SlideShare une entreprise Scribd logo
1  sur  15
Reza FARRAHI MOGHADDAM, Fereydoun FARRAHI MOGHADDAM and Mohamed CHERIET
Synchromedia Laboratory, ETS, Montreal (QC), Canada H3C 1K3
imriss@ieee.org, rfarrahi@synchromedia.ca,
ffarrahi@synchromedia.ca, mohamed.cheriet@etsmtl.ca
ICDAR 2013, Washington, DC, USA, August 25th-28th, 2013
Outline
 Why Ensemble of Experts (EoE) framework?
 EoE vs. Ensemble of Classifiers (EoC)
 The big picture
 Notations
 Endorsements and the Endorsement Graph
 The selection process
 Calculation of the EoE result and its variations
 Use cases
 Conclusions and future prospects
 Any questions!
Why Ensemble of Experts (EoE) framework?
 In recent years, a large number of binarization methods have been developed, but almost all suffer from
varying performance, generalization and strength against different benchmarks.
 There is, and will be, no winner approach in short (or even in long) term because of complexity of study
subjects (document and manuscript images) and also because of new processing goals.
 In this work, to leverage on all these methods of varying performance and interrelations, the ensemble of
experts (EoE) framework is introduced, to efficiently combine their outputs toward an output of higher
performance.
 The EoE framework can also be applied to other decision making problems:
 Medical image segmentation
 Parliament setting
 Opinion fraud detection
 However, caution should be taken when working with smart experts, such as humans, because they
could collectively adjust their behavior, having prior access to the rules of an EoE-based framework, to
win the ensemble’s result.
Ensemble of Experts vs. Ensemble of Classifiers
EoE EoC
 En ensemble.
 It work on a “set” of problems not just
one problem
 Every member is “free” to devise its
own approach to modeling and
concluding its opinion on each
problem.
 It could be seen as an enabler toward
featureless approaches.
 Performance evaluation is not easy
and straightforward.
 En ensemble.
 It (usually) works on one problem at a
time.
 Every member works on the
“regularized” representations of
problem, i.e., the feature vectors.
 Performance comparison is more
accurate and trustable because of
regularization approach used.
Basics of the EoE framework
 The proposed EoE framework offers a new expert selection process from an ensemble, by introducing three concepts:
confidentness, endorsement and schools of experts.
 The EoE framework tries to combine the outputs of an ensemble of related and unrelated experts using consolidation
and selection concepts toward an less-biased opinion.
 Endorsement graph:
 is defined based on the relations among the confidentness of the experts on their own opinions across the ensemble.
 Two generic selection principles:
 Consolidation of saturated opinions
 Selection of schools of experts
 For binarization methods, which lack the confidentness values, a confidentness map is defined.
 After building the endorsement graph of the ensemble for an input document image based on the confidentness of the
experts, the saturated opinions are consolidated, and then the schools of experts are identified by thresholding the
consolidated endorsement graph.
 The framework was successfully applied on the H-DIBCO’12 dataset. However, it is not limited only to handwritten
documents.
 A variation of the framework, in which no selection is made, is also introduced that combines the outputs of all experts
using endorsement-dependent weights (called EwEoE).
 Many aspects of the proposed framework could be improved.
EoE Framework: The Big Picture
EoE Framework is based on three concepts of
Confidentness, Endorsement, and School of Expert
0. Assemble the
Ensemble of Experts
1. Acquire the Set of
Problems
2. Get the Opinions
of experts on
problems
3. Calculate the
Confidentness of
each expert on each
problem
4. Calculate the
Endorsement Graph
among experts
5.1 Consolidate
highly-similar
experts (Reduce
Bias)
5.2 Calculate the
Schools of Experts
(clusters of experts)
6. Calculate the EoE
result by considering
only members of the
schools
7. Go back to step 1
to process a new set
of problems
Notations and application of the EoE Framework to document binarization
Currently, the methods do not provide any estimation of their
confidentness on individual pixels
EoE framework notation Equivalent in document binarization
1 An expert A binarization method
2 An Ensemble of Experts A set of binarization methods (can be
the same method with different
parameters)
3 A problem Binarization of a pixel
4 A set of problems Binarization of an image as a set of pixels
5 Opinion of an expert on a
problem
Binarization value of a method on a
pixel
6 Confidentness of an expert
on its opinion
<<To Be Defined>>
7 Endorsement (of expert A by
expert B)
Endorsement (of method A by method
B)
8 Endorsement graph Endorsement graph
Endorsement Graph Weights
The relation among confidentness maps on all pixels is used to define the weight of
corresponding edge on the endorsement graph
Confidentness of a on pixel i
masked by that of b
Endorsement b  a
EoE and EwEoE means
The selection processEoE-adjusted mean output
EwEoE-adjusted mean output
“Regular” mean output
An example of a highly-biased ensemble
84 experts using the Gb Sauvola method[1]
1. The Endorsement Matrix 2. Consolidated Endorsement Matrix 3. The selected experts
1. The Endorsement Graph 2. Consolidated Endorsement Graph 3. The selected experts (Graph)
[1] Farrahi Moghaddam, Reza, and Mohamed Cheriet. "A multi-scale framework for adaptive binarization of degraded document
images." Pattern Recognition 43, no. 6 (2010): 2186-2198. DOI: 10.1016/j.patcog.2009.12.024
EoE Framework Performance (1): H-DIBCO’12
Ensemble on H-DIBCO’12 datasetOriginal Endorsement Graph
of H-DIBCO’12 for H12
The performance
Final Schools of Expert for H12
EoE Framework Performance (2): Gb Sauvola
ensemble (84 experts) on H-DIBCO’12 datasetOriginal Endorsement Graph
of H-DIBCO’12 for H12
Final Schools of Expert for H12
EoE output for H05
“Regular” output for H05The performance
EoE Framework Performance (3): Laplacian-
energy[2] ensemble on H-DIBCO’12 dataset
The performance
H-DIBCO’12:H05 H-DIBCO’12:H09 H-DIBCO’12:H14
[2] Howe, Nicholas R. "Document binarization with automatic parameter tuning." International Journal on Document Analysis and
Recognition (IJDAR) (2012): 1-12. DOI: 10.1007/s10032-012-0192-x
Conclusions: The EoE framework
Summary Future Prospects
 The ensemble of experts (EoE) framework is
introduced, to efficiently combine the opinion of
experts methods on a set of problems.
 It is based on
 Confidentness
 Endorsement
 Schools of experts
 The EoE framework:
 combines the outputs of an ensemble of related and
unrelated experts using consolidation and selection concepts
toward reducing the bias of opinions.
 Endorsement graph is defined based on the
confidentness of the experts.
 Two generic principles of the EoE framework:
 Consolidation of saturated opinions
 Selection of schools of experts
 It has been applied to the H-DIBCO’12 database using various
ensembles of experts: H-DIBCO’12 participants, Gb Sauvola, and
Laplacian-energy.
 Generalization to other applications in other
decision making problems:
 Medical image segmentation
 Parliament setting
 Opinion fraud detection
 Improving the selection processes:
 Especially the consolidation step
 Adding another level of selection by selecting
one school out of all the EoE schools
 Improving the endorsement definition
 Standardization of the confidentness value as
the secondary output of an expert (a
binarization method) in addition to its opinion
value (binary output).
Thank you; any questions!
imriss@ieee.org, rfarrahi@synchromedia.ca
Synchromedia Lab ETS NSERC
http://arxiv.org/abs/1305.2949

