[PPT] - OUTLINE Introduction Supervised ,Unsupervised And Semi-Supervised PowerPoint Presentation

SLIDE 1

Presentation in

A Bayesian Approach To Satellite Aircraft

Image Identification Using Invariant Moments. Dickson Gichaga Wambaa F56/76676/2009 Supervised By Professor Elijah Mwangi University Of Nairobi Electrical And Information Engineering Dept.

19th September 2013 SEMINAR PRESENTATION

SLIDE 2

OUTLINE

Introduction
Supervised ,Unsupervised And Semi-Supervised

Learning.

Classifiers
Feature Selection
Clustering
Results
Optimization
Conclusion
References

SLIDE 3

COMPUTER VISION

A Computer vision system

captures images via a camera and analyzes them to produce descriptions

f what is imaged.

SLIDE 4

PATTERN RECOGNITION

Pattern

recognition is the discipline whose goal is the classification of objects into a number of categories or classes. These objects can be images or signal waveforms.

SLIDE 5

Aircraft Images

All aircraft are built with the same

basic elements:

– Wings – Engine(s) – Fuselage – Mechanical Controls – Tail assembly.

The differences of these elements

distinguish their structures or images.

SLIDE 6

SUPERVISED AND UNSUPERVISED CLASSIFICATION

Supervised learning involves training

data of known classes in an image database.

If the training data is not available then it

is known as unsupervised learning or clustering.

And if some of the training data is of

known classes then it is known as Semi-Supervised learning.

SLIDE 7

TASK

Design a classifier in a

pattern recognition system.

SLIDE 8

Stages in the design of a classification system.

SENSOR FEATURE GENERATION FEATURE SELECTION CLASSIFIER DESIGN EVALUATION

SLIDE 9

TYPES OF CLASSIFIERS

Classifiers Based on Bayes

Decision Theory.

SLIDE 10

BAYESIAN DECISION THEORY CLASSIFIERS

Based on probabilistic

statistical nature of generated features.

SLIDE 11

LINEAR CLASSIFIERS

Result from a set of

linear discriminant functions.

SLIDE 12

NON-LINEAR CLASSIFIERS

Problems that are not

linearly separable

SLIDE 13

FEATURE SELECTION

The number of features

should be reduced to a sufficient minimum to minimize on Computational complexity.

SLIDE 14

CLASSIFIER PARAMETERS

For a finite number N of training

patterns, the number of features l should be as small as possible so as to design a classifier with good generalization capabilities. The ratio N/l is also used in the performance evaluation stage .

SLIDE 15

FEATURE GENERATION

If I (x, y) Is a continuous image function, Its

geometric moment of order p + q is defined as

Geometric moments provide rich information about an image.

SLIDE 16

The Seven Moments of Hu

A set of seven moments

that are invariant under the actions

f

translation, scaling, and rotation.

SLIDE 17

HU Moments

f1 = h20 + h02
f2 = (h20 - h02)2 + 4h11

2

f3 = (h30 - 3h12)2 + (h03 - 3h21)2
f4 = (h30 + h12)2 + (h03 + h21)2
f5 = (3h30 - 3h12)(h30 + h12)[(h30 + h12)2-3(h21 + h03)2] +

(3h21 - h03)(h21 + h03) [3(h30 + h12)2 - (h21 + h03)2]

f6 = (h20 - h02)[(h30 + h12)2 - (h21 + h03)2]+ 4h11(h30 +

h12)(h21 + h03)

f7 = (3h21 - h03)(h30 + h12)[(h30 + h12)2- 3(h21 + h03)2] +
(3h12 - h30)(h21 + h03) [3(h30 + h12)2 - (h21 + h30)2]

SLIDE 18

CLUSTERING

Clustering is unsupervised

Classification.

VECTOR QUANTIZATION IS A

COST FUNCTION BASED CLUSTERING.

SLIDE 19

STEPS IN CLUSTERING TASK

FEATURE SELECTION
PROXIMITY MEASURES
CLUSTERING CRITERION
CLUSTERING ALGORITHM
VALIDATION OF RESULTS

SLIDE 20

RESULTS

Фi Original Image Scaled Image Rotated Image Φ1 6.500 6.500 6.500 Φ2 16.3201 16.3200 16.3201 Φ3 25.5669 25.5669 25.5669 Φ4 25.8880 25.8880 25.8880 Φ5 43.3000 43.3001 43.3000 Φ6 34.0960 34.0960 34.0961 Φ7 47.395 47.395 47.395

SLIDE 21

INVARIANT MOMENTS FOR 10 AIRCRAFTS

Aircraft Image

Ф1 Ф2 Ф3 Ф4 Ф5 Ф6 Ф7

1

Apache

7.1729 16.6723 19.7413 21.8784 42.8038 30.2146 47.1336

2

A5

7.1487 20.2793 22.4129 24.4962 48.0614 34.6401 50.1980

3

C5

7.0341 16.2207 18.1325 20.3399 39.7008 28.6312 45.6856

4

Mig23

7.1921 17.7858 19.5198 21.7067 42.4404 30.7621 44.8716

5

A-4 SKYHAWK 5.1226 15.9883 11.9615 14.1234 27.4048 22.3531 28.3778

6

A-10 A THUNDER BOL TII

5.6575 17.9239 19.3685 21.0637 41.5321 30.3261 42.2122

7

A-6 INTRUDER 5.0326 16.9623 15.3296 17.1987 36.6321 26.7596 33.8529

8

A-7 CORSAIRII 5.7388 15.3942 16.9992 19.1147 37.6194 27.1467 38.3100

9

ALPHA JET

4.9672 15.3168 12.4693 14.6741 28.4333 22.5569 29.9875

10

AMX

7.2238 16.5452 18.3967 20.5945 40.2130 29.0375 45.2979

SLIDE 22

TEST IMAGE FEATURES

Aircraft Image Ф1 Ф2 Ф3 Ф4 Ф5 Ф6 Ф7 Sample 6.1939 16.1073 21.7588 23.2762 45.8489 33.3117 46.4124

