Spatial Stratification Mode and Differentiation Evaluation for Accuracy Assessment of Remote Sensing Classification

被引：0
|
作者：
Wu Y. [1 ]
Dong S. [2 ,3 ]
Xiao C. [2 ,3 ]
Li X. [1 ]
Pan Y. [2 ,3 ]
Niu C. [4 ]
机构：
[1] College of Information Science and Engineering, Shandong Agricultural University, Tai'an
[2] Beijing Research Center for Information Technology in Agriculture, Beijing
[3] National Engineering Research Center for Information Technology in Agriculture, Beijing
[4] Geological Surveying and Mapping Institute of Shandong Province, Ji'nan
来源：
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery | 2021年 / 52卷 / 08期
关键词：
Accuracy assessment; Geographic detector; Remote sensing classification; Sampling; Spatial differentiation; Spatial stratification;
D O I：
10.6041/j.issn.1000-1298.2021.08.014
中图分类号：
学科分类号：
摘要：
In order to evaluate the sampling accuracy of remote sensing classification, taking Beijing-Tianjin-Hebei remote sensing data products with different spatial resolutions as an example, the internal and boundary objects of remote sensing image were firstly divided based on land use types, and different spatial stratification modes were constructed, including without considering internal and boundary objects, considering boundary objects, considering internal objects, both considering internal and boundary objects. Secondly, direct land use types, image eight-neighborhoods algorithm, multi-scale spatial differentiation method, coupling method of image eight-neighborhoods and multi-scale spatial differentiation were adopted for spatial stratification, respectively. Finally, a comparative experiment of K-means clustering method was set up, and the differentiation effects of different spatial stratification modes were quantitatively evaluated based on geographic detector. The results suggested that the mean and standard deviation of q of the corresponding five groups of sampling sites for the spatial stratification modes of without considering internal and boundary (6 strata), considering boundary (12 strata), considering internal (18 strata), both considering internal and boundary objects (24 strata), K-means (12, 18, 24 strata) in the Beijing-Tianjin-Hebei regions were 0.252±0.022 66, 0.259±0.022 45, 0.321±0.019 01, 0.318±0.018 06, 0.269±0.006 98, 0.304±0.010 56, and 0.317±0.011 25, respectively. Internal objects played a leading role for spatial stratification differentiation and boundary objects slightly improved spatial stratification differentiation, and the number of strata also affected the differentiation of spatial stratification. The research results can better understand the contributions of internal and boundary objects on improving spatial stratification differentiation, and had a certain research value and guiding significance for developing spatial stratification methods with high differentiation. © 2021, Chinese Society of Agricultural Machinery. All right reserved.
引用
收藏
页码：147 / 153
页数：6
相关论文
共 28 条
  • [21] WAN Changjun, WU Xiaodan, LIN Xingwen, Impact of spatial and temporal scales of remote sensing data on the spatiotemporal change in geographic elements, Journal of Remote Sensing, 23, 6, pp. 1064-1077, (2019)
  • [22] ZENG Hongwei, LI Lijuan, Accuracy validation of TRMM 3B43 data in Lancang River basin, Acta Geographica Sinica, 66, 7, pp. 994-1004, (2011)
  • [23] WANG J F, LI X H, CHRISTAKOS G, Et al., Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun region, China, International Journal of Geographical Information Science, 24, 1, pp. 107-127, (2010)
  • [24] WANG J F, HU Y., Environmental health risk detection with GeogDetector, Environmental Modelling & Software, 33, pp. 114-115, (2012)
  • [25] WU Bingfang, ZHANG Miao, Remote sensing: observations to data products, Acta Geographica Sinica, 72, 11, pp. 2093-2111, (2017)
  • [26] DONG S W, GAO B B, PAN Y C, Et al., Assessing the suitability of FROM-GLC10 data for understanding agricultural ecosystems in China: Beijing as a case study, Remote Sensing Letters, 11, 1, pp. 11-18, (2020)
  • [27] WU Tingtian, DING Shan, CHEN Zongzhu, Et al., Dynamic analysis of mangrove wetlands based on LUCC and landscape pattern change in Dongzhai Port, Forest Research, 33, 5, pp. 154-162, (2020)
  • [28] LI Guoqing, HUANG Jinghua, LIU Guan, Et al., A study of the landscape fragmentations of land cover structure based on Landsat 8 remote sensing image: a case study of Mata watershed in Yan'an, Shaanxi Province, Remote Sensing for Land & Resources, 32, 3, pp. 121-128, (2020)
123