SlideShare a Scribd company logo
Seeing is Not
Believing: Camouflage
Attacks on Image
Scaling Algorithms
Junyaup Kim
20-03-11
TL DR
• This paper suggests image generation algorithm to form as convex optimization
to attack image scaling function. The objective of attack is
1. Make image 𝐴 to 𝐵 when the 𝐴 is resized.
2. The distortion should be small enough that attack image should be almost looks
like 𝐴
• By this, we can assure that this attack is model-free. You can attack any model
that is using certain frame work and certain scaling function. And scaling
function is used in any code line
• They suggests effective querying process to reveal Cloud Vision API provider’s
scaling size
Prerequisite
What is scaling?
• Scaling function is resizing function that make input
image to match specific shape.
• Deep learning model is basically matrix calculation.
So we should make our input static (solid shape)
Prerequisite
Inconsistency in DL model input shape and camera size
Basic Camera resolution chart Deep learning model input shape
Image scaling function is essential to every deep learning model.
Prerequisite
Interpolation and sampling
Bilinear interpolation
Interpolation: A type of estimation, a method of
constructing new data points within the range of a
discrete set of known data points. [Inter + pole]
Linear InterpolationGiven set Spline Interpolation
Prerequisite
Interpolation and sampling
Sampling: sampling is the reduction of a
continuous-time signal to a discrete-time signal
Bit depth : Quantization of input signal
Sampling rate : Quantization of time segment
Prerequisite
The Nyquist theorem specifies that a sinusoidal function in time or
distance can be regenerated with no loss of information as long as it is
sampled at a frequency greater than or equal to twice per cycle.
Alias and Nyquist Theorem
Nyquist Theorem
Prerequisite
Nyquist Theorem
Let’s suppose that pixel values are the discrete signal.
When we scale down the input image, we have not sufficient information of original image
value. To prevent aliasing artifacts, we must use filter to erase the aliasing artifacts.
Without optical low-pass filter With optical low-pass filterWithout optical low-pass filter
Prerequisite conclusion
1. Pixels are discrete signal.
2. We need filter with coefficient to scale down the image.
3. Scaling the image can be considered at data under-sampling
4. Due to physical limitations, scaling is used almost every deep
learning model.
Main subject
1. Background
• A lot of DL Framework provide
their own image resize
method.
• Order of interpolation is
Horizonal to vertical.(element
wise to channel-wise)
Main subject
1. Background
• Even though you are not using
resize function, somewhere in
the framework might inferring
the resize function.
Main subject
2. Objective
• The objective of this goal is map
the perturbations on Source
image that after scaling function
𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥) , the attack image
turns into target image.
𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥)
Source Image
Attack Image Target Image
𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥)
Source Image
Attack Image Target image
Main subject
3. Taxonomy
• Source image (𝑆 𝑚∗𝑛): the image that an
attacker wants the attack image to look like
