基于多特征组合优选与随机森林算法的石漠化信息提取——以云南省昭通市为例

doi:10.16843/j.sswc.2022187

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (5314 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract [Background] Rocky desertification is one of the most important geo-ecological disasters in southwestern China. It causes land resources loss, ecosystem degradation, drought and water shortage, which seriously threatens the ecological balance, food security and the absence of large-scale return to poverty in southwest China. Accurate extraction of rock desertification information is crucial to the sustainable development of regional economy and society. [Methods] Aiming at the problems such as single temporal phase, poor timeliness and low accuracy of regional scale extraction results in the current rocky desertification information extraction, this study took Zhaotong city of Yunnan as an example by proposing an optimized classification method incorporating multi-features. Based on the preferential selection of samples and features, the multiple features such as spectra, indices, fractional vegetation cover, bedrock exposure rate, texture and topography were extracted using Sentinel-2 imagery and DEM data, and five classification schemes were constructed, as well as the extraction was completed using the random forest classification algorithm. [Results] 1) When the Jeffries-Matusita (JM) distance algorithm was applied to evaluate separability of input features, the input features with the maximum average JM distance were BSI and Albedo, followed by TF₁ and slope, and the input features with the minimum average JM distance were B6 and B8. For rocky desertification land and other land cover types, slope, TF₁, BSI and Albedo had JM distance greater than 1.9, indicating a significant effect on the classification accuracy. 2) The importance of all input features was analyzed by the forest classification algorithm. The slope feature contributed the most to the classification accuracy, followed by the texture feature TF₁, NDVI and BSI, and the contribution of B4 and B6 bands in the spectral feature was relatively small. 3) In the case of the same number and distribution of sample points, compared with the other four classification schemes, the overall accuracy (OA) of the feature selection scheme obtained by using JM distance was 88.0%, and the Kappa coefficient was 0.85. The producer accuracy (PA) and user accuracy (UA) of rocky desertification land reached 91.2% and 83.8%, respectively. Finally, the rocky desertification land area of Zhaotong in 2020 was 2 820 km², accounting for 11.11% of the total land area of the region. The classification results were also in good agreement with the field survey area. [Conclusions] In this study, the input samples and characteristics are optimized by combining land use and land cover data and JM distance algorithm respectively, which effectively improves the phenomenon of misclassification, omission and large error in the fractured area of rocky desertification distribution in plateau mountainous areas. By the method proposed in this study, high classification accuracy at regional scale can be achieved, which provides reference for relevant departments to carry out rocky desertification prevention and monitoring.

Key words： Sentinel-2 rocky desertification random forest algorithm feature optimization Google Earth Engine (GEE)

Received: 02 September 2022

PACS:

TP75

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LIU Zicheng
	CHEN Guokun
	WEN Qingke
	YI Ling
	ZHAO Jingjing

Cite this article:

LIU Zicheng,CHEN Guokun,WEN Qingke, et al. Extraction of rocky desertification information based on multi-feature combination optimization and random forest algorithm: A case study of Zhaotong city in Yunnan province[J]. SSWC, 2024, 22(1): 95-105.

URL:

http://journal12.magtechjournal.com/Jweb_stbc/EN/10.16843/j.sswc.2022187 OR http://journal12.magtechjournal.com/Jweb_stbc/EN/Y2024/V22/I1/95

