免费文献传递   相关文献

Study on Highland Wetlands Remote Sensing Classification Based on Decision Tree Algorithm

基于决策树的高寒湿地类型遥感分类方法研究



全 文 :!"#$%&
 2011,24(4):464 469
ForestResearch
  
!"#$
:10011498(2011)04046406
k234š5Ä56G@¤78?@VWÎÏ
zLb

{|
,
}~

D
 
P

(!"#$%&)ó;B¯%&*

56
 100091)
ABCD
:20101111
EFGH

þÿIJK#%)*Em#%"!OMGóF

ST³èBí|Ì;ŸæUŽM³è%&
(IFRIT200906)”;
(+

ç,
P

#fdC01GHÕQ

•fg@_ºô‡ÈÉÿÈ
(2006BAD23B06)”
PQRS

{n€
(1982—),
$

é!¿Hþ

XY

%&]^¼GŽM¢•fg%&

/PQ:ÜD:(1973—),$,Çî8ëþ,%&Z,EYmƒš,[\%&°4!"1¢è¼fg‡ŽM¢•fg。
89

~Íê

_0#U4!

BC”|Ì;U4¼GŽMl‘—\]Õ

ÊÉ
TM
÷FÎDŽ
DEM
eGH³h
à™×



Ž×

~ec,è\@ÏÎ
(NDWI)
äZ†­µÏÐ

£:nI+¼æ

z%&UéÌM\“LU
—

¡à/¥‡pœ(=s¡Õ¢J—\*ô䁙š“Lz·

™š›œ

ÊÉE½Ï΁nI+—\]Õz
¼GŽM\æ“L—,
¦pœ(=s¡Õ¢J—\"@N×p¼
12.05%;
"@
kappa
_Îp¼
0.1407;



ÇK



MMäŽM\æ

mëQN׎ÉNNחµp¼”
6.06%,6.25%;0.12%,3.13%;6.99%,
25.00%;6.12%,28.13%,
·¢J—\»Øœ®p¼

ȜE½Ï΁nI+—\]Õ%¼GU4ŽMl‘—
\,cØvêL

:;<

|Ì;¼GŽM

nI+—;NDWI;
GH³h
=>?@$
:S7718
!ABCD
:A
StudyonHighlandWetlandsRemoteSensingClassification
BasedonDecisionTreeAlgorithm
ZOUWentao,ZHANGHuaiqing,JUHongbo,LIUHua
(ResearchInstituteofForestResourceInformationTechniques,ChineseAcademyofForestry,Beijing 100091,China)
Abstract:SuojiaQumaheNatureReserve,whichlocatesinthesourceregionofThreeRivers(YangtzeRiver,
YelowRiverandLancangRiver),wastakenastheresearchfieldtodiscussthepropermethodforremotesensing
classificationofhighlandwetlands.TheTMimages,DEM,NDWI(NormalizedDiferenceWaterIndices)andthe
brightness,greennessandhumidityafterthetasseledcaptransformationwereusedastheindicatorstoestablishthe
decisiontreemodeltodistinguishthediferentwetlandsandotherlandcovertypes.Theauthorscomparedthere
sultswiththetraditionalmaximumlikelihoodsupervisedclassification,itshowedthatthedecisiontreemethodbased
ontheindicescanimprovetheoveralaccuracyby12.05%,andtheoveralkappacoeficientby0.1407.Forriv
ers,lakes,swampsandfloodplains,theproducer’saccuracyanduser’saccuracyincreasedby6.06%,6.25%;
0.12%,3.13%;6.99%,25.00% and6.12%,28.13% respectively.Theresultsofthisstudysuggestthatthe
decisiontreemethodbasedonindicesisanefectivetoolforwetlandsremotesensingclassificationinhighlandarea.
Keywords:highlandwetlands;decisiontree;NDWI;tasseledcaptransformation
  
