O
pen
A
rchive
T
oulouse
A
rchive
O
uverte
(OATAO)
OATAO is an open access repository that collects the work of some Toulouse
researchers and makes it freely available over the web where possible.
This is
an author's
version published in:
https://oatao.univ-toulouse.fr/19746
Official URL :
https://informatique-mia.inra.fr/resste/sites/informatique-mia.inra.fr.resste/files/RESSTEmai2017_ALagrange.pdf#overlay-context=hierarchique
To cite this version :
Any correspondence concerning this service should be sent to the repository administrator:
tech-oatao@listes-diff.inp-toulouse.fr
Lagrange, Adrien and Dobigeon, Nicolas and Fauvel, Mathieu and May, Stéphane A Bayesian model for
joint unmixing, clustering and classification of hyperspectral data. (2017) In: Séminaire des doctorants, 28
April 2017 (Toulouse, France). (Unpublished)
OATAO
A Bayesian model for joint unmixing, clustering
and classification of hyperspectral data
Adrien Lagrange
Ph.D. student at IRIT/INP-ENSEEIHT
Supervisors: Nicolas Dobigeon (IRIT/INP-ENSEEIHT), Mathieu Fauvel (DYNAFOR - INRA/INP-ENSAT) and Stéphane May (CNES)
Context Model Experiments Conclusions and perspectives
Context
Hyperspectral imaging
Objective
Model
Spectral unmixing
Clustering
Classification
Experiments
Synthetic data
Real data
Context Model Experiments Conclusions and perspectives Hyperspectral imaging
Nature of an hyperspectral image
A remote sensing hyperspectral image is:
same area at different wavelength → hundreds of measurements per pixel,
poor spatial resolution due to sensor limitations, e.g., resolution around 10x10m per
pixel for aerial applications
0
100
200
# bands
reflectance
0
100
200
# bands
reflectance
A. Lagrange Supervisors: Nicolas Dobigeon (IRIT/INP-ENSEEIHT), Mathieu Fauvel (DYNAFOR - INRA/INP-ENSAT) and Stéphane May (CNES) A Bayesian model for joint unmixing, clustering and classification of hyperspectral data 3 of 28
eoooo
o
000000000000 0000o
•
•
Context Model Experiments Conclusions and perspectives Hyperspectral imaging
Hyperspectral image analysis
Spectral unmixing
y
p≈ Ma
py
p: p-th observation
M: endmember matrix (spectra of
elementary components)
a
p: p-th abundance vector
Classification
Maximum a posteriori (MAP) rule:
y
pbelongs to j ⇔ j = arg max
j∈Jp(j|y
p),
⇔ j = arg max
j∈Jp(j)p(y
p|j).
y
p?
buildings
water
vegetation
y
pbuildings
water
vegetation
oeooo
o
•
•
•
0000 000000000000""-...
-0----Context Model Experiments Conclusions and perspectives Hyperspectral imaging
Spectral unmixing
One illustrative example
Image: 50 × 50 pixels (Moffett field), L = 224 bands,
3 materials: vegetation, water, soil.
ooeoo
o
000000000000 0000o
•
•
Context Model Experiments Conclusions and perspectives Hyperspectral imaging
Spectral unmixing
One illustrative example
Image: 50 × 50 pixels (Moffett field), L = 224 bands,
3 materials: vegetation, water, soil.
oooeo
o
000000000000 0000o
•
•
8
[r]
5
s
0.40
Vegetot;onSl)ect<\Jm
O.Ol•
Woter•pect,um
B
5o
;
1•pect,um
1:
1:
~
E:
/
0 0 0 10 20 30 40 50 0.5 1 1.5 2 2.5 0.5 1 1.5 2 2.5 0.5 1 1.5 2 2.5wavelength (µm) wavelength (µm) wavelength (µm) AbundallCe of Yegetation Abundance of water Abundance of soil
-.:-. r·
,·
.
\_rt_ • - I
Context Model Experiments Conclusions and perspectives Hyperspectral imaging
Spectral unmixing
A matrix factorization, latent factor modeling or blind source separation problem: Y ≈ MA
1.
Principal Component Analysis (PCA)
I
Searching for
orthogonal
“principal components” (PCs) m
r,
IPCs = directions with maximal variance in the data,
IGenerally used as a dimension reduction procedure.
2.
Independent Component Analysis (ICA) (of Y
T)
I
Maximizing the statistical
independence
between the sources m
r,
ISeveral measures of independence ⇒ several algorithms.
