宁夏东部半干旱区典型植物群落遥感分类特征

doi:10.13866/j.azr.2022.06.23

摘要/Abstract

摘要：

以宁夏哈巴湖国家级自然保护区为研究区域，对区域尺度上典型植物群落遥感提取信息进行研究，验证基于多时相Landsat 8数据对该地区植物群落提取的适用性。在最佳指数因子的基础上，确定最优波段组合；同时结合面向对象的分类方法，对比分析采用单期影像与2期影像不同波段组合的共计8组分类实验，探究多时相数据对分类精度的影响。结果表明：（1）不同分割参数设置对分类精度有一定影响，紧致度因子和形状因子分别在0.7和0.1时，达到实验最优分类效果；（2）研究区内人工大面积种植的植被，其分类效果较好，白刺、芨芨草等天然混生的植物群落容易造成误分混分；（3）由最终分类精度可知，采用多时相数据进行分类可大大提高分类精度，较单时相数据总体分类精度和Kappa系数最大提升了8.24%和0.10，可有效提高研究区植被信息的提取精度。

关键词: 面向对象分类, 多时相数据, 植物群落, 波段组合, 分类精度, 半干旱区

Abstract:

The vegetation information extraction technology by remote sensing has been widely used in environmental monitoring and ecological protection. It can be seen that this technology has more important significance and application value in desertification frontier areas, such as arid and semi-arid regions. Taking the Haba Lake National Nature Reserve as the research area, the remote sensing extraction of typical plant communities at the regional scale was studied, verifying the applicability of the extraction of plant communities in this area based on the multi-temporal Landsat8 data. Based on the optimal index factor, the optimal band combination was determined. Simultaneously, combined with the object-oriented classification method, eight groups of classification experiments using single-phase and two-phase images with different band combinations were compared and analyzed to explore the influence of multi-temporal data on classification accuracy. The research results show that: (1) Different segmentation parameter settings influence the classification accuracy. Among these, when the compactness factor and shape factor are 0.7 and 0.1, respectively, the optimal classification effect of the experiment can be achieved. (2) The vegetation planted artificially in a large area of the study area has a better classification effect. Natural mixed plant communities, such as Bletilla striata and Achnatherum splendens, easily cause misclassification and mixed classification. (3) According to the final classification accuracy, the classification of multi-temporal data can significantly improve the classification accuracy. Compared with the overall classification accuracy and Kappa coefficient of single-phase data, the maximum improvement is 8.24% and 0.10, which can effectively improve the extraction accuracy of vegetation information in the study area.

Key words: object oriented classification, mulit-temporal image, plant community, band combination, classification accuracy, semi-arid areas

庞海威,余殿,任成宝,张玉,郑彩之,郭佳诚,边振,桑国庆. 宁夏东部半干旱区典型植物群落遥感分类特征[J]. 干旱区研究, 2022, 39(6): 1930-1941.

PANG Haiwei,YU Dian,REN Chengbao,ZHANG Yu,ZHENG Caizhi,GUO Jiacheng,BIAN Zhen,SANG Guoqing. Remote sensing classification characteristics of typical plant communities in the semi-arid areas of eastern Ningxia[J]. Arid Zone Research, 2022, 39(6): 1930-1941.