SLIDE 23

POSTERIOR PROBABILITIES 10 AIRCRAFTS

p(Ф1)/Ci p(Ф2)/Ci p(Ф3)/Ci p(Ф4)/Ci p(Ф5)/Ci p(Ф6)/Ci p(Ф7)/Ci Posterior probabilitie s

AIRCRAFT

1

Apache

0.53010155 0.239142429 0.128713538 0.130153337 0.066254708 0.113322114 0.055174814 8.79767E-07

2

A5

0.540385913 0.025736543 0.04331936 0.044934173 0.022337103 0.027296774 0.037127917 6.12847E-10

3

DELTA WING 0.580429962 0.237774382 0.150376929 0.151342242 0.076931454 0.121520153 0.062799134 1.84401E-06

4

Mig23

0.521569992 0.198120308 0.134533841 0.134759936 0.068735317 0.104895961 0.063198026 8.53643E-07

5

A-4 SKYHAWK

0.020378607 0.215134605 0.008885482 0.008545815 0.004052047 0.015835887 0.003047201 6.50934E-14

6

A-10 A THUNDER BOL TII

0.123731866 0.187618545 0.13809907 0.147746366 0.073731031 0.111805648 0.062535275 2.44177E-07

7

A-6 INTRUDER

0.014081437 0.237710062 0.08224835 0.071969767 0.063053236 0.096938093 0.018266329 2.2122E-09

8

A-7 CORSAIRII

0.153442028 0.171167628 0.135409074 0.135027707 0.069972999 0.104256364 0.044140037 1.54633E-07

9

ALPHA JET

0.010636652 0.164596904 0.013795254 0.013853378 0.006447256 0.017920509 0.00559365 2.16238E-13

10

AMX

0.506820616 0.237494915 0.151036155 0.152098166 0.077253421 0.121735458 0.062231457 1.6183E-06

SLIDE 24

K-MEANS CLUSTERING

AIRCRAFT 5 CLUSTERS 10 CLUSTERS

1 123 RF 1 1 2 CAW4F62P 5 9 3 CA8Q39AI 1 1 4 CAXF7IYK 3 6 5 CAORC6ZQ 3 6 6 B2 4 5 7 DELTA WING 1 1 8 F15 5 9 9 F35A 1 4 10 SAMPLE AIRCRAFT 1 1

SLIDE 25

CLUSTER ANALYSIS

CLUSTER NAME CENTROIDS

DATA No.

C1 7.0183 18.9995 21.5296 23.5403 46.1832 33.3138 49.5230 6 C2 7.0351 16.7412 18.6322 20.7872 40.6296 29.3285 44.8721 6 C3 7.2015 17.1527 19.6870 21.9202 42.8392 30.6037 48.7196 7 C4 6.6613 15.7419 16.9512 19.1145 37.5655 27.2579 39.7862 7 C5 5.4466 14.1287 12.2083 14.4152 27.9367 21.8117 29.7311 6

SLIDE 26

CONCLUSION

Invariant

moments feature extraction combined with Bayesian classifiers has been successful in classifier design: given a training set

f patterns of known class, and a

test pattern, a classifier has been designed that is optimal for the expected operating conditions.

SLIDE 27

REFERENCES

[1] Felix O. Owalla,Elijah Mwangi,“A Robust

Image Watermarking Scheme Invariant to Rotation,Scaling and Translation Attacks”, 16th IEEE Mediterranean Electrotechnical Conference, March , 2012.

[2] Dickson G. Wambaa,Elijah Mwangi, “Aircrafts

identification using moments invariants feature extraction and Bayesian Decision Theory Classification”, IEK Conference May, 2012.

[3] R. O. Duda,P. E. Hart and D. G. Stork, John

Wiley & Sons, Pattern Classification (2nd ed) 2000

SLIDE 28

REFERENCES

[4] William K. Pratt. Digital image processing 4th

edition John Wiley,US,2007

[5] M.-K. Hu, “Visual pattern recognition by moment

invariants,” IRE Trans. Information Theory, vol. IT-8,

pp. 179- 187, 1962.
[6] Richard O. Duda,Peter E. Hart and David

G.Stork.Pattern Classification 2nd edition John Wiley and Sons,US,2007

[7] Rafael C. Gonzalez,Richard E. Woods and Steven L.

Eddins . Digital image processing using matlab 2nd edition Pearson/Prentice Hall,US,2004

SLIDE 29

REFERENCES

[8] Dickson G. Wambaa,Elijah Mwangi,

“Aircrafts identification using moments invariants feature extraction and Bayesian Decision Theory Classification”, SAICSIT 2012 Masters and Doctoral Symposium 1 October 2012 Irene Country Lodge, Centurion South Africa.