Jason Parent Qian Lei University of Connecticut Land cover and land use Land cover the physical material on the earths surface eg water grass asphalt etc Land use the use of the land by humans eg reservoir agriculture parking lot etc ID: 595640
Download Presentation The PPT/PDF document "A Fully Automated Approach to Classifyin..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
A Fully Automated Approach to Classifying Urban Land Use and Cover from LiDAR, Multi-spectral Imagery, and Ancillary Data
Jason Parent
Qian Lei
University of ConnecticutSlide2
Land cover and land use
Land cover: the physical material on the earth’s surface (e.g. water, grass, asphalt, etc.)Land use: the use of the land by humans (e.g. reservoir, agriculture, parking lot, etc.)Fundamental to landscape analyses and urban planning.
2Slide3
Opportunities and challenges for high resolution data
Increasing availability of airborne light detection and ranging (LiDAR) and aerial imagery offers opportunities to study landscapes in great detail.Technically challenging to process…require lots of hard drive
space.datasets must be divided into small subsets for processing.c
onventional algorithms not well suited to processing large numbers of subsetsSlide4
Study objectives and justificationDevelop fully automated algorithm
to classify high resolution (1-meter) land cover / land use which is applicable over large areas.no previously presented algorithm has been feasible to apply over large areas.Specifically, we developed python scripts with ArcGIS to…
classify 1-meter land cover from LiDAR and multispectral data.i
nfer land use from object geometry and spatial context of land cover features.4Slide5
Study areaLocated in eastern Connecticut in the northeastern U.S.
Semi-random stratified sample of 30 1x1 km tiles.Stratified by % impervious cover (according to Connecticut’s Changing Landscape land cover data).
5
0 -
33
33
-
66
66
–
100
% impervious
4800 km
2Slide6
DataLiDAR
2010 leaf-off fall acquisitionSmall footprint (44 cm)Near-infrared (1064 nm)> 1.5 pts/m2
6
Aerial
orthophotos
2012 leaf-off spring acquisition
Blue
, green, red, and
NIR
0.3 meter
resolutionSlide7
Land cover classification rules7
Land cover
Primary characteristicsBuildingHeight > 2.5m; no ground returns
Low impervious cover (low IC)Low NDVI; no returns 2 to 4.5 meters above groundDeciduous
forestHeight > 3m; high NDVIConiferous forestHeight > 3m; very high NDVIMedium vegetation
Height 0.5
to
3m; high NDVI
Water
No returns
Riparian
wetlands
Low
reflectance in all bands; adjacent to water
Low vegetation
High return
intensity
Pixel- and object-based rules using structural and spectral propertiesSlide8
Land cover classification example
8
deciduous
coniferous
med. veg.
low veg.
water
wetland
building
low ICSlide9
Land cover class accuracies
ClassUser acc. (%)Prod. acc. (%)Water
9685
Building99
97Low vegetation91
94
Wetland
26
35
Low impervious
93
91
Med. vegetation
61
60
Coniferous trees
90
76
Deciduous
trees
95
96
93%
overall
Kappa = 0.90
n = 3200
User accuracy
: probability that a cell label is
correct.
Producer accuracy
: probability that a cell is correctly labelled
.Slide10
Land use classification rules10
Building use
Primary characteristicNon-ResidentialLarge parking area;
flat roof; large building sizeMulti-family residential
Large parking area; narrow building width; similar building shapesSingle family residentialSmall parking area;
peak roof; small building size
Parcel cadastral information not used because of limited availability.
Object- and parcel-based rules using object shape/size and parcel land cover compositionSlide11
Land use preliminary results
deciduous
coniferous
med. veg.
low veg.waterwetland
building
low IC
multi-family
non-
resid
.
single-familySlide12
Land use classification assessment
12small commercial buildings misclassified as single family due to similar structural characteristics
problems caused by mismatch between land cover and parcel data
Qualitative assessment notes…
12Slide13
Conclusions and future work
Land cover classification:
Use of airborne LiDAR and multi-spectral data proved highly effective in classification of high resolution land cover.
Developed fully automated algorithm that performs well over large area.
13
Land use classification:
Use of building shape and context is promising
Future
work will develop rules for classification
of…
r
oads
vs. parking
lots
u
rban
vs. non-urban
forest
a
griculture
vs.
turfSlide14
A Fully Automated Approach to Classifying Urban Land Use and Cover from LiDAR, Multi-spectral Imagery, and Ancillary Data
Jason Parent
Qian Lei
University of Connecticut
Questions?