顾及多特征约束的建筑物组合化简方法
马京振(1993— ),男,山东平邑人,博士,讲师,主要从事数字地图制图技术研究。E-mail: zb50mjz@163.com |
收稿日期: 2022-06-26
修回日期: 2022-07-28
网络出版日期: 2023-04-19
基金资助
国家自然科学基金项目(42101454)
河南省中原学者资助项目(202101510001)
智慧中原地理信息技术河南省协同创新中心和时空感知与智能处理自然资源部重点实验室基金资助项目(212102)
Combined Building Simplification Approach Considering Multi-Feature Constraints
Received date: 2022-06-26
Revised date: 2022-07-28
Online published: 2023-04-19
Supported by
National Natural Science Foundation of China(42101454)
The Fund Project of ZhongYuan Scholar of Henan Province(202101510001)
The Joint Fund of Collaborative Innovation Center of Geo-Information Technology for Smart Central Plains, Henan Province and Key Laboratory of Spatiotemporal Perception and Intelligent processing, Ministry of Natural Resources(212102)
建筑物是大比例尺地图中的重要要素,其自动化简是制图综合领域的重要研究内容。针对当前化简方法仅侧重于某一方面的问题,本文提出一种顾及多特征约束的建筑物组合化简方法。首先,对建筑物进行冗余点删除、尖角去除等预处理操作;然后根据化简阈值筛选出需要化简的短边,并对其局部结构进行自动分类识别;然后,充分考虑建筑物的位置、面积、方向和形状等特征约束的影响,针对不同的结构类型采用不同的化简操作,在化简的过程中采用阈值逐渐增大的渐进式确定方法,直到满足化简要求为止;最后,从多个特征角度对化简结果进行评估,对不合理的化简实体进行调整修改。以OpenStreetMap数据为例进行实验,并与其它化简方法进行对比,实验结果表明,本文所提的方法可以有效保持建筑物的面积、形状、直角等基本特征,可以实现建筑物的自动连续化简,具备一定的通用性。
马京振 , 周炤 , 张付兵 . 顾及多特征约束的建筑物组合化简方法[J]. 地球信息科学学报, 2023 , 25(2) : 277 -287 . DOI: 10.12082/dqxxkx.2023.220381
Settlement generalization is an important step for scale transformation, data fusion and integration, data mining, etc. The methods of settlement generalization include selection, simplification, dimension reduction, merging, and typification, among which simplification has been commonly used method. In large scale data, settlement usually exists in the form of building polygon, so the building simplification has always been the focus of settlement generalization. The essence of building simplification is to delete or edit the points that constitute the boundary, and extract the original figure as a simple polygon, while keeping the basic features of the building. In view of the problem that the current simplification method only focuses on one aspect, this paper proposes a combined building simplification method considering multi-feature constraints. Firstly, redundant points and sharp corners are deleted. Then the short edges that need to be simplified are screened according to the simplification threshold, and their local structures are automatically classified and recognized. Furthermore, the influence of the characteristic constraints such as the location, area, direction, and shape of the building is fully considered, and different simplification operations are adopted for different structural types. In the process of simplification, a progressive method with gradually increasing threshold value is adopted until the simplification requirements are met. Finally, the simplification results are evaluated from multiple features and the unreasonable simplification entities are adjusted and modified. Taking OpenStreetMap data as an example, and compared with other simplification methods, the experimental results show that the proposed method can effectively keep the basic features of buildings, such as area, shape, and orthogonal angle, and realize automatic and continuous simplification of buildings, with certain universality.
表1 与其他方法化简结果对比Tab. 1 Simplification results comparison with other methods |
尺寸 | 方法 | 位置 | 面积 | 方向 | 形状 |
---|---|---|---|---|---|
1:25 000 | 本文方法 | 0.9871 | 0.9873 | 0.9741 | 0.9020 |
邻近四点方法 | 0.9845 | 0.9930 | 0.9748 | 0.5887 | |
递归回归方法 | 0.9833 | 0.9597 | 0.9760 | 0.4331 |
表2 渐进式化简结果统计Tab. 2 Results statistics of progressive simplification |
化简阈值/m | 2.5 | 2.81 | 3.29 | 4.67 | |
最短边长度/m | 2.35 | 2.65 | 3.04 | 3.92 | 7.5 |
是否满足要求 | 否 | 否 | 否 | 否 | 是 |
建筑物表达 | ![]() | ![]() | ![]() | ![]() | ![]() |
[1] |
晏雄锋. 深度卷积学习支持下的建筑物模式分析[D]. 武汉: 武汉大学, 2019.
[
|
[2] |
高晓蓉, 闫浩文, 禄小敏, 等. 利用“计算区”进行建筑物短边结构识别和渐进式化简[J]. 武汉大学学报·信息科学版, 2021, 46(11):1782-1790.
[
|
[3] |
|
[4] |
尹烁, 闫小明, 晏雄锋. 基于特征边重构的建筑物化简方法[J]. 测绘学报, 2020, 49(6):703-710.
[
|
[5] |
|
[6] |
许文帅, 龙毅, 周侗, 等. 基于邻近四点法的建筑物多边形化简[J]. 测绘学报, 2013, 42(6):929-936.
[
|
[7] |
连超, 李成名, 殷勇, 等. 改进的邻近四点法建筑物多边形化简算法[J]. 测绘科学, 2016, 41(2):26-29,75.
[
|
[8] |
|
[9] |
刘鹏程, 艾廷华, 邓吉芳. 基于最小二乘的建筑物多边形的化简与直角化[J]. 中国矿业大学学报, 2008, 37(5):699-704.
[
|
[10] |
|
[11] |
|
[12] |
|
[13] |
刘鹏程, 艾廷华, 胡晋山, 等. 基于原型模板形状匹配的建筑多边形化简[J]. 武汉大学学报·信息科学版, 2010, 35(11):1369-1372.
[
|
[14] |
陈文瀚, 龙毅, 沈婕, 等. 利用约束D-TIN进行建筑物多边形凹部结构识别与渐进式化简[J]. 武汉大学学报·信息科学版, 2011, 36(5):584-587,592.
[
|
[15] |
|
[16] |
杨好, 何宗宜, 陈华洋, 等. 顾及路网结构保持的城镇居民地自动综合[J]. 武汉大学学报·信息科学版, 2018, 43(6):965-970.
[
|
[17] |
|
[18] |
晏雄锋, 艾廷华, 杨敏, 等. 地图空间形状认知的自编码器深度学习方法[J]. 测绘学报, 2021, 50(6):757-765.
[
|
[19] |
|
[20] |
王辉连, 武芳, 张琳琳, 等. 数学形态学和模式识别在建筑物多边形化简中的应用[J]. 测绘学报, 2005, 34(3):269-276.
[
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
周培德. 计算几何算法分析与设计[M]. 北京: 清华大学出版社, 2000.
[
|
[26] |
|
[27] |
|
/
〈 |
|
〉 |