图/表 17

图1

图2

表1

表2

表3

图3

图4

表4

表5

图5

图6

图7

图8

表6

表7

图9

图10

参考文献 28

[1]	尼加提·卡斯木, 师庆东, 刘素红, 等. 基于卷积网络的沙漠腹地绿洲植物群落自动分类方法[J]. 农业机械学报, 2019, 50(1): 217-225.
	[ Nijat Kasim, Shi Qingdong, Liu Suhong, et al. Automatic classification method of oasis plant community in desert hinterland based on VGGNet and ResNet models[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(1): 217-225. ]
[2]	邬亚娟, 刘廷玺, 童新, 等. 基于面向对象的干旱半干旱地区植被分类[J]. 干旱区研究, 2020, 37(4): 1026-1034.
	[ Wu Yajuan, Liu Tingxi, Tong Xin, et al. Vegetation classification in arid and semi-arid areas using an object-oriented method[J]. Arid Zone Research, 2020, 37(4): 1026-1034. ]
[3]	张贵花, 王瑞燕, 赵庚星, 等. 基于物候参数和面向对象法的濒海生态脆弱区植被遥感提取[J]. 农业工程学报, 2018, 34(4): 209-216.
	[ Zhang Guihua, Wang Ruiyan, Zhao Gengxing, et al. Extraction of vegetation information in coastal ecological vulnerable areas from remote sensing data based on phenology parameters and object-oriented method[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(4): 209-216. ]
[4]	刘润红, 梁士楚, 赵红艳, 等. 中国滨海湿地遥感研究进展[J]. 遥感技术与应用, 2017, 32(6): 998-1011.
	[ Liu Runhong, Liang Shichu, Zhao Hongyan, et al. Progress of Chinese Coastal Wetland based on remote sensing[J]. Remote Sensing Technology and Application, 2017, 32(6): 998-1011. ]
[5]	陈仲新, 任建强, 唐华俊, 等. 农业遥感研究应用进展与展望[J]. 遥感学报, 2016, 20(5): 748-767.
	[ Chen Zhongxin, Ren Jianqiang, Tang Huajun, et al. Progress and perspectives on agricultural remote sensing research and applications in China[J]. Journal of Remote Sensing, 2016, 20(5): 748-767. ]
[6]	Tang Lina, Shao Guofan. Drone remote sensing for forestry research and practices[J]. Journal of Forestry Research, 2015, 26(4): 791-797. doi: 10.1007/s11676-015-0088-y
[7]	王鹏, 万荣荣, 杨桂山. 基于多源遥感数据的湿地植物分类和生物量反演研究进展[J]. 湿地科学, 2017, 15(1): 114-124.
	[ Wang Peng, Wan Rongrong, Yang Guishan. Advance in classification and biomass estimation of plants in wetlands based on multi-source remote sensing data[J]. Wetland Science, 2017, 15(1): 114-124. ]
[8]	徐凯健, 田庆久, 岳继博, 等. 基于多光谱影像的森林树种识别及其空间尺度响应[J]. 应用生态学报, 2018, 29(12): 3986-3994. doi: 10.13287/j.1001-9332.201812.011
	[ Xu Kaijian, Tian Qingjiu, Yue Jibo, et al. Forest tree species identification and its response to spatial scale based on multispectral and multi-resolution remotely sensed data[J]. Chinese Journal of Applied Ecology, 2018, 29(12): 3986-3994. ] doi: 10.13287/j.1001-9332.201812.011
[9]	Brewer C K, Barber J A, Willhauck G, et al. Multi-source and multi-classifier system for regional landcover mapping[J]. Advances in Techniques for Analysis of Remotely Sensed Data, 2003 : 143-149.
[10]	Bross L C. Using Landsat TM Imagery to Monitor Vegetation Change Following Flow Restoration to the Lower Owens River, California[D]. California: Portland State University, 2015.
[11]	付伟, 徐涵秋, 王美雅, 等. 南方红壤典型水土流失区植被分类及植被类型变化的遥感评估——以福建省长汀县河田地区为例[J]. 遥感技术与应用, 2017, 32(3): 546-555.
	[ Fu Wei, Xu Hanqiu, Wang Meiya, et al. Vegetation classification and variation assessment in a typical red soil erosion area in southern China: Hetian, changting country of Fujian Province[J]. Remote Sensing Technology and Application, 2017, 32(3): 546-555. ]
[12]	章晓洁, 邓艳芬, 张亚超, 等. 利用Sentinel-2A多光谱成像仪与Landsat 8陆地成像仪影像进行普陀山岛植被分类效果比较[J]. 测绘通报, 2019(10) : 97-100.
	[ Zhang Xiaojie, Deng Yanfen, Zhang Yachao, et al. Comparison of vegetation classification performances on Putuoshan island using Sentinel-2A MSI and Landsat 8 OLI images[J]. Bulletin of Surveying and Mapping, 2019(10): 97-100. ]
[13]	朱永森, 曾永年, 张猛. 基于HJ卫星数据与面向对象分类的土地利用/覆盖信息提取[J]. 农业工程学报, 2017, 33(14): 258-265.
	[ Zhu Yongsen, Zeng Yongnian, Zhang Meng. Extract of land use/cover information based on HJ satellites data and object-oriented classification[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(14): 258-265. ]
[14]	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): 479-492.
	[ Zhang Meng, Zeng Yongnian, Zhu Yongsen. Wetland mapping of Donting Lake Basin based on time-series MODIS data and object-oriented method[J]. Journal of Remote Sensing, 2017, 21(3): 479-492. ]
