Study on Vegetation Extraction Method in Karst Areas Based on Visible Light Band of Sentinel 2

doi:10.13466/j.cnki.lyzygl.2020.06.023

Abstract

Abstract:

In order to accurately extract vegetation information from visible band,study the characteristics of Karst vegetation in the visible light range and use the visible light band of the Sentinel 2 image,the spatial characteristics of the three bands of red,green and blue are created.The results showed that the spatial ratio characteristics of vegetation spectrum in blue-green are separable from other ground features and the brightness value of the vegetation is relatively low in the three visible lights.Comparing the extracted vegetation results with the normalized difference vegetation index,over-green index,ultra-green and ultra-red difference index,vegetation index,visible light band difference index,normalized green-red difference index,and normalized green-blue difference index,the study finds its precision is 92.67%,which is superior to other visible light extraction methods.It shows that the method is effective for the extraction of vegetation in Karst areas in the visible light band.

Key words: visible band, vegetation extraction, Karst landform, vegetation index, Sentinel 2 image

CLC Number:

TP79
P237

DUAN Jiwei, ZHONG Jiusheng, JIANG Li, DAI Renli, HE Zhiyuan, LIN Shuangshuang, HE Xin. Study on Vegetation Extraction Method in Karst Areas Based on Visible Light Band of Sentinel 2[J]. FOREST RESOURCES WANAGEMENT, 2020, 0(6): 143-152.

Figures/Tables 11

Tab.1

Fig.1

Tab.2

Fig.2

Fig.3

Fig.4

Tab.3

Fig.5

Tab.4

Fig.6

Tab.5

References 25

[1]	Tucker C J. Red and photographic infrared linear combinations for monitoringvegetation[J]. Remote Sensing of Environment, 1979,8(2):127-150. doi: 10.1016/0034-4257(79)90013-0
[2]	田庆久, 闵祥军. 植被指数研究进展[J]. 地球科学进展, 1998,13(4):3-5.
[3]	Huete A R. A soil-adjusted vegetation index(SAVI)[J]. Remote Sensing of Environment, 1988,48(25):295-309.
[4]	高志海, 魏怀东, 丁峰. TM影像VI提取植被信息技术研究[J]. 干旱区资源与环境, 1998,12(3):3-5.
[5]	Liu Zhanyu, Huang Jingfeng, Wu Xinhong, et al. Comparison of Vegetation Indices and Red-edge Parameters for Estimating Grassland Cover from Canopy Reflectance Data[J]. Journal of Integrative Plat Biology, 2007,49(3):299-306.
[6]	郭铌. 植被指数及其研究进展[J]. 干旱气象, 2003,21(4):71-75.
[7]	张仁华, 饶农新, 廖国男. 植被指数的抗大气影响探讨[J]. 植物学报, 1996(1):53-62.
[8]	张立福, 张良培, 村松加奈子, 等. 利用MODIS数据计算陆地植被指数VIUPD[J]. 武汉大学学报:信息科学版, 2005,30(8):699-702.
[9]	Gitelson A A, Kaufman Y J, Stark R, et al. Novel algorithms for remote estimation of vegetation fraction[J]. Remote Sensing of Environment, 2002,80(1):76-87.
[10]	Meyer G E, Neto J C. Verification of color vegetation indices for automated crop image a pplication[J]. Computers and Electronics in Agriculture, 2008,63(2):282-293
[11]	Woebbecke D M, Meyer G E, Von Bargen K, et al. Color indices for weed identification under various soil,residue,and lighting conditions[J]. Transactions of the ASAE, 1995,38(1):259-269.
[12]	谢小曼, 李俐, 张迁迁, 等. SAR遥感指数研究进展[J]. 中国农业信息, 2019,31(5):13-24.
[13]	何霄嘉, 王磊, 柯兵, 等. 中国喀斯特生态保护与修复研究进展[J]. 生态学报, 2019,39(18):6577-6585.
[14]	刘筱笛. 贵阳市周边地区特色村落规划方法研究[D]. 西安:长安大学, 2012.
[15]	周晓芳. 贵阳市发展都市农业旅游探讨[J]. 山地农业生物学报, 2003,22(1):66-69.
[16]	白洁, 葛全胜, 戴君虎. 贵阳木本植物物候对气候变化的响应[J]. 地理研究, 2009,28(6):1606-1614.
[17]	肖红伟, 肖化云, 王燕丽. 贵阳大气降水化学特征及来源分析[J]. 中国环境科学, 2010,30(12):1590-1596.
[18]	冯小铭, 郭坤一, 王敬东. 对中国城市环境地质工作的思考[J]. 安全与环境工程, 2001,8(4):1-4
[19]	谷祥辉, 张英, 桑会勇, 等. 基于哨兵2时间序列组合植被指数的作物分类研究[J]. 遥感技术与应用, 2020,35(3):702-711.
[20]	杨栩, 朱大明, 杨润书, 等. 一种提取不透水面的可见光波段遥感指数[J]. 农业工程学报, 2020,36(8):127-134.
[21]	孙国祥, 汪小旵, 闫婷婷, 等. 基于机器视觉的植物群体生长参数反演方法[J]. 农业工程学报, 2014,30(20):187-195.
[22]	Hague T, Tillett N D, Wheeler H. Automated crop and weed monitoring in widely spaced cereals[J]. Precision Agriculture, 2006,7:21-32
[23]	汪小钦, 王苗苗, 王绍强, 等. 基于可见光波段无人机遥感的植被信息提取[J]. 农业工程学报, 2015,31(5):152-159.
[24]	Gitelson A A, Kaufman Y J, Stark R, et al. Novel algorithms for remote estimation of vegetation fraction[J]. Remote Sensing of Environment, 2002,80(1):76-87. doi: 10.1016/S0034-4257(01)00289-9
[25]	Hunt E R, Michel C, Daughtry C S T, et al. Evaluation of Digital Photography from Model Aircraft for Remote Sensing of Crop Biomass and Nitrogen Status[J]. Precision Agriculture, 2005,6(4):359-378.

