spectral classification of worldview-2 multi-angle sequence atlanta city-model derived from a...
TRANSCRIPT
Spectral classificationof WorldView-2 multi-angle sequence
Atlanta city-model derived from a WorldView-2 multi-sequence acquisition
N. Longbotham, C. Bleilery, C. Chaapel,C. Padwick, W. J. Emery, and F. Pacifici
Outline
2
This presentation illustrates the unique aspects of the WorldView-2 satellite platform by combining multi-spectral information with multi-angle observations
The previous presentation dealt with very high spatial resolution imagery with multi-angle observations
What can we do with this kind of data set?
Four experiments have been carried out to investigate the classification contribution of multi-angle reflectance (MAR) as well as different feature extraction data sets (reducing the large size of the raw data space)
Methodology (1/2)
3
13 Multispectral Images
Atmospheric Correction
13 Panchromatic Images
Digital Surface Model
13 Multispectral True-Ortho ImagesPolynomial Multispectral
Nadir Multispectral
Multi-angle Multispectral
Principal Component Analysis
Methodology (2/2)
4
Polynomial Multispectral
Nadir Multispectral
Multi-angle Multispectral PCA
y = ax2 + bx + cPoly fit standard error
104 bands 32bands 10 bands8 bands
Information Sources
7
The MAR contains a partial bidirectional reflectance distribution function (BRDF) over a single satellite track at a single sun angle
Objects with pitched surfaces, such as trees and residential roofs, will present a different observational cross-section at each angle
Surfaces with varying reflectance in both time and angle can be described by an error term that encapsulates the variation of a pixel through the multi-angle sequence
Varying reflectance in both time and angle (1/2)
10
Differentiates land-use of similar spectral signature– low vs. high volume traffic roads
Multi-angle spectral variability– stationary vehicles
Four Experiments
The most-nadir multi-spectral image is used as base-case
12
Exp. 9 Exp. 10 Exp. 11 Exp. 12
Nadir Multispectral X
Multi-angle Multispectral X
Polynomial Multispectral X
Principal Component Analysis X
Classification and Validation
15 classes of interest have been selected representing a wide variety of both natural and man-made land-covers, including different kind of roof, roads, and vegetation
Training: 50 samples per class
Validation: 90,000 of independent samples
Each of the classification experiments are conducted using the Random Forest algorithm
13
Flat Roof
Pitched Roof
Concrete
Pavement
Parking Lot
Healthy Vegetation
Stressed Vegetation
Dormant Vegetation
Soil
Evergreen Trees
Deciduous Trees
Parked Cars
Recreational
Shadow
Water
Results (1/2)
14
Exp. 9 Exp. 10 Exp. 11 Exp. 12
Nadir X
Multi-angle X
Polynomial X
PCA X
Exp. 9
Exp. 10
Exp. 11
Exp. 12
Flat Roof 55.4 62.8 54.6 57.2
Pitched Roof 35.6 54.8 63.2 65.8
Concrete 84.7 94.2 89.1 91.0
Pavement 54.1 80.5 80.2 84.0
Parking Lot 76.4 82.0 90.7 89.3
Healthy Vegetation 94.1 96.7 96.0 96.2
Stressed Vegetation 94.7 94.8 91.4 90.4
Dormant Vegetation 80.8 80.8 82.0 85.4
Soil 90.2 94.8 92.5 96.7
Evergreen Trees 77.2 85.6 92.4 93.0
Deciduous Trees 85.9 94.8 95.8 91.0
Parked Cars 28.0 43.0 64.7 49.6
Recreational 88.8 96.2 96.7 97.1
Shadow 89.3 94.8 93.0 89.9
Water 95.1 95.9 83.4 83.2
Exp. 9 Exp. 10 Exp. 11 Exp. 12
Results (2/2)
15
Flat Roof
Pitched Roof
Concrete
Pavement
Parking Lot
Healthy Vegetation
Stressed Vegetation
Dormant Vegetation
Soil
Evergreen Trees
Deciduous Trees
Parked Cars
Recreational
Shadow
Water
Detail
16
• Parked Cars
• Empty Parking Spots
• Pitched Roofs
• Deciduous Trees
• Stressed/Dormant Grass
• Road
Flat Roof
Pitched Roof
Concrete
Pavement
Parking Lot
Healthy Vegetation
Stressed Vegetation
Dormant Vegetation
Soil
Evergreen Trees
Deciduous Trees
Parked Cars
Recreational
Shadow
Water
Conclusions
18
This study showed that there is significant improvement in classification accuracy available from the spectral data in a multi-angle WorldView-2 image sequence.
Four spectral classification experiments were separately presented using a nadir multi-spectral image, the full multi-angle multispectral data set, and two feature extraction techniques.
The multi-angle spectral information provided 14% improvement in kappa coefficient over the base case of a single nadir multispectral image.
Specific classes benefited from the unique aspects of the multi-angle information:– The classes car and highway are of particular interest