的融合、相离且语义邻近图斑的聚合、狭长图斑的综合以及要素边线的化简等。通过以下4种方法实现图斑综合:① 以要素间语义相似度为先决条件并顾及空间拓扑关系建立邻近度分析模型,通过模型检索要素的最邻近图斑实现数据融合;② 以相离要素缓冲区交集中的节点建立多边形要素填补桥梁区域实现要素聚合;③ 以相邻要素的缓冲区将狭长区域剖分,并融合到周围要素的方法来实现狭长图斑的综合;④ 以Douglas-Peucker算法实现要素边线化简。上述方法对普通图斑及特征图斑均提出了各自适应的综合算法。实验结果表明上述模型及算法能最大限度的保证各地类面积平衡、也能有效控制要素变形、合理压缩数据并保证地图简洁、美观,保证自动化制图综合的简捷高效。
杨俊 , 席建超 , 孔凡强 , 葛全胜 , 李雪铭 , 高航 . 基于语义优先的土地利用图斑综合的研究——以大连旅顺口区北海街道为例[J]. 地理科学, 2013 , 33(8) : 949 -956 . DOI: 10.13249/j.cnki.sgs.2013.08.949
Generalization of land use patch is the key technology in the multi-scale representation of land-use data. It consists essentially of the combination of adjacent patch, the aggregation of separated semantic similar patches, the decomposition and merging of strip patches and the simplification of boundaries of features, etc. This article has gain it from the following four methods: 1) the proximity analysis model was built based on the semantic similarity of features and the spatial topological relationship. and then the nearest adjacent patch is retrieved through it to complete generalization;2) Polygon elements are established to fill bridge areas using the nodes that contained separated factor buffer intersections;3) The narrow regions will be split by its adjacent features buffer, and then merged with the surrounding features to integrate the regions; 4) The Douglas-Peucker algorithm is used to conduct line simplification. The above methods put respective synthesis algorithm for general and particular patches. And the experiment result from the Beihai sub-district of Lushun Port, Dalian City, shows that the models and algorithms can maximally balance area of various types of land, effectively control elements distortion, reasonably compress data to ensure the appearance of map and easiness of automatic cartography generalization.
Fig.1 Hierarchical tree of land use type图1 土地类型语义层次树 |
Fig.2 An example of neighborhood degree图2 图斑邻近度示例 |
Fig.3 Workflow of the model图3 模型流程图 |
Fig.4 Process of land use patch aggregation图4 图斑聚合过程 |
Fig.5 Process of handling strip feature图5 狭长要素处理过程 |
Fig.6 Douglas-Peucker algorithm procedure图6 Douglas-Peucker算法过程 |
Table 1 The change of area before and after generalization表1 综合前后地类面积变化 |
地类(二级) | 1∶10000 面积(m2) | 1∶100000 面积(m2) | 变化率 |
---|---|---|---|
城乡建设用地 | 3591616.8 | 3581008.5 | 0.3% |
耕地 | 7310081.2 | 7263653.0 | 0.6% |
交通水利用地 | 344421.8 | 331129.5 | 3.9% |
林地 | 10688829.7 | 10975084.1 | -2.7% |
牧草地 | 4378.3 | 10685.9 | -144.1% |
其他建设用地 | 41656.6 | 41656.6 | 0.0% |
其他农用地 | 1478613.1 | 1405879.1 | 4.9% |
水域 | 857290.4 | 866583.7 | -1.1% |
园地 | 2170817.9 | 2137894.1 | 1.5% |
自然保留地 | 3091709.5 | 2965840.8 | 4.1% |
Fig.7 The generalization result of Beihai Sub-district data图7 北海街道数据综合前后对比 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
/
〈 |
|
〉 |