|Ì;U4¼GŽMß%&ˆ‹õ(¼

#
$(=

\æ(½vŽMm_œ

ÒÚUi
WáST³¬÷ø

|Ì;UŽMm_œ.
N~m”™£³èŽ?uc4

ÊÉl‘fg
%&|Ì;U¼GŽMô?[1-2],
Šý%&͸ÉþòH¢+ž]ՓLl‘÷F
—,
võV

N3\«¼2Ɣ^Ձ‹É

þò
µN>¶·Õ

d^^]‘]Õ

#^z<—\än
û4
D {nۊ

E½nI+¼GŽM\æl‘—\]Õ%&
Epy]Õ
[3-5]
1.NTf¾ä”ŽMB¯py


0ŠýnEpy]Õ0×ÕZƒ

¼—Oõ÷
Fôm¦¼äPQ

„uFÈÉ؋É

F=È
ɁŽMB¯py

3èLbŸÕ
[6]、
èL5_@
×·¦Õ
[7]
Íûɽz=½,–#$=Pæœ@“Lpy
[8]。
z½LÍ

MMä}6\恎M

3½‹ÉÎDmk“LU—ú·¦Qj

0w

,
ý1~–]›ÿŽMŽ×vž

Œ
:NDWI,
GH³
hàŽ×—]
)、
jTFGvž
(NDVI)
äE½è
L@ׁÏΰ³]Tf¾”ŽM—\%&

Š\]Õî3”ŽMmkŽ×¦=vž

)
î3”Šý\æŽMjTFGvž

饊ý]
Õ4¡?F,–,2

Œ
:NDWI
zlœ\@\æ
ŽMnÈ4z¦›
;NDVI
o½}mkzVjTFG
U4nÈél‘

*~/()®#MM

³D®G
MŽVG׿MœF,g

ˆKŠý­1‹ôŠ
ý]ÕFtzjTFG×ÂV伣צ=U
4

zŽM\æ—\e}6M\—\—\N×
zëm/=÷ø

nI+]Õ.NF>Í°4TȜ%,cN×
¦¼l‘—\]Õ

0F|Ì;U¼GŽMB¯
pyðaØ89

0w

m%&/¥nI+—]Õ

E½÷F³hàvžÏÎ

3ÈnI+]Õ
F¼GŽMB¯l‘—\vš

1 
VWX\]
dyB½ü‹9GîARÍ<Ž_0ÿ<÷Ø
Íê

_0#n¡]÷U%&z<

^UN½
92°47′ 95°00′E,33°34′ 35°51′N,
MAUBee
ê ü~1—åŒÒ

*ÿ

Ò
1 
%&UMABQUBeeê~1—åÿtÒ
%&U!Qì…

CST²

XuzXYD

Ú
»SO
-2.5℃,
Ú»î\]5
400mm,
‹õ
4230 5600m,
{Ÿæ¼÷=áKST

[\j
T\ætÇtU
(StipapurpureaGriseb.)、
VW}
DR
(SausureawoodianaHemsl.)、
rX¿
(Leon
topodiumnanum(Hook.f.etThoms)Hand.Mazz.)
ä¼G¿÷jT

Ž~ґ¿
(Kobresiaschoe
noides)、
r‘¿
(Kobresiahumilis(C.A.Mey)Serg.)
[LÍjT

[\ef\æ¼÷¿÷e

¼
÷¿)eŽ¼÷LÍe

2 
no623
2.1 
QRr*9,Œ
¸Él‘÷FFG%&U
2006
Ú

D


136-36
Ž
2007
Ú

D
19

137-35/36
|

Landsat-5TM
÷F

䂗Oõ
30m。
Êl
‘÷FÎDê

ð̺”FG%&U
1∶10
ÿMŒ
Ò
、1∶10
ÿŽM—åÒ
、1∶10
ÿjTÒäPÂ"
ÎD

2.2 
78?@^S
wD
2008
Ú


2009
Ú

Dz%&U“L
eê ü*Ry
652
|eê~1

Äî

ŽM
I5

—\_œ~e

W(ŽMó; ü‡¢•f
g¤Ï

(+ŽM¦—Ðå

™†%&U2´M—åŽjT—åEs

®m%&[\M”\æñ
–#Ç

ÇK

¿M



MM

³D®GMŽ7
M

ceMFG\æ

2.3 
34š?@VW
nI+%o,_:=Y+£ô+Œ—§

wD¤–Ùͤç

éÍM®÷FF>—Z
ô4zÌúÎD¤ºUñ–÷Fx|F>*{
`ñ\æ

nI+—\]ÕÅد—R[

1L


Ö×]v1
[9]。
oÓnI+—\ð1~
¦þMz—\¥ÏŽ™š“L+?