• Attack image (𝐴 𝑚∗𝑛): the crafted image
eventually created and fed to the scaling
function
• Output image (𝐷 𝑚′∗𝑛′): the output image of
the scaling function
• Target image (𝑇 𝑚′∗𝑛′): the image that the
attacker wants the outImg to look like
• Scale function (ScaleFunc): The scaling function
of image.
𝑆 𝑚∗𝑛 + ∆1 = 𝐴 𝑚∗𝑛
∆1 = 𝐴 𝑚∗𝑛- 𝑆 𝑚∗𝑛
∆2 = 𝐷 𝑚′∗𝑛′ - 𝑇 𝑚′∗𝑛′
Main subject
4. Attack method
• Strong attack form: we KNOW the source image
that wants to make it to attack image
• Weak attack: we DON’T know the source image
that wants to make it to attack image.
Unknown +
Example output image
=
Main subject
4.1 Strong attack form
• Strong attack form: we KNOW the source image
that wants to make it to attack image
• Weak attack: we DON’T know the source image
that wants to make it to attack image.
Objective function: min(| 𝐴 𝑚∗𝑛− 𝑆 𝑚∗𝑛 |2
)
Constraints: ||𝑇 𝑚′∗𝑛′ - 𝐷 𝑚′∗𝑛′||∞ ≤ 𝜀 ∗ 𝐼𝑁 𝑚𝑎𝑥
Main subject
4.1 Coefficient analysis
• As we said before, we need filter matrix to
resize not to alias the image.
• And because of overlapping in filter
matrix(like CNN) we need to calculate
separately to make perturbation
Main subject
4.2 Coefficient analysis
• Horizontal scaling (𝑚 ∗ 𝑛 → 𝑚 ∗ 𝑛′)
• Vertical scaling(𝑚 ∗ 𝑛′ → 𝑚′ ∗ 𝑛′)
Coefficient matrix example
Vertical Horizontal
Craft sequence
Main subject
4.3 Strong attack form
• Constraints is a upper boundary of pixel
value(Constant function). So this constraints is
Linear.
• By that, we can calculate this as a convex form
Objective function: min(| 𝐴 𝑚∗𝑛− 𝑆 𝑚∗𝑛 |2
)
Constraints: ||𝑇 𝑚′∗𝑛′ - 𝐷 𝑚′∗𝑛′||∞ ≤ 𝜀 ∗ 𝐼𝑁 𝑚𝑎𝑥
WLOG
Main subject
4.4 Strong attack form algorithm analysis
• Decomposition into sub matrix problem.
Main subject
4.5 Cloud inference attack(black box)
• We have to know the exact size of
cloud DL model input size.
• inferring model image serach space is
𝑂 𝑁4
= (𝑃𝑎𝑘𝑐𝑎𝑔𝑒 ∗
𝑆𝑐𝑎𝑙𝑖𝑛𝑔 𝑚𝑒𝑡ℎ𝑜𝑑 ∗ ℎ𝑒𝑖𝑔ℎ𝑡 ∗ 𝑤𝑖𝑑𝑡ℎ)
 setting range[201,300] in H, W
 Infer different class by k times at the same
time (k=4)
Main subject
5.1 Result
• Attack target: Azure, Baidu, Aliyun, Tencent
• Testing Dataset: 935 (Crafted)
 Class except Sheep or sheep-like animal
 Set as 800*600 image
 𝜀 = 0.01
 Target = Sheep
• Baidu , Aliyun ,Tencent got 100% success ratio
where as Azuzre is more complex
• CDF(cumulative distribution function) shows
that Tag and description is successfully attacked
by this algorithm.
Main subject
Main subject
5.2 Possible attack scenario
• Data poisoning on database.
• Detection evasion and Cloaking on CNN
based deep learning models.
• Fraud by Leveraging Inconsistencies
between Displays. (ex mobile)
Main subject
5.3 Detection of attack
• Color-histogram-based Detection
• Color-scattering-based Detection
Conclusion
6. Pros
• This attack is model-free attack. It means
that we can use this attack in any situation
(not only limited in Deep learning)
• This attack is more light-weight than
adversarial attack by deep learning.
• Attack success ratio & confidence is high.
6. Cons
• If the model do not use the resize method(such
as yolo based object detection). It cannot be
successful.
• Only can be applied on smaller attack image
• The perturbations are easily recognizable by
human. The key of this kind of attack is should
be out of human-eye. You can easily recognize
that this image is somewhat wrong.

