27
Image recognition using analysis of the frequency domain features 1

Image recognition using analysis of the frequency domain features 1

Embed Size (px)

Citation preview

Page 1: Image recognition using analysis of the frequency domain features 1

Image recognition using analysis of the frequency domain features

1

Page 2: Image recognition using analysis of the frequency domain features 1

Image Recognition

Image recognition problem is a problem of recognition of some certain objects that are located in an image.

2

Page 3: Image recognition using analysis of the frequency domain features 1

Image Recognition

To solve any pattern recognition/classification problem, it is necessary to find a relevant set of those features that can exhaustively describe an object to be recognized.

We never will confuse recognizing where is a tiger and where is a rabbit, but how an automatic tool can decide who is who?

3

Page 4: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection

Can you propose a set of features using which we can definitely distinguish a tiger from a rabbit?

4

Page 5: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection

It is often difficult to find a proper set of those features that would be really exhaustive and would not be redundant (redundancy complicates both processes of learning and recognition).

Another problem is a formal representation of the selected features.

5

Page 6: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection PCA (Principal Component Analysis)

is a method, which is often used for obtaining the objective features.

PCA is based on the Karhunen-Loeve transformation of a signal (a transformation by the eigenvectors of the covariance matrix of the ensemble of signals), which is computationally very costly.

6

Page 7: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection

The idea behind PCA is to find a small amount of those eigenvectors (and spectral coefficients, respectively) that make a major contribution to the formation of a signal

The question: is it possible to find another approach to obtaining the objective features?

7

Page 8: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection Oppenheim, A.V.; Lim, J.S., The

importance of phase in signals, IEEE Proceedings, v. 69, No 5, 1981,pp.:  529- 541

In this paper, it was shown that phase in the Fourier spectrum of an image is much more informative than magnitude: phase contains the information about all shapes, edges, orientation of all objects, etc.

8

Page 9: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection

Thus the Fourier Phase Spectrum can be a very good source of the objective features that describe all objects located in images.

The Power Spectrum (magnitude) describes global image properties (blur, noise, cleanness, contrast, brightness, etc.).

9

Page 10: Image recognition using analysis of the frequency domain features 1

Phase and Magnitude

(a) (b)

Phase (a) & Magnitude (b) Phase (b) & Magnitude (a)

Phase contains the information about an object presented by a signal

10

Page 11: Image recognition using analysis of the frequency domain features 1

Phase and Magnitude

(a) (b)

Phase (a) & Magnitude (b) Phase (b) & Magnitude (a)

Magnitude contains the information about the signal’s properties

11

Page 12: Image recognition using analysis of the frequency domain features 1

Phase and Magnitude

Blur with a symmetric point-spread function practically does not affect the phase, while the magnitude may be distorted significantly.

This property may be use for recognition of blurred images using a phase spectrum as a feature space.

12

Page 13: Image recognition using analysis of the frequency domain features 1

Image Recognition: Features Selection

Since the Fourier Transform is computationally much simpler and more efficient than the Karhunen-Loeve transform (because of the existence of a number of Fast Fourier Transform algorithms), the use of the Fourier phases as the features for object recognition is very attractive.

13

Page 14: Image recognition using analysis of the frequency domain features 1

Image Recognition: Decision Rule and Classifier The next question is: is it possible to formulate

(and formalize!) the decision rule, using which we can classify or recognize our objects basing on the selected features?

Can you propose the rule using which we can definitely decide is it a tiger or a rabbit?

14

Page 15: Image recognition using analysis of the frequency domain features 1

Image Recognition: Decision Rule and Classifier

Once we know our decision rule, it is not difficult to develop a classifier, which will perform classification/recognition using the selected features and the decision rule.

However, if the decision rule can not be formulated and formalized, we should use a classifier, which can develop the rule from the learning process

15

Page 16: Image recognition using analysis of the frequency domain features 1

Image Recognition: Decision Rule and Classifier

In the most of recognition/classification problems, the formalization of the decision rule is very complicated or impossible at all.

A neural network is a tool, which can accumulate knowledge from the learning process.

After the learning process, a neural network is able to approximate a function, which is supposed to be our decision rule

16

Page 17: Image recognition using analysis of the frequency domain features 1

Why neural network?

1( ,..., )nf x x

0 1( , ,..., )nw w w

- unknown multi-factor decision rule

Learning process using a representative learning set

- a set of weighting vectors is the result of the learning process

1

0 1 1

ˆ ( ,..., )

( ... )n

n n

f x x

P w w x w x

- a partially defined function, which is an approximation of the decision rule function 17

Page 18: Image recognition using analysis of the frequency domain features 1

Image Recognition: Approach

We will use the low frequency Fourier phases as the features. They contain the most important information about those objects that we want to recognize

We will use a neural network as a classifier

18

Page 19: Image recognition using analysis of the frequency domain features 1

Features Selection

Features are selected from the low frequency part of theFourier phase spectrum

19

Page 20: Image recognition using analysis of the frequency domain features 1

Example: Classification of Gene Expression Patterns

20

Page 21: Image recognition using analysis of the frequency domain features 1

Gene expression patterns We have studied spatio-temporal

expression patterns of genes controlling segmentation in the embryo of fruit fly Drosophila.

A problem is to perform temporal classification of the gene expression patterns taken form the confocal electronic microscope (8 temporal classes are considered)

21

Page 22: Image recognition using analysis of the frequency domain features 1

Image of gene expression data

in Drosophila embryoobtained by confocalscanning microscopy

22

Page 23: Image recognition using analysis of the frequency domain features 1

A problem of the classification

Class 1 Class 2 Class 3 Class 4

Class 5 Class 6 Class 7 Class 8

Representatives of 8 temporal classes:

23

Page 24: Image recognition using analysis of the frequency domain features 1

Phases as the features

Class 1 Class 8

Phase Cl.1+Amplitude Cl.8

Phase Cl.8 +Amplitude Cl.1

24

Page 25: Image recognition using analysis of the frequency domain features 1

Learning process

From 28 up to 32 images from each class a priori correctly classified as “representative” from biological view were used for the learning

From 60 inputs up to 144 inputs (from 5 to 8 low frequency coefficients) have been used as the features

25

Page 26: Image recognition using analysis of the frequency domain features 1

The Classification Results

Final classification results

Classes 1 2 3 4 5 6 7 8

Number of frequencies/

inputs

1-8/144

1-8/144

1-5/144

1-5/ 144

1-8/ 144

1-8/ 144

1-5/144

1-5/144

Number of Images

48 43 68 55 78 89 66 46

Recognized46

(95%)33

(76%)53

(77%)47

(85%)56

(71%)65

(73%)48

(72%)32

(69%)

Unrecognized 0 1 1 1 0 0 0 0

Misclassified 2 9 14 7 22 24 18 14

Recognized using Discrimination Analysis (previously used approach)

- - - 84% 49% 59% 61% 68%

26

Page 27: Image recognition using analysis of the frequency domain features 1

Problems that we will consider

Textures classification (automatic classification of different Gaussian and uniform textures)

Blurred images recognition

27