土地覆盖变化信息自适应抽样策略及其精度评估

梅莹莹; 张景雄

doi:10.11947/j.AGCS.2018.20170262

测绘学报 >

2018 , Vol. 47 >Issue 5: 644 - 651

DOI: https://doi.org/10.11947/j.AGCS.2018.20170262

地图学与地理信息

土地覆盖变化信息自适应抽样策略及其精度评估

梅莹莹 ,
张景雄

展开

1. 武汉大学遥感信息工程学院, 湖北武汉 430079;
2. 武汉大学测绘学院, 湖北武汉 430079

梅莹莹(1990-),女,博士,研究方向为地理类别信息与分析。E-mail:myy2014@whu.edu.cn

收稿日期: 2017-05-17

修回日期: 2018-02-26

网络出版日期: 2018-06-01

基金资助

国家自然科学基金（41471375）

收起

An Adaptive Sampling Strategy for Land Cover Change Information and Its Accuracy Characterization

MEI Yingying ,
ZHANG Jingxiong

Expand

1. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China;
2. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China

Received date: 2017-05-17

Revised date: 2018-02-26

Online published: 2018-06-01

Supported by

The National Natural Science Foundation of China (No.41471375)

Fold

摘要

提出了一种面向土地覆盖变化信息局域精度评估的自适应型抽样策略。结合待研究地图的土地覆盖变化信息局部特征（如土地覆盖变化类别、斑块大小、异质性和优势度），探讨与土地覆盖变化信息精度显著相关的协变量，以预测精度的标准误差作为判断标准，识别需要提高精度预测结果可靠性的区域，以自适应地和逐步定位的方式进行样本采集。基于武汉地区的精度评价结果，自适应的抽取增加100个训练样本使得预测精度的确定系数提高了50.66%，而简单随机抽取的增加样本使得预测精度的确定系数提高了17.22%。试验表明，自适应型抽样策略能显著提高土地覆盖变化信息局域精度预测的抽样效益，减少预测精度的不确定性。模型选择的结果表明，土地覆盖变化类别和优势度指数是最优的协变量组合。

关键词： 土地覆盖变化; 精度; 逻辑回归; 采样

本文引用格式

梅莹莹 , 张景雄 . 土地覆盖变化信息自适应抽样策略及其精度评估[J]. 测绘学报, 2018 , 47(5) : 644 -651 . DOI: 10.11947/j.AGCS.2018.20170262

Abstract

An adaptive sampling strategy is proposed for location-specific characterization of accuracy in land cover change information. The local accuracy characterization strategy was established based on local patterns of land cover change maps (e.g land cover change classes, patch size, heterogeneity and dominance), which include exploring covariates significantly relate to accuracy. Standard error of prediction accuracy was used for identifying the area which needs to improve the reliability of prediction accuracy and locating samples adaptively and progressively. The performance of different sampling methods for accuracy prediction was evaluated at the same testing samples in Wuhan. It was indicated that 100 more training samples selected by adaptive sampling strategy lead to about 50.66% increase in prediction accuracy, as measured by sums-of-squares. In comparison, for random sampling, the same increase in training sample size led to about 17.22% increase in prediction accuracy, as measured by sums-of-squares. This confirms that adaptive sampling strategy improves the sampling efficiency while reduces the uncertainty in local accuracies prediction. Model selection reveals that land cover change classes and dominance are the highest significant covariates.