More Related Content

What's hot

Image Processing: Spatial filters
Image Processing: Spatial filtersImage Processing: Spatial filters
Image Processing: Spatial filters
A B Shinde
 
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
JaeJun Yoo
 
Digital image processing Tool presentation
Digital image processing Tool presentationDigital image processing Tool presentation
Digital image processing Tool presentation
dikshabehl5392
 
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Jedha Bootcamp
 
An overview of gradient descent optimization algorithms
An overview of gradient descent optimization algorithms An overview of gradient descent optimization algorithms
An overview of gradient descent optimization algorithms
Hakky St
 
Image Enhancement
Image Enhancement Image Enhancement
Image Enhancement
Deven Sahu
 
Contrast enhancement in digital images
Contrast enhancement in digital imagesContrast enhancement in digital images
Contrast enhancement in digital images
Sakher BELOUADAH
 
Image Enhancement in Spatial Domain
Image Enhancement in Spatial DomainImage Enhancement in Spatial Domain
Image Enhancement in Spatial Domain
DEEPASHRI HK
 
Image enhancement
Image enhancementImage enhancement
Image enhancement
Kuppusamy P
 
Spatial filtering
Spatial filteringSpatial filtering
Spatial filtering
DeepikaT13
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun Yoo
JaeJun Yoo
 
JonathanWestlake_ComputerVision_Project1
JonathanWestlake_ComputerVision_Project1JonathanWestlake_ComputerVision_Project1
JonathanWestlake_ComputerVision_Project1Jonathan Westlake
 
Image enhancement
Image enhancementImage enhancement
Image enhancement
Dr INBAMALAR T M
 
Simultaneous Smoothing and Sharpening of Color Images
Simultaneous Smoothing and Sharpening of Color ImagesSimultaneous Smoothing and Sharpening of Color Images
Simultaneous Smoothing and Sharpening of Color Images
Cristina Pérez Benito
 
Spatial enhancement
Spatial enhancement Spatial enhancement
Spatial enhancement abinarkt
 
Smoothing Filters in Spatial Domain
Smoothing Filters in Spatial DomainSmoothing Filters in Spatial Domain
Smoothing Filters in Spatial Domain
Madhu Bala
 
Remote Sensing Lec 10
Remote Sensing Lec 10Remote Sensing Lec 10
Remote Sensing Lec 10polylsgiedx
 
image enhancement
 image enhancement image enhancement
image enhancement
Rajendra Prasad
 
Enhancement in Digital Image Processing
Enhancement in Digital Image ProcessingEnhancement in Digital Image Processing
Enhancement in Digital Image Processing
International Institute of Information Technology (I²IT)
 
Improved Alpha-Tested Magnification for Vector Textures and Special Effects
Improved Alpha-Tested Magnification for Vector Textures and Special EffectsImproved Alpha-Tested Magnification for Vector Textures and Special Effects
Improved Alpha-Tested Magnification for Vector Textures and Special Effects
ナム-Nam Nguyễn
 

What's hot (20)

Image Processing: Spatial filters
Image Processing: Spatial filtersImage Processing: Spatial filters
Image Processing: Spatial filters
 
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
 
Digital image processing Tool presentation
Digital image processing Tool presentationDigital image processing Tool presentation
Digital image processing Tool presentation
 
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
Faire de la reconnaissance d'images avec le Deep Learning - Cristina & Pierre...
 
An overview of gradient descent optimization algorithms
An overview of gradient descent optimization algorithms An overview of gradient descent optimization algorithms
An overview of gradient descent optimization algorithms
 
Image Enhancement
Image Enhancement Image Enhancement
Image Enhancement
 
Contrast enhancement in digital images
Contrast enhancement in digital imagesContrast enhancement in digital images
Contrast enhancement in digital images
 
Image Enhancement in Spatial Domain
Image Enhancement in Spatial DomainImage Enhancement in Spatial Domain
Image Enhancement in Spatial Domain
 
Image enhancement
Image enhancementImage enhancement
Image enhancement
 
Spatial filtering
Spatial filteringSpatial filtering
Spatial filtering
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun Yoo
 
JonathanWestlake_ComputerVision_Project1
JonathanWestlake_ComputerVision_Project1JonathanWestlake_ComputerVision_Project1
JonathanWestlake_ComputerVision_Project1
 