[1]	王世杰,李阳兵,李瑞玲. 喀斯特石漠化的形成背景、演化与治理[J]. 第四纪研究,2003,23(6):657. WANG Shijie,LI Yangbing,LI Ruiling. Formation background,evolution and governance of karst rocky desertification[J]. Quaternary Sciences,2003,23(6):657.
[2]	苏维词. 中国西南岩溶山区石漠化的现状成因及治理的优化模式[J]. 水土保持学报,2002,16(2):29. SU Weici. Present situation causes of rocky desertification in karst mountainous areas of Southwest China and its optimization model[J]. Journal of Soil and Water Conservation,2002,16(2):29.
[3]	赵英时. 遥感应用分析原理与方法[M]. 北京:科学出版社,2013:1. ZHAO Yingshi. Principles and methods of remote sensing application analysis[M]. Beijing:Science Press,2013:1.
[4]	张盼盼,胡远满,肖笃宁,等. 一种基于多光谱遥感影像的喀斯特地区裸岩率的计算方法初探[J]. 遥感技术与应用,2010,25(4):510. ZHANG Panpan,HU Yuanman,XIAO Duning,et al. A calculation method of bare rock rate in karst area based on multispectral remote sensing image[J]. Remote Sensing Technology and Application,2010,25(4):510.
[5]	闫利会,周忠发,谢雅婷,等. 贵州高原石漠化敏感性与宏观地貌的空间关联分析[J]. 中国岩溶,2018,37(3):400. YAN Lihui,ZHOU Zhongfa,XIE Yating,et al. Spatial correlation analysis of rocky desertification sensitivity and macrogeomorphology in Guizhou Plateau[J]. Carsologica Sinica,2018,37(3):400.
[6]	MA Shuyong,ZHU Xinglei,AN Yulun. Remote sensing image classification based on decision tree in the karst rocky desertification areas:A case study of Kaizuo Township[J]. Asian Agricultural Research,2014,6(7):58.
[7]	凌成星,林辉,薛晓坡,等. 基于植被指数特征和空间分析的石漠化信息提取[J]. 水土保持研究,2009,16(1):167. LING Chengxing,LIN Hui,XUE Xiaobo,et al. Extraction of rocky desertification information based on vegetation index and spatial analysis[J]. Research of Soil and Water Conservation,2009,16(1):167.
[8]	李阳兵,罗光杰,王世杰,等. 黔中高原面石漠化演变典型案例研究:以普定后寨河地区为例[J]. 地理研究,2013,32(5):828. LI Yangbing,LUO Guangjie,WANG Shijie,et al. Typical case study on evolution of rocky desertificationin central Guizhou Plateau:A case study of Houzhai river area in Puding[J]. Geographical Research,2013,32(5):828.
[9]	陈棋,张超,田湘云,等. 基于GEE的曲靖石漠化时空演变过程分析[J/OL]. 西南林业大学学报(自然科学),2022,42(5):1. CHEN Qi, ZHANG Chao, TIAN Xiangyun, et al. Spatial-temporal evolution analysis of Qujing rocky desertification based on GEE[J/OL].Journal of Southwest Forestry University(Natural Sciences),2022,42(5):1.
[10]	王明明,王世杰,白晓永,等. 典型小流域喀斯特石漠化演变特征及其关键表征因子与驱动因素[J]. 生态学报,2019,39(16):6083. WANG Mingming,WANG Shijie,BAI Xiaoyong,et al. Evolution characteristics,key characterization factors and driving factors of karst rocky desertification in typical small watershed[J]. Acta Ecologica Sinica,2019,39(16):6083.
[11]	刘鹏,张紫霞,杨波,等. 基于信息熵与层次分析法的石漠化形成驱动因素分析[J]. 西南林业大学学报(自然科学),2019,39(5):100. LIU Peng,ZHANG Zixia,YANG Bo,et al. Analysis of driving factors of rocky desertification based on information entropy and analytic hierarchyprocess[J]. Journal of Southwest Forestry University(Natural Sciences),2019,39(5):100.
[12]	种国双,海月,郑华,等. 石漠化遥感信息提取方法进展[J]. 应用科学学报,2021,39(6):961. ZHONG Guoshuang,HAI Yue,ZHENG Hua, et al. Progress of remote sensing information extraction methods for rocky desertification[J]. Journal of Applied Sciences,2021,39(6):961.