Key words： land cover change; accuracy; logistic regression; sampling

参考文献

[1] 陈军, 陈利军, 李然, 等. 基于-GlobeLand30的全球城乡建设用地空间分布与变化统计分析[J]. 测绘学报, 2015, 44(11):1181-1188. DOI:10.11947/j.AGCS.2015.20140677. CHEN Jun, CHEN Lijun, LI Ran, et al. Spatial Distribution and Ten Years Change of Global Built-up Areas Derived from GlobeLand30[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(11):1181-1188. DOI:10.11947/j.AGCS.2015.20140677.
[2] FOODY G M. Assessing the Accuracy of Land Cover Change with Imperfect Ground Reference Data[J]. Remote Sensing of Environment, 2010, 114(10):2271-2285.
[3] 明冬萍, 邱玉芳, 周文. 遥感模式分类中的空间统计学应用——以面向对象的遥感影像农田提取为例[J]. 测绘学报, 2016, 45(7):825-833. DOI:10.11947/j.AGCS.2016.20150520. MING Dongping, QIU Yufang, ZHOU Wen. Applying Spatial Statistics into Remote Sensing Pattern Recognition:with Case Study of Cropland Extraction Based on GeOBIA[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(7):825-833. DOI:10.11947/j.AGCS.2016.20150520.
[4] TSENDBAZAR N E, DE BRUIN S D, HEROLD M. Assessing Global Land Cover Reference Datasets for Different User Communities[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 103:93-114.
[5] 陈军, 陈晋, 宫鹏, 等. 全球地表覆盖高分辨率遥感制图[J]. 地理信息世界, 2011, 9(2):12-14. CHEN Jun, CHEN Jin, GONG Peng, et al. Higher Resolution Global Land Cover Mapping[J]. Geomatics World, 2011, 9(2):12-14.
[6] FOODY G M. Status of Land Cover Classification Accuracy Assessment[J]. Remote Sensing of Environment, 2002, 80(1):185-201.
[7] SMITH J H, STEHMAN S V, WICKHAM J D, et al. Effects of Landscape Characteristics on Land-cover Class Accuracy[J]. Remote Sensing of Environment, 2003, 84(3):342-349.
[8] GOODCHILD M F, GUOQING S, SHIREN Y. Development and Test of an Error Model for Categorical Data[J]. International Journal of Geographical Information Systems, 1992, 6(2):87-103.
[9] BURNICKI A C. Modeling the Probability of Misclassification in a Map of Land Cover Change[J]. Photogrammetric Engineering & Remote Sensing, 2011, 77(1):39-49.
[10] PONTIUS R G JR, LI X X. Land Transition Estimates from Erroneous Maps[J]. Journal of Land Use Science, 2010, 5(1):31-44.
[11] ZHANG Jingxiong, ATKINSON P M, GOODCHILD M F. Scale in Spatial Information and Analysis[M]. Boca Raton, FL:CRC Press, 2014.
[12] STEELE B M, WINNE J C, REDMOND R L. Estimation and Mapping of Misclassification Probabilities for Thematic Land Cover Maps[J]. Remote Sensing of Environment, 1998, 66(2):192-202.
[13] 陈志强, 陈健飞. 福建土地利用/覆被人为影响指数及其变化的地统计学分析[J]. 资源科学, 2008, 30(11):1700-1705. CHEN Zhiqiang, CHEN Jianfei. Geostatistical Analysis of Human Impact Indexes for Land Use/Cover Change in Fujian[J]. Resources Science, 2008, 30(11):1700-1705.
[14] 王库. 回归克里格在土壤全氮空间预测上的应用[J]. 中国农学通报, 2013, 29(20):142-147. WANG Ku. Application of Regression Kriging on the Spatial Prediction of Total Soil Nitrogen[J]. Chinese Agricultural Science Bulletin, 2013, 29(20):142-147.
[15] CARMEL Y, FLATHER C H, DEAN D J. Combining Location and Classification Error Sources for Estimating Multi-Temporal Database Accuracy[J]. Photogrammetric Engineering and Remote Sensing, 2001, 67(7):865-872.
[16] KEMPEN B, BRUS D J, HEUVELINK G B M. Soil Type Mapping Using the Generalised Linear Geostatistical Model:A Case Study in a Dutch Cultivated Peatland[J]. Geoderma, 2012, 189-190:540-553.
[17] EMERY X. A Disjunctive Kriging Program for Assessing Point-Support Conditional Distributions[J]. Computers & Geosciences, 2006, 32(7):965-983.
[18] BURNICKI A C. Impact of Error on Landscape Pattern Analyses Performed on Land-cover Change Maps[J]. Landscape Ecology, 2012, 27(5):713-729.