Image enhancement
Image enhancementImage enhancement
Image enhancement
 
Simultaneous Smoothing and Sharpening of Color Images
Simultaneous Smoothing and Sharpening of Color ImagesSimultaneous Smoothing and Sharpening of Color Images
Simultaneous Smoothing and Sharpening of Color Images
 
Spatial enhancement
Spatial enhancement Spatial enhancement
Spatial enhancement
 
Smoothing Filters in Spatial Domain
Smoothing Filters in Spatial DomainSmoothing Filters in Spatial Domain
Smoothing Filters in Spatial Domain
 
Remote Sensing Lec 10
Remote Sensing Lec 10Remote Sensing Lec 10
Remote Sensing Lec 10
 
image enhancement
 image enhancement image enhancement
image enhancement
 
Enhancement in Digital Image Processing
Enhancement in Digital Image ProcessingEnhancement in Digital Image Processing
Enhancement in Digital Image Processing
 
Improved Alpha-Tested Magnification for Vector Textures and Special Effects
Improved Alpha-Tested Magnification for Vector Textures and Special EffectsImproved Alpha-Tested Magnification for Vector Textures and Special Effects
Improved Alpha-Tested Magnification for Vector Textures and Special Effects
 

Similar to Seeing is Not Believing: Camouflage Attacks on Image Scaling Algorithms Review

cnn ppt.pptx
cnn ppt.pptxcnn ppt.pptx
cnn ppt.pptx
rohithprabhas1
 
Computer vision-nit-silchar-hackathon
Computer vision-nit-silchar-hackathonComputer vision-nit-silchar-hackathon
Computer vision-nit-silchar-hackathon
Aditya Bhattacharya
 
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
IRJET Journal
 
cvpresentation-190812154654 (1).pptx
cvpresentation-190812154654 (1).pptxcvpresentation-190812154654 (1).pptx
cvpresentation-190812154654 (1).pptx
PyariMohanJena
 
ppt 20BET1024.pptx
ppt 20BET1024.pptxppt 20BET1024.pptx
ppt 20BET1024.pptx
ManeetBali
 
Computer Vision.pptx
Computer Vision.pptxComputer Vision.pptx
Computer Vision.pptx
GDSCIIITDHARWAD
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
PyData
 
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
IRJET Journal
 
Deceiving Autonomous Cars with Toxic Signs
Deceiving Autonomous Cars with Toxic SignsDeceiving Autonomous Cars with Toxic Signs
Deceiving Autonomous Cars with Toxic Signs
LeonardoSalvucci1
 
Computer Vision - Real Time Face Recognition using Open CV and Python
Computer Vision - Real Time Face Recognition using Open CV and PythonComputer Vision - Real Time Face Recognition using Open CV and Python
Computer Vision - Real Time Face Recognition using Open CV and Python
Akash Satamkar
 
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
JacobSilbiger1
 
Image De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural NetworkImage De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural Network
aciijournal
 
Image De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural NetworkImage De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural Network
aciijournal
 
IMAGE DE-NOISING USING DEEP NEURAL NETWORK
IMAGE DE-NOISING USING DEEP NEURAL NETWORKIMAGE DE-NOISING USING DEEP NEURAL NETWORK
IMAGE DE-NOISING USING DEEP NEURAL NETWORK
aciijournal
 
IMAGE PROCESSING
IMAGE PROCESSINGIMAGE PROCESSING
IMAGE PROCESSING
ABHISHEK MAURYA
 
Visual CryptoGraphy
Visual CryptoGraphyVisual CryptoGraphy
Visual CryptoGraphy
pallavikhandekar212
 
BMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorialBMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorial
potaters
 
OpenCV.pdf
OpenCV.pdfOpenCV.pdf
OpenCV.pdf
sagarawasthi5
 

Similar to Seeing is Not Believing: Camouflage Attacks on Image Scaling Algorithms Review (20)

cnn ppt.pptx
cnn ppt.pptxcnn ppt.pptx
cnn ppt.pptx
 
Computer vision-nit-silchar-hackathon
Computer vision-nit-silchar-hackathonComputer vision-nit-silchar-hackathon
Computer vision-nit-silchar-hackathon
 
