邻域扩展法辅助的陆地争议区划界
作者简介:冯长强(1988-),男,博士生,主要从事地理信息系统应用、地缘环境分析与建模研究。E-mail: luckystar1682006@126.com
收稿日期: 2017-04-26
要求修回日期: 2017-07-11
网络出版日期: 2017-10-20
基金资助
国家科技支撑计划项目(2012BAK12B00)
国家自然科学基金面上项目(41471336)
地理信息工程国家重点实验室开放研究基金项目(SKLGIE2015-M-4-4)
Delimitation of Land Disputed Area by Neighborhood Expansion Method
Received date: 2017-04-26
Request revised date: 2017-07-11
Online published: 2017-10-20
Copyright
已有计算机辅助划界方法难以充分维护单方资源利益且效率不够理想,本文基于利益密度及区域完整性视角提出一种邻域扩展法辅助的陆地争议区划界方法。首先,利用六角格构建争议区环境模型,计算并处理六角格的利益密度;其次,以利益密度及区域完整性为导向,利用邻域扩展法对争议区进行初始划分;再次,通过扩展单源六角格最优路径算法求解通往飞地的最佳路径,并基于一定原则对飞地归属进行再次划分;最后,辅以邻域扩展法对争议区进行精确分割。采用模拟数据、不同格网尺寸及不同面积比例进行实验,结果表明本文方法具有如下特点:① 能够对双方既定的必争区域及不可穿越区域进行正确划分;② 能够依双方约定面积比例对争议区进行快速精确分割;③ 能够充分维护单方资源利益优势,验证了本文方法的有效性及可靠性。
关键词: 谈判划界; 利益密度; 区域完整性; 邻域扩展法; 多源六角格最优路径算法
冯长强 , 华一新 , 张晓楠 , 曹一冰 , 武丽丽 , 崔虎平 . 邻域扩展法辅助的陆地争议区划界[J]. 地球信息科学学报, 2017 , 19(10) : 1306 -1314 . DOI: 10.3724/SP.J.1047.2017.01306
Now, there are still many land border territory disputes causing local wars or unstableness between countries. Under the current international situation, negotiation is the most effective method of solving territory disputes and it needs lots of demarcation technology. Current demarcation methods cannot fully safeguard unilateral resource interest and it costs lots of time. Thus, we proposed a new solution using neighborhood expansion method (NEM). Firstly, the disputed area was split using hexagon where related information such as resource reserve was mapped, and benefit density (BD), which is the comprehensive evaluation value of related resources in each grid, was calculated and disposed. Secondly, the disputed area was initially divided using NEM under the guide of regional integrity, BD and bilateral agreed area ratio, where most hexagons with higher BD were assigned to the related country. Thirdly, the single-source optimal path algorithm based on hexagon was improved to solve the optimal path from non-enclave to enclave caused during the initial segmentation of disputed area. The ascriptions of all the enclaves were determined once again based on some rules. Finally, the integrity of unilateral region was optimized, the gap between the unilateral area and the agreed area was reduced to the extent smaller than the area of single complete grid using NEM. The disputed zone was split accurately according to the agreed area ratio. Tests were made to compare our method with the other one using genetic algorithm based on simulated data, different hexagon sizes and agreed area ratios. The results indicated that our method owns the following characteristics: (1) it can correctly assign bilateral agreed never-lost regions and impenetrable areas like ethnic settlements; (2) the disputed area can be divided fast and precisely according to agreed area ratio; (3) it can fully safeguard unilateral resource interest. These features indicate that our method is effective and reliable and it can provide important reference and guide for one-side delimitation.
Fig. 1 Three types of regular grids and their adjacent grids图1 3种规则格网及其邻接格网示意图 |
Fig. 2 Distribution of benefit density in disputed area图2 争议区利益密度分布 |
Fig. 3 Narrow area with single-grid width图3 单格网宽度狭长区域示意图 |
Fig. 4 Initial division of disputed area and optimal path from enclave to its country图4 争议区初始分割及飞地通路寻优结果 |
Fig. 5 Searching optimal paths from one-side enclaves to its country图5 甲方飞地通路寻优流程 |
Fig. 6 Accurate segmentation of grids图6 格网精确分割示意图 |
Tab. 1 Weight of resource importance表1 资源重要性权值 |
类别 | 水资源 | 石油 | 天然气 | 稀有金属 |
---|---|---|---|---|
权值 | 0.1053 | 0.3628 | 0.2446 | 0.2873 |
Tab. 2 Comparison between results of two methods表2 2种方法结果对比 |
格网尺寸/km | 甲乙约定面积比例 | 遗传诱导法 | 本文方法 | |||||
---|---|---|---|---|---|---|---|---|
甲乙实际所 得面积比例 | 甲乙实际所得利益比例 | 耗时/s | 甲乙实际所得面积比例 | 甲乙实际所得利益比例 | 耗时/s | |||
2 | 1:2 | 1.09:2 | 1.21:2 | 33 | 1:2 | 1.92:2 | 7 | |
1:1 | 1.05:1 | 1.31:1 | 27 | 1:1 | 2.31:1 | 6 | ||
2:1 | 2.05:1 | 2.88:1 | 29 | 2:1 | 5.17:1 | 8 | ||
3 | 1:2 | 1.06:2 | 1.13:2 | 16 | 1:2 | 1.90:2 | 5 | |
1:1 | 1.03:1 | 1.27:1 | 14 | 1:1 | 2.30:1 | 4 | ||
2:1 | 2.04:1 | 2.57:1 | 17 | 2:1 | 5.08:1 | 5 | ||
4 | 1:2 | 1.07:2 | 1.05:2 | 9 | 1:2 | 1.83:2 | 4 | |
1:1 | 1.06:1 | 1.10:1 | 8 | 1:1 | 2.15:1 | 3 | ||
2:1 | 2.04:1 | 2.18:1 | 6 | 2:1 | 4.53:1 | 3 |
Fig. 7 Experiment results图7 实验结果 |
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] |
[
|
/
〈 | 〉 |