u×/þ›
ÿéÌ\æ߂4i5_

0w
,¸
ÉnI+
]Õz%&U¼GU4eMÊÉ

eMFG“Ll
‘—\%&

564
! 
"
 
#
 
$
 
%
 
& û
24
ü
2.4 
?@6¦5
2.4.1 
fg@h7iX

jXklX>m
 
GH³
h%F%&
MSS
÷FnP›P”ŽjTm²¥
ÏpqU

%,cͺ Ó䂁«K³
h
[10],
/¥z Ó䂵¶‹jT‡ef Óv
žô~Øv—’

GH³hà™šØœñ”A


ÝÓ^³h¥ÏB¯]\]/ƒ

z
TM
ÎD“LGH³hNA
[11],
ym%&\‹É-

|—]

Å@Ö×]όí

™×
=0.3037(TM1)+02793(TM2)+0.474
3(TM3)+0.5585(TM4)+0.5082
(TM5)+0.1863(TM7)

=-0.2848(TM1)-0.2435(TM2)-
0.5436(TM3)+0.7243(TM4)+0.084
0(TM5)-0.1800(TM7)
Ž×
=0.1509(TM1)+0.1973(TM2)+0.327
9(TM3)+0.3406(TM4)-0.7112
(TM5)-0.4572(TM7)
GH³hàAח]1~zM›jTFG
?s“L/þÃK

Ž×—]ç1~¦þnP
\@~eef\—3]

m%&mØdº
ND
VI
PzjTFGÏׁÏÿÏÐ

%0
NDVI
zefB³è¦l‘

FzVFG×jTF
GU4‹É
NDVI


zƒx¥¼oÖjTFGÏ
×

}z¼FG×jTFGU4çz?F^ŽÑ
<

ÝF%&U

=]—åMVFG׿M

¼F
G׿M

~e³D®GMŽ7M

.؁%&™š
›œ
,NDVI
oéuçèz^U4jTFGÏׁ¦
—

GH³h÷A%/¥«K³hŽÍºä‚
¶h

®efŽjTB¯“÷äͺä‚

oÓFÍ
ºä‚jT?sŽef™×;%4i³¯

0
w

1~¦þ®jT‡efvž—’=

ÂÝcƒ
efBzjTFG?s÷ø

*~

î3ÉG
H³hàA׎™×—]Uz}“L¦—

2.4.2 
no?89P4aNDWI 1996
Ú
McFeeters[12]
pq”c,謭\@ÏÎ
(Normal
izedDiferenceWaterIndex,NDWI),
ɽ\@p
y

}I½Œí

NDWI=(Green-Nir)/(Green+Nir)

,Green
A èL
,Nir
ÒÏêèL

/¥z%&U
NDWI
ÒF“L—˜

ŒšÊ
É
NDWI
Upy\@

zƒx=]³D®GM‡
}œ_

î3
NDWI
Å؄z=,Kœ\@nÈÑ
`

ÝzL͎Mnțv1

1~É}UU—N
ý#ǟ4èŒôL͎M

Š\L͎Mo½
}\—3]·¦=

FŽ×—]ˆ)½Ì#Ç

ÇK
4œ_

2.4.3 
pqrstuvwx
 
̺%&U
1∶10
ÿ
MŒÒ

¡à/¥a]èRy伫a]Ò

wDä
¼«mô
DEM
ÎD

N×
30m),
»~
DEM
ÎD
Eý

mô%&UK×Ò

oP,|vž]

F
nI+ć—。
3 
45678
3.1 
G@:;/0?U*34šö¤5õç
dy~ˆ³]PnI+—\vž³]