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
 
cvpresentation-190812154654 (1).pptx
cvpresentation-190812154654 (1).pptxcvpresentation-190812154654 (1).pptx
cvpresentation-190812154654 (1).pptx
 
ppt 20BET1024.pptx
ppt 20BET1024.pptxppt 20BET1024.pptx
ppt 20BET1024.pptx
 
Computer Vision.pptx
Computer Vision.pptxComputer Vision.pptx
Computer Vision.pptx
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
 
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
An Approach for Image Deblurring: Based on Sparse Representation and Regulari...
 
Deceiving Autonomous Cars with Toxic Signs
Deceiving Autonomous Cars with Toxic SignsDeceiving Autonomous Cars with Toxic Signs
Deceiving Autonomous Cars with Toxic Signs
 
Computer Vision - Real Time Face Recognition using Open CV and Python
Computer Vision - Real Time Face Recognition using Open CV and PythonComputer Vision - Real Time Face Recognition using Open CV and Python
Computer Vision - Real Time Face Recognition using Open CV and Python
 
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
YU CS Summer 2021 Project | TensorFlow Street Image Classification and Object...
 
Image De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural NetworkImage De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural Network
 
Image De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural NetworkImage De-Noising Using Deep Neural Network
Image De-Noising Using Deep Neural Network
 
IMAGE DE-NOISING USING DEEP NEURAL NETWORK
IMAGE DE-NOISING USING DEEP NEURAL NETWORKIMAGE DE-NOISING USING DEEP NEURAL NETWORK
IMAGE DE-NOISING USING DEEP NEURAL NETWORK
 
IMAGE PROCESSING
IMAGE PROCESSINGIMAGE PROCESSING
IMAGE PROCESSING
 
Visual CryptoGraphy
Visual CryptoGraphyVisual CryptoGraphy
Visual CryptoGraphy
 
BMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorialBMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorial
 
OpenCV.pdf
OpenCV.pdfOpenCV.pdf
OpenCV.pdf
 
N046047780
N046047780N046047780
N046047780
 
DL.pdf
DL.pdfDL.pdf
DL.pdf
 

Recently uploaded

Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
Adtran
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
sonjaschweigert1
 
By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024
Pierluigi Pugliese
 
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
Jen Stirrup
 
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
UiPathCommunity
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
Kari Kakkonen
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
DanBrown980551
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
KatiaHIMEUR1
 
Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
RinaMondal9
 
Elevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object CalisthenicsElevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object Calisthenics
Dorra BARTAGUIZ
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
Safe Software
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Albert Hoitingh
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
SOFTTECHHUB
 
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdfFIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
Laura Byrne
 
GraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge GraphGraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge Graph
Guy Korland
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
James Anderson
 

Recently uploaded (20)

Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
 
By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024
 
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
 
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
 
Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
 
Elevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object CalisthenicsElevating Tactical DDD Patterns Through Object Calisthenics
Elevating Tactical DDD Patterns Through Object Calisthenics
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
 
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdfFIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
 
GraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge GraphGraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge Graph
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
 