3.
Nonnegative Matrix Factorization (NMF)
I
Searching for M et A with
positive
entries,
I
Several measures of divergence d (·|·) ⇒ several algorithms.
4.
(Fully Constrained) Spectral Mixture Analysis (SMA)
I
Positivity
constraints on m
r⇒ positive “sources”
IPositivity
and
sum-to-one
constraints on a
r⇒ mixing coefficients = proportions/concentrations/probabilities.
Context Model Experiments Conclusions and perspectives Objective
Objective
Spectral unmixing
Classification
Low-level biophysical information
High-level semantic information
Abundance vector per pixel
Unique label per pixel
Unsupervised
Supervised
=⇒
Scarcely considered jointly.
Objective
Propose a unified framework to estimate jointly a classification map
and a spectral unmixing from an hyperspectral image.
Context Model Experiments Conclusions and perspectives
Context
Hyperspectral imaging
Objective
Model
Spectral unmixing
Clustering
Classification
Experiments
Synthetic data
Real data
Context Model Experiments Conclusions and perspectives
Bayesian model
conventional linear mixing
model;
clustering of homogeneous
abundance vectors;
classification with a
non-homogeneous Markov
random field (MRF) to
promote coherence between
cluster and class labels.
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
000000 eooooooooooo 0000 o
•
•
Context Model Experiments Conclusions and perspectives
Bayesian model
conventional linear mixing
model;
clustering of homogeneous
abundance vectors;
classification with a
non-homogeneous Markov
random field (MRF) to
promote coherence between
cluster and class labels.
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
•
•
0
Context Model Experiments Conclusions and perspectives Spectral unmixing
Linear Mixture Model (1)
Linear combination of elementary signatures corrupted by an additive Gaussian noise
y
p= Ma
p+ n
pwith
y
p: measured spectrum (p = 1, . . . , P where P is the total number of pixels)
M: endmember matrix
(i.e., spectra of R elementary components, assumed to be known here)
a
p: abundance vector
n
p: noise
•
•
•
•
Context Model Experiments Conclusions and perspectives Spectral unmixing
Linear Mixture Model (2)
Prior model for the noise:
n
p|s
2∼ N (0
D, s
2I
D),
s
2|δ ∼ IG(1, δ),
p(δ|s
2) ∝
1
δ
1
R+(δ).
Context Model Experiments Conclusions and perspectives Spectral unmixing
Hierarchical model
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
000000
o
ooe
oooooooo
0000o
Context Model Experiments Conclusions and perspectives Clustering
Clustering (1)
Assumption: several unknown spectrally coherent clusters with statistically
homogeneous abundance vectors, ∀k ∈ {1, ..., K },
a
p|z
p= k, ψ
k, Σ
k∼ N (ψ
k, Σ
k) with Σ
k= diag(σ
k,1, . . . , σ
k,R)
where z1
, . . . , z
pare discrete labels identifying the belonging to the clusters.
Vague priors for cluster parameters:
I
ψ
k∼ Dir(1)
→ ensures nonnegativity and sum-to-one constraints of E [a
p|z
p= k]
(soft constraints on a
p)
I
σ
k,r∼ IG(1, 0.1)
eoooo
Context Model Experiments Conclusions and perspectives Clustering
Clustering (1)
Assumption: several unknown spectrally coherent clusters with statistically
homogeneous abundance vectors, ∀k ∈ {1, ..., K },
a
p|z
p= k, ψ
k, Σ
k∼ N (ψ
k, Σ
k) with Σ
k= diag(σ
k,1, . . . , σ
k,R)
where z1
, . . . , z
pare discrete labels identifying the belonging to the clusters.
Vague priors for cluster parameters:
I
ψ
k∼ Dir(1)
→ ensures nonnegativity and sum-to-one constraints of E [a
p|z
p= k]
(soft constraints on a
p)
Iσ
k,r∼ IG(1, 0.1)
eoooo
•
Context Model Experiments Conclusions and perspectives Clustering
Hierarchical model
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
000000
oooo
oeooo
ooo
0000o
Context Model Experiments Conclusions and perspectives Clustering
Clustering (2)
Clustering with a non-homegeneous Markov random field
P[z
p= k|zV(p)
, ω
p, q
k,ωp] ∝ exp
V
1(k, ωp, q
k,ωp) +
X
p0∈V(p)V
2(k, zp0)
with V(p) neighborhood of p, ω
pclassification label of p.