z
%&U[\M\FŠývž]ˆ ÓvK“Lœ
֗˜

œÖ™šu›
1。
Z
1 
Å9G@5:;/0
vž] ÏÐ \@ ¿M LÍ MM ³D®GM
1
³D®GM
2
7M
™× (Ÿ
29.00 90.00 51.00 80.00 95.00 67.00 112.00
(=Ÿ
60.00 112.00 87.00 89.00 119.00 82.00 158.00
»Ÿ
43.53 102.27 62.44 83.88 109.43 72.21 138.67
Ðå¬
7.53 6.37 6.72 1.61 4.42 3.45 5.26
A× (Ÿ
111.00 167.00 184.00 154.00 142.00 164.00 141.00
(=Ÿ
149.00 180.00 211.00 162.00 155.00 171.00 157.00
»Ÿ
122.69 174.32 196.87 157.79 148.02 167.59 149.26
Ðå¬
12.92 2.40 5.10 1.63 2.17 1.20 2.51
Ž× (Ÿ
113.00 58.00 73.00 79.00 62.00 82.00 51.00
(=Ÿ
138.00 79.00 101.00 88.00 79.00 93.00 83.00
»Ÿ
126.64 64.40 92.88 83.84 71.73 87.70 59.83
Ðå¬
6.86 4.78 6.42 1.43 3.19 2.37 3.67
NDWI
(Ÿ
136.00 60.00 31.00 82.00 85.00 68.00 73.00
(=Ÿ
228.00 75.00 68.00 95.00 99.00 81.00 86.00
»Ÿ
211.67 66.17 46.69 89.11 92.09 74.55 80.29
Ðå¬
12.19 2.36 6.81 2.19 2.34 2.02 2.29
664
û4
D {nۊ

E½nI+¼GŽM\æl‘—\]Õ%&
wDéÌM\ ÓvžœÖŸ

/¥nZ;
3þ‘„i]½

ÊÉH¢Ù́]Õdy(V
N3KΟPnI+‚1bŸ

‚1bŸ
ñ–e X¥Ï

/¥z·éÌbŸ—\àH
ÐM\U—Ï×%Úåñ

UÙÍ^bŸdy
%چA

™š

FŽ×—]ˆ


77
PŽM

ˆ‰#

Ç



MM

ŽlŽM\¦—b
Ÿ

‘Ž×—]ˆŸ½
77


ù^U‚
lŽM\*FU‚

‘Ž×—]ŸB½
[77,
102]


ù^bŸØLÍ[\B½U‚

Ž
ח]Ÿ=½
102


ù^U‚œ\@—å


FlŽM\U‚

¿M

³D®GMŽ7M£ô
”Š|U‚[\M。
—˜Aח]ˆéÌMyŸÈÉ

ñ–Aח]Ÿ½
153
¿M‡l
jTFGU4bŸ

z½^U47M

³D®
GM/¥™×—]ŽKדL7è

LÍU4*FU‚Ø

1~wDL͎×

A
   
×1/=v1

®LÍpyqU

z½œp¼G
׿M

/¥K×ê~bÊ

z½cí}6M,
/¥Aח]ˆdybŸ®¿M“L¦—

z½M
M

ÊÉ}F™×

A׎Kח]ˆ1·¦v
1

—µy
82,158
Ž

P|QbŸ

®}py
qU

Ý^U4؁³D®GMŽ7Mä}6M)%/¥MŒ0¤“Lpy

¦—

Fœ\@*FU4

ʔœ\@ßê

ð?FØ
#ǟL͎ƒ]§÷FG³D

³D/¥K
×e¼Ï0¤ê~bÊ

#ǟæLÍç/¥
ND
WI
‡œ\@ê~U—

z½pqœ\@

wD#
ǎÇKŒ?vž

ÊɌ?ÏÎ槡S/Pê~U
—

/¥23

ùŒ?ÏÎ
>0.11
ÇK

/¥~ˆzmU4[\M\ Óvž—˜e
éÌM\Fvž]ˆbŸñ–

:[Œí¼GŽ
Ml‘—\nI+¼æ

Ò
2),^
¼æ

[ñ‚1
ç融\«M,
Zñ‚1éçè\«
M。
Ò
2 
nI+—\¼æ
3.2 
?@¸û>
@LnI+¼æ

ôä%&Uðâ—\Ò

¡
àÊÉ0Î/

¹\ŽÞʗ˜“L—\àNA

1ôäŒí%&U—\™šÒ

Ò
3)。
3.3 
?@<ÓF=Šê 
z·3ÈE½Ï΁nI+]ÕF%&¼G
ŽMm_œl‘—\ûÈEs

/¥Ù‘{
»d1]½

dy
224
|1
,¸
ÉH¢Ù́]
Õ

znI+—\Ž(=s¡Õ¢J—\™š“
LN×3È

¢c—\]Ձœ_de

M\N
×e
kappa
_Îz·u›
2 4。
¢J—\œ_de

›
2)
1L

#LJÇK

764
! 
"
 