[19] 孟雯, 童小华, 谢欢, 等. 基于空间抽样的区域地表覆盖遥感制图产品精度评估——以中国陕西省为例[J]. 地球信息科学学报, 2015, 17(6):742-749. MENG Wen, TONG Xiaohua, XIE Huan, et al. Accuracy Assessment for Regional Land Cover Remote Sensing Mapping Product Based on Spatial Sampling:A Case Study of Shaanxi Province, China[J]. Journal of Geo-Information Science, 2015, 17(6):742-749.
[20] LECHNER A M, LANGFORD W T, BEKESSY S A, et al. Are Landscape Ecologists Addressing Uncertainty in Their Remote Sensing Data?[J]. Landscape Ecology, 2012, 27(9):1249-1261.
[21] 柏延臣, 王劲峰. 遥感数据专题分类不确定性评价研究:进展、问题与展望[J]. 地球科学进展, 2005, 20(11):1218-1225. BO Yanchen, WANG Jinfeng. Assessment on Uncertainty in Remotely Sensed Data Classification:Progresses, Problems and Prospects[J]. Advances in Earth Science, 2005, 20(11):1218-1225.
[22] 张景雄, GOODCHILD M F. 野外空间采样的渐进式策略[J]. 武汉大学学报(信息科学版), 2008, 33(5):441-445. ZHANG Jingxiong, GOODCHILD M F. Towards Progressive Strategies for Spatial Sampling in the Field[J]. Geomatics and Information Science of Wuhan University, 2008, 33(5):441-445.
[23] MODIS K, PAPAODYSSEUS K. Theoretical Estimation of the Critical Sampling Size for Homogeneous Ore Bodies with Small Nugget Effect[J]. Mathematical Geology, 2006, 38(4):489-501.
[24] 农宇, 王坤, 杜清运. 利用多分类Logistic回归进行土地利用变化模拟——以湖北省嘉鱼县为例[J]. 武汉大学学报(信息科学版), 2011, 36(6):743-746. NONG Yu, WANG Kun, DU Qingyun. Modeling Land Use Change Using Multinomial Logistic Regression[J]. Geomatics and Information Science of Wuhan University, 2011, 36(6):743-746.
[25] HENGL T, HEUVELINK G B M, ROSSITER D G. About Regression-Kriging:from Equations to Case Studies[J]. Computers & Geosciences, 2007, 33(10):1301-1315.
[26] AGRESTI A. Categorical Data Analysis[M]. 3rd ed. Hoboken, NJ:Wiley, 2013:58-65.
[27] GOTWAY C A, STROUP W W. A Generalized Linear Model Approach to Spatial Data Analysis and Prediction[J]. Journal of Agricultural, Biological, and Environmental Statistics, 1997, 2(2):157-178.
[28] ZHANG Jingxiong, MEI Yingying. Integrating Logistic Regression and Geostatistics for User-oriented and Uncertainty-Informed Accuracy Characterization in Remotely-Sensed Land Cover Change Information[J]. ISPRS International Journal of Geo-Information, 2016, 5(7):113.
[29] VAN OORT P A J, BREGT A K, DE BRUIN S, et al. Spatial Variability in Classification Accuracy of Agricultural Crops in the Dutch National Land-cover Database[J]. International Journal of Geographical Information Science, 2004, 18(6):611-626.
[30] SMITH J H, STEHMAN S V, WICKHAM J D. Impacts of Patch Size and Land-cover Heterogeneity on Thematic Image Classification Accuracy[J]. Photogrammetric Engineering and Remote Sensing, 2002, 68(1):65-70.
[31] LU Dengsheng, BATISTELLA M, MORAN E, et al. Fractional Forest Cover Mapping in the Brazilian Amazon with a Combination of MODIS and TM Images[J]. International Journal of Remote Sensing, 2011, 32(22):7131-7149.
[32] 马京振, 孙群, 肖强, 等. 河南省GlobeLand30数据精度评价及对比分析[J]. 地球信息科学学报, 2016, 18(11):1563-1572. MA Jingzhen, SUN Qun, XIAO Qiang, et al. Accuracy Assessment and Comparative Analysis of GlobeLand30 Dataset in Henan Province[J]. Journal of Geo-Information Science, 2016, 18(11):1563-1572.
[33] 杨波, 程泽凯, 秦锋. 用AUC评估分类器的预测性能[J]. 情报学报, 2007, 26(2):275-279. YANG Bo, CHENG Zekai, QIN Feng. Using AUC to Evaluate Predictive Performance of Classifiers[J]. Journal of the China Society for Scientific and Technical Information, 2007, 26(2):275-279.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献