波段	中心波长/nm	波段宽度/nm	空间分辨率/m	备注
2	492.4	98
3	559.8	45	10
4	664.6	38
8	832.8	145
5	704.1	19
6	740.5	18	20	窄波段红边
7	782.8	28
8a	864.7	33
11	1616.7	143	20
12	2202.4	242
1	442.7	27
9	945.1	26	60
10	1373.5	75

地物名称	红波段			绿波段			蓝波段
地物名称	均值	标准差	植被阈值内像素比/%	均值	标准差	植被阈值内像素比/%	均值	标准差	植被阈值内像素比/%
植被	450.72	94.87	100.00	709.79	134.97	100.00	402.56	76.68	100.00
水体	427.74	205.60	72.63	580.57	247.22	29.21	487.56	207.98	72.10
其他城市建筑	1156.81	354.36	19.21	1358.98	435.68	33.15	1156.81	354.36	9.73
裸土	2293.95	463.41	0	1871.77	437.28	0	1413.10	409.54	1.57
蓝色建筑	1296.93	277.43	5.52	1407.75	322.88	11.05	2274.54	298.77	0.26
火燎迹地	616.69	72.61	99.47	504.68	61.680	72.1	389.65	58.90	98.15
红色建筑	2088.54	714.92	0.26	1089.97	301.91	52.89	915.82	295.20	18.42
高速	1052.31	174.69	9.21	1038.19	142.65	43.42	888.84	132.88	2.36

光谱比值地类	斜率k值	截距b值	R²值	点线阈值	计算结果
绿比蓝植被	1.4676	108.03	0.834	(-138.3195,105.8213)	A2
绿比蓝水体	1.0125	51.588	0.8455	(-39.4251,67.3981)	B2
红比绿植被	0.6493	-5.224	0.6799	(-271.1123,171.7994)	C2
红比绿火烧迹地	1.0763	75.293	0.7138	(-97.6862,110.3322)	D2

植被指数提取方法	植被指数提取方法中文全称	公式	文献
NDVI	归一化差异植被指数	(NIR-R)/(NIR+R)	[3]
EXG	过绿指数	2g-r-b	[11]
EXGR	超绿超红差异指数	EXG-1.4r-g	[21]
VEG	植被指数	g/r⁰^.⁶⁷b⁰^.³³	[22]
VDVI	可见光波段差异指数	(2G-R-B)/(2G+R+B)	[23]
NGRDI	归一化绿红差异指数	(G-R)/(G+R)	[24]
NGBDI	归一化绿蓝差异指数	(G-B)/(G+B)	[25]

植被指数提取方法	总体精度/%	生产精度/%		用户精度/%		Kappa系数
植被指数提取方法	总体精度/%	植被	非植被	植被	非植被	Kappa系数
NDVI	94.13	94.59	93.67	93.63	94.63	0.8826
EXG	90.99	96.93	85.14	86.51	96.58	0.8200
EXGR	92.60	94.43	90.80	90.99	94.31	0.8521
VEG	91.57	96.590	86.63	87.66	96.27	0.8315
VDVI	90.12	97.60	82.76	84.77	97.23	0.8026
NGRDI	92.45	93.60	91.33	91.39	93.55	0.8491
NGBDI	77.49	97.09	58.21	69.56	95.32	0.5513
本文方法	92.67	94.46	90.92	91.09	94.34	0.8535