#
 
$
 
%
 
& û
24
ü
Ò
3 
nI+—\™š
¿MŽMM?F,–œ—Ñ<

MM)?Fé—ô


¿MÑ<

³D®GM)zé—ô¿M‡M
M

0œ—Ñ<(ÿ0%L͎MŽ¿M

L͎
M*d
32
|3È1

Ø
12
|é—ô”¿M

Ý
¿M)?FM,–é—ôL͎MEs

/¥—
\Nכ

›
4)
1~Sä

ÇKŽ7M—\N×
¦¼



L͎³D®GM—\N×Ø*í
î

¿M~eMM—\N×1·¦V

MMÉN
N×V½
60%。
¢J—\™š"@N×)>
75.45%,
ãéä2´ÈɁ\«

"@
kappa
_
Î
0.7135,
{½¦V\{

/¥ˆK—˜

(
=s¡Õ¢J—\*ô™šH2´ÈɁN×\«
ئ=¬H

éuçè2´ÈÉ

Z
2 2007
>?@

ABޖàCD?@EFGH
M\ #Ç ÇK ¿M LÍ MM ³D®GM 7M




    27 1 2 0 2 0 0 32
ÇK
    3 28 0 0 0 0 1 32
¿M
    1 0 26 4 0 1 0 32

    0 0 12 20 0 0 0 32
MM
    2 0 8 0 18 4 0 32
³D®GM
0 0 7 0 3 22 0 32
7M
    0 0 3 0 0 1 28 32


)    33 29 58 24 23 28 29 224
ÂnI+—\œ_de

›
3)
1L

tz¢
J—\#LJÇK

¿M‡MM

¿MŽL͎M?
Fœ—Ñ<

nI+—\]Õ1ؔ¦=Ïׁ
s{

L͎Mé—ô¿M3È1|Îo
12
|
î

|

¿M

MMŽ³D®GM‚é—1Î)
=´íî

Z
3 2007
>?@

ABޖà34š?@EFGH
M\ #Ç ÇK ¿M LÍ MM ³D®GM 7M




    29 1 0 0 1 1 0 32
ÇK
    2 29 0 0 1 0 0 32
¿M
    0 0 28 3 1 0 0 32

    0 0 4 28 0 0 0 32
MM
    2 0 1 0 27 2 0 32
³D®GM
0 0 4 0 2 26 0 32
7M
    0 0 3 0 0 0 29 32


)    33 30 40 31 32 29 29 224
/¥—\Nכ

›
4)
ð1~Sä

nI+—
\™š

ÇK





³D®GMŽ7M
—\Nצ¼

mëQN׎ÉNNחµ1ã

96.67%、90.63%;90.32%、87.50%;87.88%、
9063%;89.66%、81.25%
Ž
100%、90.63%。
}
6M\—\N×)1ؔ¦=Ïׁp¼

"@
—\N×1ã
87.50%,
¦¢J—\Ø
12.05%



æW1~çè2´òPÁ。kappa
_Î]
#

"@
kappa
_Î
0.8542,
¦¢J—\™š

kappa
_Îp¼”
0.1407(19.72%)。
ïˆ*K

m%&¸ÉnI+—\]ÕéB
F3|M\N×

"@N×ð%
kappa
_Î]#

1
¦¢J—\*ô™šØ¦=´×p¼

¦¼N
×u×çèF2´òPzRyÎDåñK


864
û4
D {nۊ

E½nI+¼GŽM\æl‘—\]Õ%&
Z
4 
CD?@Š34š?@MG@<Ó*
kappa^
Qê 
GH
¢J—mëQN×
/%
ÉNN×
/%
"@N×
/%
nI+—mëQN×
/%
ÉNN×
/%
"@N×
/%

81.82 84.38 87.88 90.63
ÇK
96.55 87.50 96.67 90.63
¿M
44.83 81.25 70.00 87.50

83.33 62.50 75.45 90.32 87.50 87.50
MM
78.26 56.25 84.38 84.38
³D®GM
78.57 68.75 89.66 81.25
7M
96.55 87.50 100.00 90.63
Kappa