[15]	马燕妮, 明冬萍, 杨海平. 面向对象影像多尺度分割最大异质性参数估计[J]. 遥感学报, 2017, 21(4): 566-578.
	[ Ma Yanni, Ming Dongping, Yang Haiping. Scale estimation of object-oriented image analysis based on spectral-spatial statistics[J]. Journal of Remote Sensing, 2017, 21(4): 566-578. ]
[16]	任传帅, 叶回春, 崔贝, 等. 基于面向对象分类的芒果林遥感提取方法研究[J]. 资源科学, 2017, 39(8): 1584-1591. doi: 10.18402/resci.2017.08.14
	[ Ren Chuanshuai, Ye Huichun, Cui Bei, et al. Acreage estimation of mango orchards using object-oriented classification and remote sensing[J]. Resources Science, 2017, 39(8): 1584-1591. ] doi: 10.18402/resci.2017.08.14
[17]	Douglas G, Goodin. Mapping land cover and land use from object-based classification: an example from a complex agricultural landscape[J]. International Journal of Remote Sensing, 2015, 36(18): 4702-4723. doi: 10.1080/01431161.2015.1088674
[18]	向海燕, 罗红霞, 刘光鹏, 等. 基于Sentinel-1A极化SAR数据与面向对象方法的山区地表覆被分类[J]. 自然资源学报, 2017, 32(12): 2136-2148.
	[ Xiang Haiyan, Luo Hongxia, Liu Guangpeng, et al. Land cover classification in mountain areas based on Sentinel-1A polarimetric SAR data and object oriented method[J]. Journal of Nature Resources, 2017, 32(12): 2136-2148. ]
[19]	姚金玺, 王浪, 李建忠, 等. 青海诺木洪地区多源遥感及多特征组合地物分类[J]. 农业工程学报, 2022, 38(3): 247-256.
	[ Yao Jinxi, Wang Lang, Li Jianzhong, et al. Multi-source remote sensing and multi-feature combination ground object classification in Nuomuhong areas, Qinghai Province of China[J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(3): 247-256. ]
[20]	罗开盛. 基于面向对象技术的旅游用地遥感识别[J]. 中国科学院大学学报, 2020, 37(4): 490-497. doi: 10.7523/j.issn.2095-6134.2020.04.008
	[ Luo Kaisheng. Remote sensing identification of tourism land use based on object-oriented technology[J]. Journal of University of Chinese Academy of Sciences, 2020, 37(4): 490-497. ] doi: 10.7523/j.issn.2095-6134.2020.04.008
[21]	Iabchoon S, Wongsai S, Chankon K. Mapping urban impervious surface using object-based image analysis with WorldView-3 satellite imagery[J]. Journal of Applied Remote Sensing, 2017, 11: 046015.
[22]	马梦茹, 张永彬, 王奕丹. 最佳波段选择的迁西县土地利用信息提取研究[J]. 华北理工大学学报(自然科学版), 2021, 43(3): 20-29.
	[ Ma Mengru, Zhang Yongbin, Wang Yidan. Research on extraction of land utilization information of Qianxi County based on optimum band selection[J]. Journal of North China University of Seience and Technology(Natural Science Edition), 2021, 43(3): 20-29. ]
[23]	彭佳忆, 王新军, 朱磊, 等. 基于无人机影像的荒漠地表类型信息提取[J]. 干旱区研究, 2019, 36(3): 771-780.
	[ Peng Jiayi, Wang Xinjun, Zhu Lei, et al. Information extraction of desert surface types based on UVA Image[J]. Arid Zone Research, 2019, 36(3): 771-780. ]
[24]	王芳, 杨武年, 王建, 等. 遥感影像多尺度分割中最优尺度的选取及评价[J]. 遥感技术与应用, 2020, 35(3): 623-633.
	[ Wang Fang, Yang Wunian, Wang Jian, et al. Selection and evaluation of the optimal scale in multiscale segmentation of remote sensing images[J]. Remote Sensing Technology and Application, 2020, 35(3): 623-633. ]
[25]	陈喜梅, 王庆国, 王一斐, 等. eCognition在西藏全区草原普查遥感影像信息快速提取中的应用[J]. 中国草地学报, 2017, 39(2): 117-120.
	[ Chen Ximei, Wang Qingguo, Wang Yifei, et al. Application of eCognition to rapidly extract the information form remote sensing image of grassland general investigation in Tibet region[J] .Chinese Journal of Grassland, 2017, 39(2): 117-120. ]
[26]	郑琪, 邸苏闯, 潘兴瑶, 等. 基于Rapid Eye数据的北京生态涵养区土地利用分类及变化研究[J]. 遥感技术与应用, 2020, 35(5): 1118-1126.
	[ Zheng Qi, Di Suchuang, Pan Xingyao, et al. Study of land use classification and changes in the ecological conservation region of Beijing based on Rapid Eye images[J]. Remote Sensing Technology and Application, 2020, 35(5): 1118-1126. ]
[27]	袁磊, 赵俊三, 陈国平, 等. 面向对象的土地利用多尺度时空数据模型[J]. 测绘科学, 2014, 39(11): 52-56, 71.
	[ Yuan Lei, Zhao Junsan, Chen Guoping, et al. An object-oriented multi-scale spatio-temporal data model for land-use[J]. Science of Surveying and Mapping, 2014, 39(11): 52-56, 71. ]
[28]	宋奇, 史舟, 冯春晖, 等. 基于1990—2019年多时相影像的干旱区绿洲景观格局分析[J]. 干旱区研究, 2022, 39(2): 594-604.
	[ Song Qi, Shi Zhou, Feng Chunhui, et al. Analysis of landscape pattern from 1990 to 2019 based on multi-temporal imagery in arid oasis[J]. Arid Zone Research, 2022, 39(2): 594-604. ]