Seeing is Not Believing: Camouflage Attacks on Image Scaling Algorithms Review

  • 1. Seeing is Not Believing: Camouflage Attacks on Image Scaling Algorithms Junyaup Kim 20-03-11
  • 2. TL DR • This paper suggests image generation algorithm to form as convex optimization to attack image scaling function. The objective of attack is 1. Make image 𝐴 to 𝐵 when the 𝐴 is resized. 2. The distortion should be small enough that attack image should be almost looks like 𝐴 • By this, we can assure that this attack is model-free. You can attack any model that is using certain frame work and certain scaling function. And scaling function is used in any code line • They suggests effective querying process to reveal Cloud Vision API provider’s scaling size
  • 3. Prerequisite What is scaling? • Scaling function is resizing function that make input image to match specific shape. • Deep learning model is basically matrix calculation. So we should make our input static (solid shape)
  • 4. Prerequisite Inconsistency in DL model input shape and camera size Basic Camera resolution chart Deep learning model input shape Image scaling function is essential to every deep learning model.
  • 5. Prerequisite Interpolation and sampling Bilinear interpolation Interpolation: A type of estimation, a method of constructing new data points within the range of a discrete set of known data points. [Inter + pole] Linear InterpolationGiven set Spline Interpolation
  • 6. Prerequisite Interpolation and sampling Sampling: sampling is the reduction of a continuous-time signal to a discrete-time signal Bit depth : Quantization of input signal Sampling rate : Quantization of time segment
  • 7. Prerequisite The Nyquist theorem specifies that a sinusoidal function in time or distance can be regenerated with no loss of information as long as it is sampled at a frequency greater than or equal to twice per cycle. Alias and Nyquist Theorem Nyquist Theorem
  • 8. Prerequisite Nyquist Theorem Let’s suppose that pixel values are the discrete signal. When we scale down the input image, we have not sufficient information of original image value. To prevent aliasing artifacts, we must use filter to erase the aliasing artifacts. Without optical low-pass filter With optical low-pass filterWithout optical low-pass filter
  • 9. Prerequisite conclusion 1. Pixels are discrete signal. 2. We need filter with coefficient to scale down the image. 3. Scaling the image can be considered at data under-sampling 4. Due to physical limitations, scaling is used almost every deep learning model.
  • 10. Main subject 1. Background • A lot of DL Framework provide their own image resize method. • Order of interpolation is Horizonal to vertical.(element wise to channel-wise)
  • 11. Main subject 1. Background • Even though you are not using resize function, somewhere in the framework might inferring the resize function.
  • 12. Main subject 2. Objective • The objective of this goal is map the perturbations on Source image that after scaling function 𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥) , the attack image turns into target image. 𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥) Source Image Attack Image Target Image 𝑆𝑐𝑎𝑙𝑒𝐹𝑢𝑛𝑐(𝑥) Source Image Attack Image Target image
  • 13. Main subject 3. Taxonomy • Source image (𝑆 𝑚∗𝑛): the image that an attacker wants the attack image to look like • Attack image (𝐴 𝑚∗𝑛): the crafted image eventually created and fed to the scaling function • Output image (𝐷 𝑚′∗𝑛′): the output image of the scaling function • Target image (𝑇 𝑚′∗𝑛′): the image that the attacker wants the outImg to look like • Scale function (ScaleFunc): The scaling function of image. 𝑆 𝑚∗𝑛 + ∆1 = 𝐴 𝑚∗𝑛 ∆1 = 𝐴 𝑚∗𝑛- 𝑆 𝑚∗𝑛 ∆2 = 𝐷 𝑚′∗𝑛′ - 𝑇 𝑚′∗𝑛′
  • 14. Main subject 4. Attack method • Strong attack form: we KNOW the source image that wants to make it to attack image • Weak attack: we DON’T know the source image that wants to make it to attack image. Unknown + Example output image =
