The elm sparse forest is an important component in the Otindag sandy land ecosystem, which is of great significance for windbreak and sand fixation. In order to obtain the spatial distribution information of elm trees quickly and accurately, this paper proposes a method of automatic sand elm identification based on remote sensing technology. With the data of domestic high spatial resolution satellite GF-2, the research was implemented on Zhenglan Banner, Xilin Gol League, Inner Mongolia. Combined with the characteristics of elm sparse distribution in the sand, normalizd difference vegetation index (NDVI) threshold was firstly used to quickly extract the coarse distribution of elm. Then, a method based on geographic object based image analysis (GEOBIA) was used to extract the distribution of elm accurately. To compensate for the uncertainty of GEOBIA method in feature selection and rule set construction, this study used SEaTH algorithm to optimize features and automatically calculate the feature threshold. The results show that the proposed methods reached the overall accuracy of 88.17% and Kappa coefficient of 0.76 in identifying the sparse elm. Among them, elm mapping accuracy could reach 99.14%. Therefore, it is effective to identify elms by using GF-2 and the method proposed in this study. This method can provide technical support for the further research and production practices of elm sparse forest in the Otindag sandy land.
薛传平, 高志海, 孙斌, 李长龙, 王燕, 张媛媛. 浑善达克沙地榆树疏林的高分辨率遥感识别方法[J]. 国土资源遥感, 2018, 30(4): 74-81.
Chuanping XUE, Zhihai GAO, Bin SUN, Changlong LI, Yan WANG, Yuanyuan ZHANG. Research on high resolution remote sensing recognition method of elm sparse forest in Otindag sandy land. Remote Sensing for Land & Resources, 2018, 30(4): 74-81.
Peng Y, Jiang G M, Guo L , et al. Landscape pattern of elm open forest in the center part of Hunshandake sandland[J]. Science Technology Review, 2011,29(25):45-47.
Yu S L, Chen H W . Characteristics and formation causes of temperate sparse forest grassland ecosystem in Inner Mongolia Plateau[J]. Chinese Journal of Ecology, 2007,26(4):549-554.
Li Y G, Jiang G M, Gao L M , et al. Impacts of human disturbance on elms-motte-veldt in Hunshandak sandland[J]. Acta Phytoecologica Sinica, 2003,27(6):829-835.
[4]
牛海亮 . 浑善达克沙地榆树疏林榆树种群特征分析[D]. 呼和浩特:内蒙古大学, 2008.
Niu H L . Analysis of Population Characteristics of Ulmus Pumila Sparse Forest in Hunshandake Sandy Land[D]. Hohhot:Inner Mongolia Agricultural University, 2008.
Li G T, Yao Y F, Zuo H J . Study on the relation between growth and site of Ulmus pumila L.var.sabulosa J.H.Guo in Otingdag sandy land[J]. World Forestry Research, 2008,21:82-86.
Gu H Y, Yan L, Li H T , et al. An object-based automatic interpretation method for geographic features based on random forest machine learning[J]. Geomatics and Information Science of Wuhan University, 2016.
[8]
Laliberte A S, Fredrickson E L, Rango A . Combining decision trees with hierarchical object-oriented image analysis for mapping arid rangelands[J]. Photogrammetric engineering and Remote sensing, 2007,73(2):197-207.
doi: 10.14358/PERS.73.2.197
[9]
Gibbes C, Adhikari S, Rostant L , et al. Application of object based classification and high resolution satellite imagery for savanna ecosystem analysis[J]. Remote Sensing, 2010,2(12):2748-2772.
doi: 10.3390/rs2122748
[10]
Boggs G S . Assessment of SPOT 5 and QuickBird remotely sensed imagery for mapping tree cover in savannas[J]. International journal of applied earth observation and geoinformation, 2010,12(4):217-224.
doi: 10.1016/j.jag.2009.11.001
Shi M Y, Guo C L, Yang X T , et al. Canopy extraction of Populus euphratica forest in Ejina based on Quickbird image[J].Science and Technology innovation Heraid,2011(5):13-15.
Zhang J Y . Studies on Characteristics of Trees and Grass Ulmus pumila L.var.sabulosa Woodland of Sanggendalai in Hunshandake Sandy Land[D]. Hohhot:Inner Mongolia Agricultural University, 2009.
[13]
徐文铎 . 中国沙地森林生态系统[M]. 北京: 中国林业出版社, 1998.
Xu W D. Sandy Forest Ecosystem of China[M]. Beijing: China Forestry Publishing House, 1998.
[14]
Whiteside T G, Boggs G S, Maier S W . Extraction of tree crowns from high resolution imagery over Eucalypt dominant tropical savannas[J]. Photogrammetric Engineering and Remote Sensing, 2011,77(8):813-824.
doi: 10.14358/PERS.77.8.813
[15]
Karlson M, Reese H, Ostwald M . Tree crown mapping in managed woodlands (parklands) of semi-arid West Africa using WorldView-2 imagery and geographic object based image analysis[J]. Sensors, 2014,14(12):22643-22669.
doi: 10.3390/s141222643
pmid: 25460815
[16]
Nussbaum S, Niemeyer I, Canty M J. SEATH-a new tool for automated feature extraction in the context of object-based image analysis [C]//1st International Conference on Object-based Image Analysis (OBIA).Salzburg:Austria. 2006.
Yu X M, Zhan F B, Liao M S , et al. Object-oriented feature selection algorithms based on improved SEaTH algorithms[J]. Geomatics and Information Science of Wuhan University, 2012,37(8):921-924.
[18]
Marpu P R, Niemeyer I, Nussbaum S , et al. A procedure for automatic object-based classification[J]. Lecture Notes in Geoinformation and Cartography, 2008: 169-184.
doi: 10.1007/978-3-540-77058-9_9
[19]
Ren D, Abdelsalam M G. Optimum index factor (OIF) for ASTER data:Examples from the Neoproterozoic Allaqi Suture,Egypt [C]//Geological Society of American Annual Meeting. 2001: 289-290.