Two potentials:
To promote coherence with classification → V1(k, j, q
k,j) = log(q
k,j);
To promote spatial coherence (Potts-Markov potential) → V2
(k, z
p0) = β1
δ(k, z
p0)
with δ(·, ·) Kronecker function.
000000
oooo
ooeoo
ooo
0000o
•
•
Context Model Experiments Conclusions and perspectives Clustering
Hierarchical model
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
000000
oooo
oooeo
ooo
0000o
,
I
'
II
Context Model Experiments Conclusions and perspectives Clustering
Clustering (3)
Estimation of coefficients of interaction between high-level and low-level
information:
q
j∼ Dir(1) → q
j|z, ω ∼ Dir(n1,j
, . . . , n
K ,j)
with
n
k,j= #{p|z
p= k, ω
p= j}
In particular:
E [q
k,j|z, ω] =
n
k,jPK
i=1n
i,k= P [z
p= k|ω
p= j]
Context Model Experiments Conclusions and perspectives Classification
Classification (1)
Classification rule with a Markov random field
P[ω
p= j|ωV(p)
, c
p, η
p] ∝ exp
W
1(j, c
p, η
p) +
X
p0∈V(p)W
2(j, ω
p0)
Two potentials:
To promote coherence with labeled data
W
1(j, cp, η
p) =
log(η
p),
if j = c
plog(
1−ηp J −1),
otherwise
if p ∈ L
− log(J )
otherwise
To promote spatial coherence → W2
(j, ω
p0) = β2
δ(j, ω
p0).
000000
ooooooooo
eoo
oor::0 o•
Context Model Experiments Conclusions and perspectives Classification
Hierarchical model
Y
s
2δ
M
A
Σ
ψ
z
β
1Q
ω
β
2η
c
LObservations
Unmixing
Clustering
Classification
Context Model Experiments Conclusions and perspectives Classification
Classification (2)
Robust classification:
η
p∈ (0, 1) the confidence in label c
pprovided by user
Possibility to correct labeled data when η
p< 1
•
Context Model Experiments Conclusions and perspectives
Context
Hyperspectral imaging
Objective
Model
Spectral unmixing
Clustering
Classification
Experiments
Synthetic data
Real data
Context Model Experiments Conclusions and perspectives Synthetic data
Dataset
413 spectral bands
SNR = 30dB
Clustering generated with
Potts-Markov MRF
Classes created by aggregating several
clusters
Image 1: 3 clusters, 2 classes, 3
endmenbers, 100x100px
Image 2: 12 clusters, 5 classes, 9
endmembers, 200x200px
(a)
(b)
(c)
(d)
Classification map: (a) image 1, (b) image 2;
Clusters: (c) image 1, (d) image 2
000000 000000000000
eoo
o
o
•
•
•
•
•
•
Context Model Experiments Conclusions and perspectives Synthetic data
Results
0
50
100
150
0
0.2
0.4
0.6
0.8
1
# iteration
v-mesure
0
50
100
150
0
0.2
0.4
0.6
0.8
1
# iteration
v-mesure
(a)
(b)
Proposed model in blue, model without classification stage (Eches et al.).(a) Clustering
convergence for image 1, (b) Clustering convergence for image 2
000000 000000000000
oeo
o
o
....
+ +- + +-+ +- + +-+ +T
T
T
T
Context Model Experiments Conclusions and perspectives Synthetic data
Results
Provided labeled data
Classification obtained
Deterioration of labeled data (40% of
error)
Confidence set to 60%
⇒
Correction of mislabeled pixels
000000 000000000000
ooe
o
o
•
•
Context Model Experiments Conclusions and perspectives Real data
Dataset
349 spectral bands
10 endmembers extracted with VCA
6 classes (straw cereal, summer crop,
wooded area, artificial surfaces, bare
soil, pasture)
Top half of groundtruth provided as
labeled data
(a)
(b)
(c)
(d)
Muesli dataset: (a) colored composition of
data, (b) groundtruth, (c) obtained
clustering and (d) obtained classification
000000 000000000000
ooo
e
o
•
•
•
Context Model Experiments Conclusions and perspectives
Context
Hyperspectral imaging
Objective
Model
Spectral unmixing
Clustering
Classification
Experiments
Synthetic data
Real data
Context Model Experiments Conclusions and perspectives