自然资源学报 ›› 2020, Vol. 35 ›› Issue (4): 963-976.doi: 10.31497/zrzyxb.20200417

• 研究论文 • 上一篇    下一篇

顾及空间自相关性的高分遥感影像中建设用地的变化检测

张涛1, 方宏2, 韦玉春2,3,4, 胡祺5, 徐晗泽宇2   

  1. 1. 南京市规划和自然资源局,南京 210005;
    2. 南京师范大学地理科学学院,南京 210023;
    3. 南京师范大学虚拟地理环境教育部重点实验室,南京 210023;
    4. 江苏省地理信息资源开发与利用协同创新中心,南京 210023;
    5. 南京市城市地下管线数字化管理中心,南京 210029
  • 收稿日期:2019-02-27 出版日期:2020-04-28 发布日期:2020-04-28
  • 通讯作者: 韦玉春(1965- ),男,河北玉田人,博士,教授,博士生导师,研究方向为环境遥感与地理建模。E-mail: weiyuchun@njnu.edu.cn
  • 作者简介:张涛(1962- ),男,江苏南京人,本科,高级工程师,研究方向为测绘地理信息。E-mail: 93126347@qq.com
  • 基金资助:
    国家自然科学基金项目(41471283)

Detection of the construction land change in fine spatial resolution remote sensing imagery coupling spatial autocorrelation

ZHANG Tao1, FANG Hong2, WEI Yu-chun2,3,4, HU Qi5, XU Han-ze-yu2   

  1. 1. Nanjing Municipal Bureau of Planning and Natural Resources, Nanjing 210005, China;
    2. School of Geography, Nanjing Normal University, Nanjing 210023, China;
    3. Key Laboratory of Virtual Geographic Environment (Nanjing Normal University), Ministry of Education, Nanjing 210023, China;
    4. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Jiangsu Provincial Key Laboratory of Geographical Environment Evolution, Nanjing 210023, China;
    5. Nnajing Underground Pipeline Data-managing Center, Nanjing 210029, China
  • Received:2019-02-27 Online:2020-04-28 Published:2020-04-28

摘要: 在中国城市化的进程中,建设用地常常连片分布,其开发在空间上具有明显的聚集性,表现出较强的空间自相关性,这在高空间分辨率的遥感图像中更加明显。基于2016年、2017年两期南京市域内的北京2号3.2 m多光谱遥感图像,对比分析了引入变化向量的空间自相关指数作为图像特征后建设用地遥感变化检测的性能。首先提取遥感图像光谱变化向量的局部G指数、Moran's I和Geary's C三个典型空间自相关指数,然后确定适用于变化检测的最优空间间隔(Lag)范围和最优自相关指数。结果表明:(1)光谱变化向量在空间上具有显著的正相关性。(2)全局Moran's I和半方差函数相结合可以确定最优的Lag范围。(3)在光谱变化向量的基础上加入局部G指数和局部Moran's I能够提高检测精度,F1分数表明前者优于后者。(4)在光谱变化向量的基础上加入最优Lag范围内的局部G指数作为附加图像特征,F1分数比只使用光谱变化向量提高了20%以上。融合空间自相关信息,特别是多尺度局部G指数作为遥感图像特征可有效地提高连片区域建设用地的变化检测精度。

关键词: 局部G指数, 土地覆盖, 遥感变化检测, 建设用地, 空间自相关

Abstract: In the urbanization of China, construction land is generally distributed as a continuous area, and its change shows distinct spatial aggregation leading to the strong spatial autocorrelation, which is more obvious in remote sensing imagery with a fine spatial resolution. Based on the TripleSat-2 multi-spectral remote sensing images covering Nanjing city in 2016 and 2017, the paper compares and analyzes the performance of remote sensing change detection of construction land after we introduced the spatial autocorrelation index of the change vector as the image feature. Firstly, the three typical spatial autocorrelation indices of local G, Moran's I and Geary's C are extracted, and then the optimal spatial range of Lag and the optimal autocorrelation index suitable for the change detection are determined. The results showed that: (1) The spectral change vector had significant positive correlation. (2) The optimal range of Lag can be determined by global Moran's I and semi-variance. (3) We used the local G and local Moran's I with the spectral change vectors to increase the F1 score of the change detection, and the local G showed the better performance. (4) Using the local G with the optimal range of Lag as additional image features, the F1 scores were 20% higher than that using only the spectral change vectors. Fusing local spatial autocorrelation information especially of the multi-scale local G as the additional image features can effectively improve the change detection accuracy of construction land in remfote sensing imagery.

Key words: spatial autocorrelation, local G statistic, land cover, remote sensing change detection, construction land