[13]	周迪,倪忠云,杨振宇. 面向对象的喀斯特地区石漠化遥感信息提取研究:以贵州省大方地区为例[J]. 地质力学学报,2018,24(2):263. ZHOU Di,NI Zhongyun,YANG Zhenyu. Object-oriented remote sensing information extraction of karst rocky desertification:A case study of Dafang area in Guizhou province[J]. Journal of Geomechanics,2018,24(2):263.
[14]	XU Feng,LI Zhaofu,ZHANG Shuyu,et al. Mapping winter wheat with combinations of temporally aggregated Sentinel-2 and Landsat-8 data in Shandong province,China[J]. Remote Sensing,2020,12(12):2065.
[15]	王小娜,田金炎,李小娟,等. Google Earth Engine云平台对遥感发展的改变[J]. 遥感学报,2022,26(2):299. WANG Xiaona,TIAN Jinyan,LI Xiaojuan,et al. Google Earth Engine cloud platform changes the development of remote sensing[J]. National Remote Sensing Bulletin,2022,26(2):299.
[16]	YUAN Yixin,WEN Qingke,ZHAO Xiaoli,et al. Identifying grassland distribution in a mountainous region in Southwest China using multi-source remote sensing images[J]. Remote Sensing,2022,14(6):1472.
[17]	NI Rongguang,TIAN Jinyan,LI Xiaojuan,et al. An enhanced pixel-based phenological feature for accurate paddy rice mapping with Sentinel-2 imagery in Google Earth Engine[J]. ISPRS Journalof Photogrammetry and Remote Sensing,2021,178:282.
[18]	GAO Bocai. NDWI:A normalized difference water index for remote sensing of vegetation liquid water from space[J]. Remote Sensing of Environment,1996,58(3):257.
[19]	POLYKRETIS C,GRILLAKIS M G,ALEXAKIS D D. Exploring the impact of various spectral indiceson land cover change detection using changevector analysis:A case study of crete island,Greece[J]. Remote Sensing,2020,12(2):319.
[20]	宗慧琳,甘淑,任鹏飞. 滇东南岩溶石漠化信息提取研究[J]. 价值工程,2014,33(10):211. ZONG Huilin,GAN Shu,REN Pengfei. Extraction of karst rocky desertification information in Southeast Yunnan[J]. Value Engineering,2014,33(10):211.
[21]	熊康宁,黎平,周忠发,等. 喀斯特石漠化的遥感-GIS典型研究:以贵州省为例[M]. 北京:地质出版社,2002:23. XIONG Kangning,LI Ping,ZHOU Zhongfa,et al. Remote sensing-GIS typical study of karst rocky desertification:A case study of Guizhou province[M]. Beijing:Geology Publishing House,2002:23.
[22]	张猛,曾永年. 基于多时相Landsat数据融合的洞庭湖区水稻面积提取[J]. 农业工程学报,2015,31(13):178. ZHANG Meng,ZENG Yongnian. Rice area extraction in Dongting lake area based on multi-temporal Landsat data fusion[J]. Transactions of the CSAE,2015,31(13):178.
[23]	张猛,曾永年,朱永森. 面向对象方法的时间序列MODIS数据湿地信息提取:以洞庭湖流域为例[J]. 遥感学报,2017,21(3):479. ZHANG Meng,ZENG Yongnian,ZHU Yongsen. Wetland information extraction from time series MODIS data using object-oriented method:A case study of Dongting lake basin[J]. National Remote Sensing Bulletin,2017,21(3):479.
[24]	谭深,吴炳方,张鑫. 基于Google Earth Engine与多源遥感数据的海南水稻分类研究[J]. 地球信息科学学报,2019,21(6):937. TAN Shen,WU Bingfang,ZHANG Xin. Classification of rice in Hainan based on Google Earth Engine and multi-source remote sensing data[J]. Journal of Geo-Information Science,2019,21(6):937.
[25]	李丽,童立强,李小慧. 基于植被覆盖度的石漠化遥感信息提取方法研究[J]. 国土资源遥感,2010(2):592. LI Li,TONG Liqiang,LI Xiaohui. Remote sensing information extraction method of rocky desertification based on vegetation coverage[J]. Remote Sensing for Land & Resources,2010(2):59.