Comparative Study of Multiple Water Index Methods for Surface Water Extraction based on Sentinel-2 Imagery

Shuang ZHAO; Leiku YANG; Kai LIU; Ye FENG; Xinge LIANG; Peipei CUI; Chunqiao SONG

doi:10.11873/j.issn.1004-0323.2024.2.0502

Journals >Remote Sensing Technology and Application >Volume 39 >Issue 2 >Page 502 > Article

Remote Sensing Technology and Application
Vol. 39, Issue 2, 502 (2024)

Comparative Study of Multiple Water Index Methods for Surface Water Extraction based on Sentinel-2 Imagery

Shuang ZHAO^1,2,*, Leiku YANG¹, Kai LIU², Ye FENG^1,2..., Xinge LIANG^1,2, Peipei CUI^2,3 and Chunqiao SONG²|Show fewer author(s)

Author Affiliations

¹College of Surveying and Geotechnical Engineering，Henan Polytechnic University，Jiaozuo 454000，China

²Key Laboratory of Watershed Geographic Sciences，Nanjing Institute of Geography and Limnology，Chinese Academy of Sciences，Nanjing 210008，China

³School of Geography，Geomatics and Planning，Jiangsu Normal University，Xuzhou 221116，China

show less

DOI: 10.11873/j.issn.1004-0323.2024.2.0502 Cite this Article

Shuang ZHAO, Leiku YANG, Kai LIU, Ye FENG, Xinge LIANG, Peipei CUI, Chunqiao SONG. Comparative Study of Multiple Water Index Methods for Surface Water Extraction based on Sentinel-2 Imagery[J]. Remote Sensing Technology and Application, 2024, 39(2): 502 Copy Citation Text

show less

Fig. 1. Spatial distribution of study area

Download full size | View in the Article

Fig. 2. Results of mono-temporal water extraction and distribution of validation sites in the study area

Download full size | View in the Article

Fig. 3. Details of extracted water bodies by using seven water indexes

Download full size | View in the Article

Fig. 4. Annual water frequency extraction results for seven water indexes

Download full size | View in the Article

指数名称	公式	参考文献
NDWI	$N D W I = (G R E E N - N I R) / (G R E E N + N I R)$	Mcfeeters，1996
MNDWI	$M N D W I = (G R E E N - S W I R 1) / (G R E E N + S W I R 1)$	徐寒秋，2005
AWEInsh	$A W E I n s h = 4 \times (G R E E N - N I R) - (0.25 \times N I R + 2.75 \times S W I R 2)$	Feyisa 等，2014
AWEIsh	$A W E I s h = B L U E + 2.5 \times G R E E N - 1.5 \times (N I R + S W I R) - 0.25 \times S W I R 2$	Feyisa 等，2014
WI2015	$W I 2015 = 1.7204 + 171 \times G R E E N + 3 \times R E D - 70 \times N I R - 45 \times S W I R 1 - 71 \times S W I R 2$	Fisher 等，2016
CDWI	$C D W I = 0.000 \times N D W I + 0.640 \times M N D W I + 0.008 \times A W E I n s h + 0.019 \times A W E I s h + 0.333 \times W I 2015$	Wen等，2020
MNDWI_VIs	$N D W I = (G R E E N - N I R) / (G R E E N + N I R)$ $E V I = 2.5 (N I R - R E D) / 2.5 (N I R + 6 \times R E D - 7.5 \times B L U E + 1)$ $M N D W I > N D V I$ 或者 $M N D W I > E V I$ ，且 $E V I < 0.1$	Zou等，2017

Table 1. Commonly used water index calculation formulae

View in the Article

指数名称	水体面积/km²和阈值
指数名称	样区a	阈值	样区b	阈值	样区c	阈值	样区d	阈值
NDWI	94.29	0.05	215.35	-0.09	67.20	-0.05	72.21	-0.09
MNDWI	92.99	0.21	245.40	0.04	64.26	0.19	82.05	0.33
AWEInsh	143.63	-0.37	377.81	-0.53	83.01	-0.31	107.44	0.69
AWEIsh	148.69	0.01	340.20	-0.03	76.97	-0.05	97.95	0.54
WI2015	134.49	0.80	348.24	-0.80	75.71	-0.08	102.60	0.41
CDWI	106.68	*	227.60	*	67.93	*	83.89	*
MNDWI_VIs	109.92	*	254.79	*	70.92	*	86.87	*

Table 2. Extraction of water body based on different water indices

View in the Article

地区	水体指数	总体精度 /%	Kappa 系数	错分率 /%	漏分率 /%
样区a	NDWI	90.65	0.813	0.10	9.25
	MNDWI	90.55	0.811	0.40	9.05
	AWEInsh	93.70	0.874	5.25	1.05
	AWEIsh	92.65	0.853	6.20	1.15
	WI2015	94.40	0.888	4.40	1.20
	CDWI	94.60	0.892	1.30	4.10
	MNDWI_VIs	96.00	0.924	1.95	2.05
样区b	NDWI	84.65	0.693	0.15	15.20
	MNDWI	91.10	0.822	0.15	8.75
	AWEInsh	87.25	0.745	12.55	0.20
	AWEIsh	92.60	0.852	6.80	0.60
	WI2015	91.15	0.823	8.40	0.45
	CDWI	87.55	0.751	0.25	12.20
	MNDWI_VIs	93.30	0.866	0.10	6.60
样区c	NDWI	96.30	0.926	0.25	3.45
	MNDWI	95.55	0.911	0.55	3.90
	AWEInsh	96.20	0.924	3.35	0.45
	AWEIsh	97.00	0.940	2.35	0.65
	WI2015	97.20	0.944	2.20	0.60
	CDWI	97.40	0.948	0.90	1.70
	MNDWI_VIs	97.70	0.954	1.80	0.50
样区d	NDWI	92.50	0.850	0.25	7.25
	MNDWI	95.85	0.917	1.65	2.50
	AWEInsh	95.05	0.901	4.55	0.40
	AWEIsh	95.40	0.908	4.20	0.40
	WI2015	93.10	0.862	6.50	0.40
	CDWI	96.00	0.920	1.85	2.15
	MNDWI_VIs	96.90	0.942	2.25	0.55

Table 3. Accuracy comparisons among seven water indexes

View in the Article

地区	水体指数	最大水体/km²	平均水体/km²
样区a	NDWI	73.47	61.89
	MNDWI	70.32	58.51
	AWEInsh	106.59	80.18
	AWEIsh	97.01	75.64
	WI2015	94.18	74.01
	CDWI	80.25	63.05
	MNDWI_VIs	81.35	64.23
样区b	NDWI	366.54	324.62
	MNDWI	398.55	324.86
	AWEInsh	444.24	346.48
	AWEIsh	430.49	334.89
	WI2015	432.37	336.44
	CDWI	369.84	324.60
	MNDWI_VIs	402.57	324.55
样区c	NDWI	116.59	101.11
	MNDWI	115.79	101.29
	AWEInsh	176.29	140.58
	AWEIsh	180.80	146.99
	WI2015	170.67	140.38
	CDWI	122.73	110.56
	MNDWI_VIs	125.96	109.88
样区d	NDWI	72.54	69.75
	MNDWI	80.63	76.66
	AWEInsh	120.01	109.20
	AWEIsh	95.28	88.70
	WI2015	103.91	95.95
	CDWI	82.69	77.52
	MNDWI_VIs	87.57	80.55

Table 4. Extraction of different water body areas for different water indexes

Download Citation

Save the article for my favorites

Paper Information

微信扫一扫：分享

微信扫一扫：分享