You are on page 1of 30

R&D Centre Europe Visual & Sensing Division

www.uk.mitsubishielectric-rce.eu

WORKSHOP ON MOBILE VISUAL SEARCH Fingerprinting the Web with Image Signatures

Miroslaw Bober Paul Brasnett Mitsubishi Electric R&D Centre Europe December 2009

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

Fingerprinting the Web Why?

www.uk.mitsubishielectric-rce.eu

Who are her friends? Who holds the copyright? Who is she? What web-pages does she appear at? show me all! Is that a part of a larger image? Does she know Leonardo?

In which museum can I see this painting? Do I already have a copy of that image on my laptop?

I want to see other images of her!

What is her telephone number? Somebody is taking this painting across the border is that OK?

I want a full resolution picture of her!

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

Overview
Need for standardised Visual Signature Tools MPEG7 Image and Video signature tools Image Signature Requirements Evaluation & Development Process Design of Image Signature Performance Use Scenarios Conclusions

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

Need for standardised Visual Signature Tools

www.uk.mitsubishielectric-rce.eu

There is always a need to identify items


e.g. ISBN, barcodes, car license plates etc

There is a need to identify multimedia content Identifier standards exist e.g. International Standard Audio-visual Number However:
99.999% digital content exists without any identifier even if image content is registered, no automated search/identification is possible

Content Signatures can either


1. provide a link between content and its identifier number or, 2. form an identifier itself.

Image and Video signatures are based on analysis of content


2009 R&D Centre Europe B.V. Visual & Sensing Division Export Control : NLR VIS09-PPT168

MPEG7 Image and Video signature tools

www.uk.mitsubishielectric-rce.eu

Work started in January 2006, a part of ISO/IEC 15938-part 3 (MPEG-7 Visual) Tough requirements formulated July 2007 Call for proposals issued April 2009 Image Signature Standard completed Work on Video Signature close to completion

Region Shape

Colour Layout

Camera Motion
Boom up Track right Dolly forward Boom down Dolly backward Track left

Face Detection and Recognition

Dominant Colour

Edge Histogram

Homogeneous Texture

Colour Structure

Contour Shape

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

Image Signature Requirements

www.uk.mitsubishielectric-rce.eu

How many images on the Web? 100 billion?+ What does this mean for the requirements? False alarm rate 1 in 20,000,000 Matching speed 100 million+ per second on a typical PC (global)

General Requirements: Robustness Independence Fast Matching Fast Extraction Compactness Non-alteration of content Self-contained Coding Independence
2009 R&D Centre Europe B.V. Visual & Sensing Division

Image Signature Requirements Partial matching (cropping) Heavy compression Scaling & aspect ratio change

Export Control : NLR

VIS09-PPT168

Performance Evaluation
Two Datasets

www.uk.mitsubishielectric-rce.eu

Colour/grayscale photographs, logos & graphics of varying sizes Independence: 135,000 images, ~10 billion pairs Robustness: 10,000 images x 44 modifications

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

MPEG Evaluation & Development Process

www.uk.mitsubishielectric-rce.eu

Requirements and Evaluation Conditions

Call for Proposals Issued

Technologies Received

Standard Published

Improve Technology & Prepare Standard

Evaluations Performed & Technology Selected

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

Trace Transform

www.uk.mitsubishielectric-rce.eu

The trace transform is a generalization of the Radon transform The image is traced by straight lines along which a functional of the image is computed. Each line is characterised by two parameters; Trace transform is effectively the map of the values of the functional as a function of these parameters By choosing the functionals , one can construct features that are invariant to various image distortions, like affine transforms or illumination changes. A second functional (also known as diametrical functional) is applied to the columns of the trace transform producing 1D function called a circus function Problems of occlusion can be tackled with the help of local image kernels, a localized version of the Trace transform

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

How do we get rotational invariance?

www.uk.mitsubishielectric-rce.eu

3.85

x 10

x 10

3.8
2.5

3.75
2

3.7

3.65

3.6

3.55

FFT magnitude

1.5

3.5

0.5

3.45
0 0 10 20 30 40 50 60 70 80 90 100

3.4

100

200

300

400

500

600

x 10

0.5

-0.5

-1

-1.5

10

20

30

40

50

60

70

80

90

100

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

10

Example of rotational invariance

www.uk.mitsubishielectric-rce.eu

3.85 3.9 3.85 3.85 3.8 3.8 3.8 3.75 3.75 3.75 3.7 3.7 150 150 3.7 3.65 3.65 3.6 3.6 250 250 3.55 3.55 3.55 3.5 3.5 3.5 3.45 3.45 3.45

x 104 x 10

4 4

50 50

100 100

200 200

300 300

350 350

0 0

100 100

200 200

300 300

400 400

500 500

600 600

3.4 3.4 0 0

100 100

200 200

300 300

400 400

500 500

600 600

x 10

2.5

1.5

0.5

10

20

30

40

50

60

70

80

90

100

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

11

Multiresolution Trace Transform

www.uk.mitsubishielectric-rce.eu

We need to increase robustness Idea: create a multiresolution image representation by building a pyramid of trace transforms Fast can be done directly in trace domain Linear and angular dimension

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

12

Trace Cube

www.uk.mitsubishielectric-rce.eu

A three-dimensional (3D) representation of the image is generated in terms of the parameters d, and t This new representation, which is a 3D volume, is referred to as the trace-cube. The trace-cube carries significantly more information then the conventional trace transform, including information related to non-linear regions in the image, such as circles or ellipses.

