Image Classification - Heriot
Download
Report
Transcript Image Classification - Heriot
Image Classification
MSc Image Processing
Assignment
March 2003
1
Summary
Introduction
Classification using neural networks
Perceptron
Multilayer perceptron
Applications
2
Introduction
Definition
Assignment of a physical object to one of
several pre-specified categories
Unsupervised
Supervised
For more details
See Image Processing course
3
Classification
Classification
Supervised
Pattern
recognition
Parametric
Bayes
Unsupervised
k-means
Fuzzy k-mean
Algebraic
Non-parametric
Neural nets
SVM
Minimum distance
K-nearest neighbour
Decision trees
4
Neural nets
Inspired by the human brain
Useful for
Classification
Regression
Optimization …
5
Model
x1
.
.
.
.
.
w1
wn
f
y=f(wi xi + w0)
x. n
x=(x1…xn) input vector
w=(w0…wn) weight vector
f activation function
6
Perceptron
f=sign
1
-1
2 inputs
w1x1+w2x2+w0=0
7
Perceptron (2)
Example: AND function
1
w0=1
x1
w1=1
w2=1
x2
x1 -1
x2
1
-1
-1
-1
1
-1
1
sign
x2
-1+x1+x2=0
-
+
x1
8
Perceptron (3)
Algorithm
Minimise set of misclassified examples
Gradient ascent
Converges if data linearly separable
Demo
9
Perceptron (4)
XOR problem
Problem when
Data non-linearly separable
Solution: change activation function
For more details
Matlab classification toolbox
http://tiger.technion.ac.il/~eladyt/Classification_toolbox.html
10
Multilayer Perceptron (MLP)
Able to model
complex non-linear
functions
Hidden layers with
neurons
Backpropagation
algorithm
outputs
inputs
11
MLP (2)
f=sigmoid
y
w0
1
y (x)
a
1 e
w1
x1
w2
x2
n
a w0 wi xi
i 1
12
MLP demo
Matlab Classification Toolbox
Handwritten digits classification
Discriminate between 10 digits
13
MLP demo (2)
Pre-processing
Feature extraction
Choice of neural network
Training
Test
Input
layer
For more details
See our program
1st hidden
layer
2nd hidden
layer
Output
layer
F
E
A
T
U
R
E
S
8 features
10
neurons
10
neurons
O
U
T
P
U
T
10
neurons
10
neurons
14
MLP performance
Able to model complex, nonlinear
mapping and classification
Can be trained by examples, no
mathematical description needed
In practice, shows good results
15
MLP limitations
Extensive training data must be
available
Computation time
Curse of dimensionality
Generalisation
Overfitting
To go further
See Neural Network Toolbox, demo on generalisation
16
A few applications
Medicine
Defence
Radar & Sonar
Finance …
17
Thank you.
18