  • 15. Main subject 4.1 Strong attack form • Strong attack form: we KNOW the source image that wants to make it to attack image • Weak attack: we DON’T know the source image that wants to make it to attack image. Objective function: min(| 𝐴 𝑚∗𝑛− 𝑆 𝑚∗𝑛 |2 ) Constraints: ||𝑇 𝑚′∗𝑛′ - 𝐷 𝑚′∗𝑛′||∞ ≤ 𝜀 ∗ 𝐼𝑁 𝑚𝑎𝑥
  • 16. Main subject 4.1 Coefficient analysis • As we said before, we need filter matrix to resize not to alias the image. • And because of overlapping in filter matrix(like CNN) we need to calculate separately to make perturbation
  • 17. Main subject 4.2 Coefficient analysis • Horizontal scaling (𝑚 ∗ 𝑛 → 𝑚 ∗ 𝑛′) • Vertical scaling(𝑚 ∗ 𝑛′ → 𝑚′ ∗ 𝑛′) Coefficient matrix example Vertical Horizontal Craft sequence
  • 18. Main subject 4.3 Strong attack form • Constraints is a upper boundary of pixel value(Constant function). So this constraints is Linear. • By that, we can calculate this as a convex form Objective function: min(| 𝐴 𝑚∗𝑛− 𝑆 𝑚∗𝑛 |2 ) Constraints: ||𝑇 𝑚′∗𝑛′ - 𝐷 𝑚′∗𝑛′||∞ ≤ 𝜀 ∗ 𝐼𝑁 𝑚𝑎𝑥 WLOG
  • 19. Main subject 4.4 Strong attack form algorithm analysis • Decomposition into sub matrix problem.
  • 20. Main subject 4.5 Cloud inference attack(black box) • We have to know the exact size of cloud DL model input size. • inferring model image serach space is 𝑂 𝑁4 = (𝑃𝑎𝑘𝑐𝑎𝑔𝑒 ∗ 𝑆𝑐𝑎𝑙𝑖𝑛𝑔 𝑚𝑒𝑡ℎ𝑜𝑑 ∗ ℎ𝑒𝑖𝑔ℎ𝑡 ∗ 𝑤𝑖𝑑𝑡ℎ)  setting range[201,300] in H, W  Infer different class by k times at the same time (k=4)
  • 21. Main subject 5.1 Result • Attack target: Azure, Baidu, Aliyun, Tencent • Testing Dataset: 935 (Crafted)  Class except Sheep or sheep-like animal  Set as 800*600 image  𝜀 = 0.01  Target = Sheep • Baidu , Aliyun ,Tencent got 100% success ratio where as Azuzre is more complex • CDF(cumulative distribution function) shows that Tag and description is successfully attacked by this algorithm.
  • 23. Main subject 5.2 Possible attack scenario • Data poisoning on database. • Detection evasion and Cloaking on CNN based deep learning models. • Fraud by Leveraging Inconsistencies between Displays. (ex mobile)
  • 24. Main subject 5.3 Detection of attack • Color-histogram-based Detection • Color-scattering-based Detection
  • 25. Conclusion 6. Pros • This attack is model-free attack. It means that we can use this attack in any situation (not only limited in Deep learning) • This attack is more light-weight than adversarial attack by deep learning. • Attack success ratio & confidence is high. 6. Cons • If the model do not use the resize method(such as yolo based object detection). It cannot be successful. • Only can be applied on smaller attack image • The perturbations are easily recognizable by human. The key of this kind of attack is should be out of human-eye. You can easily recognize that this image is somewhat wrong.

Editor's Notes

  1. So image scaling is basically a interpolation between the adjacent pixel matrix. Gaussian process 도 있지만 we will skip that.
  2. So what is sampling? We know sampling in statistical way. It is almost same semantic meaning in Signal processing area. We cannot know continuous values so we split the each time line (called sampling rate) and represent the signal amplitude(진폭) Let’s think
  3. If we are digging this part too far, we will be buried at mathematical equations so let me just brifly go over this. If we decompose the signal into really small time, we can see this kind of sin graph. But if our sampling rate
  4. So what is sampling? We know sampling in statistical way. It is almost same semantic meaning in Signal processing area. We cannot know continuous values so we split the each time line (called sampling rate) and represent the signal amplitude(진폭)
  5. So what is sampling? We know sampling in statistical way. It is almost same semantic meaning in Signal processing area. We cannot know continuous values so we split the each time line (called sampling rate) and represent the signal amplitude(진폭)