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

13

Combining binary descriptors

www.uk.mitsubishielectric-rce.eu

Component binary strings are combined to create the global image signature 512 bits

+ FFT

512 bits

+ FFT

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

14

Global Image Signature

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

15

Local Image Signature

www.uk.mitsubishielectric-rce.eu

A scale-space representation of the image is constructed by low-pass filtering with Gaussian kernel. Features are detected by finding maxima in the location-scale space Two detectors are used to provide good detection of the location and scale of features modified scale corrected Laplacian of Gaussian (filter) scale-adapted Harris operator Image intensity gradients are calculated using the Scharr operator

The Laplacian of Gaussianis is applied over 12 levels in scale space The locations and scales with the highest value are chosen as potential features. Candidate features corresponding to a weak response from Harris operator are removed
2009 R&D Centre Europe B.V. Visual & Sensing Division Export Control : NLR VIS09-PPT168

16

Local Image Signature II


Between 32 and 80 feature points are selected Their location and quantised gradient direction are stored The global trace transform method is used to extract a binary feature signature from circular regions centered at feature points. Regions are extracted to sub-pixel level accuracy and have scaledependent radius, normalised to 32 pixels

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

17

MPEG-7 Image Signature Syntax

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

18

Local
levels Modified Items Brightness change Colour to monochrome conversion JPEG compression with varying Quality factors Colour reduction Gaussian Noise Histogram equalization Auto-levels Blur Simple Rotation Geometric transformation Scaling Flip AVERAGE
2009 R&D Centre Europe B.V. Visual & Sensing Division Export Control : NLR

www.uk.mitsubishielectric-rce.eu

Heavy 99.81

Medium 99.91 99.99

Light 100.00

Mean 99.91 99.99

99.69

99.95 99.34

100.00 99.86 99.91

99.88 99.60 99.51 95.41 99.95

99.04

99.58 95.41 99.95

99.98 100.00 100.00

99.99 100.00 100.00 100.00

99.98 100.00 100.00

99.98 100.00 100.00 100.00 99.48


VIS09-PPT168

19

Local + Global

www.uk.mitsubishielectric-rce.eu

levels Modified Items Brightness change Colour to monochrome JPEG compression with varying Quality factors Colour reduction Gaussian Noise Histogram equalization Auto-levels Blur

Heavy 99.97

Medium 99.98 99.99

Light 100.0 0

Mean 99.98 99.99

99.97

99.99 99.84

100.0 0 99.99 99.98

99.99 99.92 99.88 93.15 100.00

99.76

99.88 93.15 100.00

100.00

100.0

100.0

100.00

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

20

Local + Global

www.uk.mitsubishielectric-rce.eu

Rotation Simple Rotation Scaling Translation Geometric transformation Flip Aspect Ratio Change Crop Skew Perspective Combined (C,T,S) AVERAGE

99.99 100.00 100.00 75.64

100.0 100.0 100.0 91.85 100.00 100.0

100.00 100.00 100.00 99.42

100.00 100.00 100.00 88.97 100.00

100.00 99.48 100.00 99.85 99.85

100.00 82.11 98.27 98.36 99.26 97.77

54.95 94.86 96.15 98.33

91.89 99.96 99.08 99.60

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

21

Usage Scenarios - The missing Link?


Media Usage Monitoring
Tracking & recording multimedia usage Is somebody using my logo illegally?

www.uk.mitsubishielectric-rce.eu

Personal library management


Managing alternative versions of the same content Have I copied this MemoryStick to my PC?

Content linking
Inferring links between content based on analysis of media content

Metadata Association
Quickly and accurately obtain content metadata information

Copyright management
Royalty collection, brand management
2009 R&D Centre Europe B.V. Visual & Sensing Division Export Control : NLR VIS09-PPT168

22

Traditional Image Search keyword based

www.uk.mitsubishielectric-rce.eu

Wojciech Machowski

Index

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

23

Web Search

www.uk.mitsubishielectric-rce.eu

Approximately 1.5 million images from Flickr Several hundred random seed keywords used e.g.
oxide, rank, perplexity, mirrors, throne, walk, hens, seat etc.

Image Signature extracted and stored in database


2009 R&D Centre Europe B.V. Visual & Sensing Division Export Control : NLR VIS09-PPT168

24

Example Query

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

25

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

26

www.uk.mitsubishielectric-rce.eu

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

27

MPEG-7 Image Signature Tools


Global Signature
512 Bits per image False alarm rate of 0.05 parts per million 99.29% correct detection Up to 100 million images per second searching

www.uk.mitsubishielectric-rce.eu

Local Signature
Between 3,584 and 7,424 bits per image False alarm rate of 10ppm 98.57% correct detection on cropping, translation Over 100,000 images per second searching

ISO/IEC 15938-3:2002/Amd. 3 Image Signature Tools


Published: April 2009

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

28

Conclusions
MPEG-7 Visual Signature Tools
Standardised Multimedia Content Identification Robust, Low False Alarm, Fast Searching

www.uk.mitsubishielectric-rce.eu

Suitable for usage monitoring, copyright management, content linking, content metadata association MPEG-7 Image Signature Tools
Published April 2009

MPEG-7 Video Signature Tools


Expected To be Published July 2010

Maybe this is the missing link that you need for your killer application!

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

29

R&D Centre Europe Visual & Sensing Division

www.uk.mitsubishielectric-rce.eu

MERCE R&D

with an eye to the future!


thank you for your attention (m.bober@uk.merce.mee.com)

2009 R&D Centre Europe B.V. Visual & Sensing Division

Export Control : NLR

VIS09-PPT168

30

You might also like