类别	数据层	对象属性名称	对象特征
光谱特征	Coastal	Mean、Max、Brightness、Standard	分割图层平均值、最大值、标准差及参与波段亮度加权平均值
	Blue
	Green
	Red
	Nir
	SWIR1
	SWIR2
植被指数	NDVI	NDVI	平均值和标准差

实验方案	波段组合
A	最优3波段（7月31日影像）
B	最优3波段（7月31日影像）+波段1（2月21日影像）
C	最优3波段（7月31日影像）+波段2（2月21日影像）
D	最优3波段（7月31日影像）+波段3（2月21日影像）
E	最优3波段（7月31日影像）+波段4（2月21日影像）
F	最优3波段（7月31日影像）+波段5（2月21日影像）
G	最优3波段（7月31日影像）+波段6（2月21日影像）
H	最优3波段（7月31日影像）+波段7（2月21日影像）

植被	群系组	群落型
植被	盐生灌丛	白刺群落（Nitraria tangutorum）
	丛生禾草盐生草甸	芨芨草群落（Achnatherum splendens）
	沙地落叶灌丛及半灌丛	柠条群落（Caragana microphylla）
	沙地落叶灌丛及半灌丛	黑沙蒿群落（Artemisia desertorum）
	沙地人工落叶灌木林	沙柳群落（Salix cheilophila.）
	盐碱地人工落叶灌木林	柽柳群落（Tamarix chinensis）
	人工寒温性针叶松林	樟子松群落（Pinus sylvestris var. mongolica）
	人工温性落叶阔叶林	杨树群落（Populus ）
	人工温性落叶阔叶林	榆树群落（Ulmus pumila）

Correlation	Band1	Band2	Band3	Band4	Band5	Band6	Band7
Band 1	1.00
Band 2	0.99	1.00
Band 3	0.98	0.99	1.00
Band 4	0.95	0.96	0.99	1.00
Band 5	0.85	0.83	0.85	0.81	1.00
Band 6	0.91	0.92	0.95	0.96	0.88	1.00
Band 7	0.90	0.91	0.95	0.97	0.82	0.99	1.00

OIF值排序	波段组合	标准差和	相关系数和	OIF
1	4、5、7	3363.75	2.60	1293.63
2	5、6、7	3389.62	2.69	1261.78
3	4、5、6	3196.66	2.65	1206.68
4	4、6、7	3440.88	2.92	1179.95
5	3、5、7	3023.22	2.62	1153.20
6	2、5、7	2774.02	2.56	1082.91
7	3、6、7	3100.36	2.89	1071.73
8	3、4、5	2830.26	2.66	1065.43
9	3、5、6	2856.13	2.68	1064.29
10	1、5、7	2721.39	2.57	1058.27