0.7135 0.8542
4 
496:9
(1)
/¥z·—˜éÌM\ Óvž

:[
nI+¼æ

ô?z|Ì;U4¼GŽMB¯“
Lpy

oyô”
87.50%
¦¼N×

5œE½
ÎD³hàvž]
,¸
ÉnI+—\]ÕF¼G
U4“LŽMm_œ—,
%,cVôm

)2Ñ
ŽIJØv]Õ

(2)
/¥‡(=s¡Õ¢J—\™šz·1
L

E½vž]nI+—\]Õ¦(=s¡Õ"
@N×p¼
12.05%,
"@
kappa
_Îp¼
0.1407;
z½#Ç

ÇK



MMäŽM\æ

mëQN×
ŽÉNNחµp¼”
6.06%,6.25%;0.12%,
313%;6.99%,25.00%;6.12%,28.13%。
(3)
Œwdº†ûvž]%nI+]Õyô
¼Nׁ5}

m%&*dyGH³hà™
×



Ž×—]~e
NDWI
Ž
DEM
ÎD

/¥2
3Ȝ”Šý³]F%&mU4¼GŽMl‘—1LK

Ž×—]1~¦þM®ŽM‡lŽMM
\U—=

Aח]1~¦þM®¿MpyqU

Ý
³D®GM

7MäM,
F™×—]ˆçÅئ¼
×Ÿ

1~Dwz}ê~U—

/¥=·!MÎ
"MŒÒRy¼ÏeKח]

FU—LÍ

³D
®GMäM\)11~PbŸ³]

(4)
m%&*Évž]Í%E½l‘ÒF Ó
vžpyvž]

ÝÓbŸñ–%/¥%&þ
ZnZ;3URy(VN3KΟ

0wF%&
þZz^U4õfÏ×~e—\fg\{ä]#
ئ¼\«

í,âòPî3F%&¼GŽMl
‘—\f¾
QUEST
Ž
CART
ä×Õ

ÂÝ2Ñn
Eñ–—\‚1bŸ

cƒ—\¥ÏòP
]

Ì

î3Fl‘ÒF—,
÷Fˆ›ÑqU
BQ5_~eéÌM\gAvž)%Ê ÓB
¯ßê0\1ÊÉB¯

“,â%&z®M\‚
BQ5_egAvžî3änI+¼æ‘


GU4ŽMè“,â7è—。
ef!A

[1]
Ðhü

„n®

‚:E

ä
.2000
Ú~U|Ì;MU\ó;³
èl‘ ü%&
[J].
ü‹#f
,2008(5):20-26
[2]
æij

¬
 
:

¬:¾

²Ì

 #;U¼GŽME³è%
&
[J].
ŽM#$
,2007,5(4):298-304
[3]
€
 
l

Ï
 
:

k
 


E½µN¶·ìbŽMeMFG
—\%&
[J].
=ŠAò=$$8
,2004,44(4):582-588
[4]
¹ö¶

Ü͐

Ã}œ

ä

E½d^^]‘l‘÷FŽM
B¯py%&
[J].
ÖבÈÉ%&
,2008,25(4):989-991
[5]GhiocaRobrechtDM,JohnstonCA,TulbureMG.Assessingthe
useofmultiseasonQuickBirdimageryformappinginvasivespeciesin
aLakeEriecoastalmarsh[J].Wetlands,2008,28(4):1028
-1039
[6]
–:è

lmã
.TM
ÒFF=æ\eeEs•ŠAÈÉ
[J].
Ÿ÷l‘
,1996,11(1):53-58
[7]HanqiuXu.Modificationofnormalizeddiferencewaterindex(ND
WI)toenhanceopenwaterfeaturesinremotelysensedimagery[J].
InternationalJournalofRemoteSensing,2006,27(12):3025
-3033
[8]
á+n

œYo
.TM
ól\@­µfgs“
[J].
Ÿ÷l‘

1992,7(1):17-23
[9]FriedlMA,BrodleyCE,StrahlerAH.Maximizinglandcoverclas
sificationaccuraciesproducedbydecisiontreeatcontinentaltoglobal
scales[J].IEEETransactionsGeoscienceandRemoteSensing,
1999,37(2):969-977
[10]
VÁ­

–%p

Ã}œ

ä

l‘ƒB
[M].
56

¼äÑ/q
ÙÁ
,2001:155-156
[11]CristEP,KauthRJ.Thetasseledcapdemystified[J].Photogram
metricEngineeringandRemoteSensing,1986,52(1):81-85
[12]McFeetersSK.TheUseofnormalizeddiferencewaterindex
(NDWI)inthedelineationofopenwaterfeatures[J].Internation
alJournalofRemoteSensing,1996,17(7):1425-1432
964