Image Registration Using Mutual Information - ScholarlyCommons

4 downloads 0 Views 1MB Size Report
Almost all imaging systems require some form of registration. A few examples are aligning medical images for diagnosis, matching stereo images to recoverĀ ...
University of Pennsylvania

ScholarlyCommons Technical Reports (CIS)

Department of Computer & Information Science

January 2000

Image Registration Using Mutual Information Geoffrey Egnal University of Pennsylvania

Kostas Daniilidis University of Pennsylvania, [email protected]

Follow this and additional works at: http://repository.upenn.edu/cis_reports Recommended Citation Geoffrey Egnal and Kostas Daniilidis, "Image Registration Using Mutual Information", . January 2000.

University of Pennsylvania Department of Computer and Information Science Technical Report No. MS-CIS-00-05. This paper is posted at ScholarlyCommons. http://repository.upenn.edu/cis_reports/117 For more information, please contact [email protected].

Image Registration Using Mutual Information Abstract

Almost all imaging systems require some form of registration. A few examples are aligning medical images for diagnosis, matching stereo images to recover shape, and comparing facial images in a database to recognize people. Given the difficulty of registering images taken at different times, using different sensors, from different positions, registration algorithms come in different shapes and sizes. Recently, a new type of solution to the registration problem has emerged, based on information theory. In particular, the mutual information similarity metric has been used to register multi-modal medical images. Mutual information compares the statistical dependence between the two images. Unlike many other registration techniques, mutual information makes few a priori assumptions about the surface properties of the object or the imaging process, making it adaptible to changes in lighting and changes between sensors. The method can be applied to larger dimensional registration and many other imaging situations. In this report, we compare two approaches taken towards the implementation of rigid 2D mutual information image registration. We look further at algorithm speedup and noise reduction efforts. A full background is provided. Comments

University of Pennsylvania Department of Computer and Information Science Technical Report No. MSCIS-00-05.

This technical report is available at ScholarlyCommons: http://repository.upenn.edu/cis_reports/117

-200

-100

0.04

0.04

0.03

0.03

0.02

0.02

0.01

0.01

100

200

-200

-100

100

200

20

20

6

18

4

16 2

10 14

0

1 -10

2

3

4

5

6

12 -2

10

-4

8

-20 -20

-10

0

10

20

30

-20

-10

0

10

20

30

4

20

20

2

10

10

0

20

40

60

80

100

20

120

40

60

80

100

120 -2

-10

-10

-20

-20

-4

40

-20

-10

0

10

20

30

-20

-10

0

10

20

30

-1

20 30 -2

20

10

-3

10 20

40

60

80

100

120 20

-10

40

60

80

100

120

-4

-10 -5

-20 -20 -30

2.5

600

1 500 2

0.5

400 1.5

300

20

40

60

80

100

120 200 1

-0.5

100

20

-1

40

60

80

100

600 500

1000

2000

3000

4000

0

10

20

30

10

20

30

0.16

400

0.14

300

0.12

5000

0.1

200 -0.5

-10

0.18

1

0.5

-20

120

0.08 100

0.06 -1

1000

2000

3000

4000

5000

-20

-10

0

350

80

300 60

250 200

40

150 100

20

50 -30.

-20.

-10.

0.

10.

20.

30.

-20.

-10.

0.

10.

20.

30.

0.4

0.3

0.2

0.1

-4

-2

2

4

v(x 1 )

v(x 2 ) w1

w2

v(T(x)) v(x 3 )

w3

w4

v(x 4 )

6

0 4

-1 2

-2 -3 -2

-1

0

1

2

3

-4 -2

-2

-1

0

1

2

3