arnt-børre salberg and rune solberg norwegian computing center
Post on 10-Feb-2016
42 Views
Preview:
DESCRIPTION
TRANSCRIPT
www.nr.noearthobs.nr.no
Land cover classification of cloud- and snow-contaminated multi-temporal high-resolutionsatellite images
Arnt-Børre Salberg and Rune SolbergNorwegian Computing Center
3rd Workshop of the EARSeL SiG Remote Sensing of Land Use and Land Cover, 25 - 27 November 2009, Bonn, Germany
www.nr.noearthobs.nr.no
Overview
► Motivation & challenges
► Missing data mechanism
► Classification with missing observations
► Image restoration
► Experiments & Results
► Summary and Discussions
www.nr.noearthobs.nr.no
Motivation –Land cover classification
Classifier
Multi-spectral image Thematic map
Featurevector
Label
www.nr.noearthobs.nr.no
Multi-temporal land cover classification► Land cover classification using high-resolution
optical remote sensing can be challenging since:▪ In Northern Europe clouds and snow prevent us
from observing the surface of the earth.▪ High-resolution images has often a low temporal
coverage.
► Multi-temporal land cover classification▪ Enhanced performance since we observe the
vegetation at different phenological states.▪ The set of cloud contaminated images have
observed a higher portion of the earth’s surface than a single image.
www.nr.noearthobs.nr.no
Multi-temporal land cover classification by pixel level fusion
Multi-temporal & Multi-spectral imagesThematic map
Featurevector Label
Classifier
www.nr.noearthobs.nr.no
Challenges – Pixel level fusion?
Image 1 X X X X
Image 2 X X X X X
Image 3 X X X X X X
Pixel no. 1 2 3 4 5 6 7 8 9 10 11 12
Typical missing data pattern
► How should we handle the missing observations?
www.nr.noearthobs.nr.no
Handling missing observations
Proposed approach:► Identify the missing observations.
► Identify the missing data mechanism.
► Construct classifiers capable of handling data with missing features and a given missing data mechanism.
www.nr.noearthobs.nr.no
Identify missing observations
► Cloud/snow detection▪ Classify the images into the categories: Cloud,
snow, water and vegetation/soil/rock.▪ Constructed a missing data indicator ri for each
pixel
► Assume perfect cloud/snow detection
www.nr.noearthobs.nr.no
Identify the missing data mechanisms► Missing completely at random (MCAR)
▪ Landsat 7 sensor failure.
► Missing at random (MAR)▪ Clouds
► Not MAR▪ Snow, censoring of measurements
www.nr.noearthobs.nr.no
Classification with missing observations
Some existing approaches► Mean value or zero substitution
▪ Biased estimates
► Remote sensing▪ Aksoy et al. 2009 ▪ Decision tree based approach
www.nr.noearthobs.nr.no
Classification with missing observationsLet x(k) denote the part of x corresponding to the missing data indicator vector rk
Optimal classifier (Mojirsheibani & Montazeri, 2007)
Let be a binary vector with 0 at the element j if the jth element of x is missing, and 1 otherwise
www.nr.noearthobs.nr.no
Classification with missing observations
► Missing data mechanism introduces an additional probability
▪ Depends on feature vector and land cover class.
► MCAR:
▪ Classifier reduces to the marginal distribution where the missing features are integrated out.
www.nr.noearthobs.nr.no
► Unknown parameters need be estimated when applying parametric classifiers
► Only use complete feature vectors for learning ▪ May be only a few available
► Expectation Maximization algorithm often applied for Gaussian distributions or mixture Gaussian distributions
► Parametric classifiers difficult since
is unknown and hard to estimate.
Parametric classifiers
www.nr.noearthobs.nr.no
► K-NN classifier for not MAR scenarios:
▪ kNN classifier works on the selection of samples among the training data that has the exact same missing data pattern as the test vector, and perform the kNN rule among these samples
Non-parametric classifiers
www.nr.noearthobs.nr.no
Two-stage classifier
www.nr.noearthobs.nr.no
► Assume that a land cover map is available (from the classification module)
► Minimum mean-squared error estimator (assuming Gaussian distributions)
▪ Dependent on the land cover class of the given pixel. c and c estimated using the EM algorithm (MAR
assumption)
Image restoration
www.nr.noearthobs.nr.no
Experiments & Results
► Land cover classification of mountain vegetation important for biomass estimation of lichen.▪ Remote sensing data: 4 Landsat 7 ETM+ images
(2004-05-31, 2000-07-23, 2002-08-14, and 2002-09-15)
▪ Ancillary data: Slope and elevation derived from a digital elevation model (DEM).
▪ In situ data: 4861 pixels were labeled according to the classes: water, ridge, leeside, snowbed, mire, forest and rock.
www.nr.noearthobs.nr.no
Results– Land cover classification
Input images Missing data indicators Thematic map
www.nr.noearthobs.nr.no
Results – Cloud removal
Input image Restored imageCloud shadows
► Image restoration of July 23 using Aug. 14 and Sep. 15 images.
www.nr.noearthobs.nr.no
Results – snow and sensor failure removal
Input image Restored image
► Image restoration of May 31 image using July 23, Aug. 14 and Sep. 15 images.
► Note that at May 31 the vegetation is in a different phenological state than for the other images.
www.nr.noearthobs.nr.no
Classification resultsMethod July 23
2000Aug 14 2002
Sep. 15 2002
DEM Acc. excl. missing data
Acc. incl. missing data
Portion classified
Gauss. X 69% 52% 63%
X 63% 57% 76%
X 67% 66% 99%
X X X X 78% 78% 100%
K-NN X 68% 51% 63%
X 63% 57% 76%
X 67% 66% 99%
X X X X 81% 81% 100%
www.nr.noearthobs.nr.no
Summary and discussions
► Proposed a two-stage approach▪ Cloud/snow classification▪ Vegetation type classification with missing observations
► Obtained increased classification power by pixel level fusion of cloud and snow contaminated satellite images
► Image restoration natural by product and seem to work good for some areas.▪ Cloud shadows remains a challenge.▪ Difficult for not MAR
www.nr.noearthobs.nr.no
Summary and discussions
► Further improvement in classification accuracy expected by▪ Proper feature extraction▪ Contextual classification (e.g. Markov Random Field)▪ Including ancillary data important for mountain
vegetation (e.g. bio-climatic variables)▪ Multi-sensor fusion with full polarimetric SAR
images?▪ Identification of cloud shadows▪ Topographic illumination correction (c-correction)
www.nr.noearthobs.nr.no
Thank you
top related