Contenu connexe

En vedette (14)

Image enhancement
Image enhancement Image enhancement
Image enhancement
 
Penal especial
Penal especialPenal especial
Penal especial
 
Compresent pyramid by coon
Compresent pyramid by coonCompresent pyramid by coon
Compresent pyramid by coon
 
Dubrovnik Pres
Dubrovnik PresDubrovnik Pres
Dubrovnik Pres
 
insect legs
insect legsinsect legs
insect legs
 
conicoid
conicoidconicoid
conicoid
 
Contagious Messages
Contagious MessagesContagious Messages
Contagious Messages
 
Recommendation From ABC Industries 2014
Recommendation From ABC Industries 2014Recommendation From ABC Industries 2014
Recommendation From ABC Industries 2014
 
Threshold Selection for Image segmentation
Threshold Selection for Image segmentationThreshold Selection for Image segmentation
Threshold Selection for Image segmentation
 
Homicidio concausal
Homicidio concausalHomicidio concausal
Homicidio concausal
 
Logic Worktext
Logic Worktext Logic Worktext
Logic Worktext
 
Conodont
ConodontConodont
Conodont
 
SSS Condonation Program
SSS Condonation ProgramSSS Condonation Program
SSS Condonation Program
 
Conferring
ConferringConferring
Conferring
 

Similaire à Unsupervised ensemble of experts (EoE) framework for automatic binarization of document images

Testing survey by_directions
Testing survey by_directionsTesting survey by_directions
Testing survey by_directionsTao He
 
2cee Master Cocomo20071
2cee Master Cocomo200712cee Master Cocomo20071
2cee Master Cocomo20071CS, NcState
 
UNIT V TESTING.pptx
UNIT V TESTING.pptxUNIT V TESTING.pptx
UNIT V TESTING.pptxanguraju1
 
Fulcher and Davidson Unit a5
Fulcher and Davidson Unit a5Fulcher and Davidson Unit a5
Fulcher and Davidson Unit a5Farzaneh Hamidi
 
Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...butest
 
Hydraulics Team Full-Technical Lab Report
Hydraulics Team Full-Technical Lab ReportHydraulics Team Full-Technical Lab Report
Hydraulics Team Full-Technical Lab ReportAlfonso Figueroa
 
Object Oriented Design
Object Oriented DesignObject Oriented Design
Object Oriented DesignAMITJain879
 
Répondre à la question automatique avec le web
Répondre à la question automatique avec le webRépondre à la question automatique avec le web
Répondre à la question automatique avec le webAhmed Hammami
 
Gap Survey, Assessment and Analysis for DevSecOps
Gap Survey, Assessment and Analysis for DevSecOpsGap Survey, Assessment and Analysis for DevSecOps
Gap Survey, Assessment and Analysis for DevSecOpsMarc Hornbeek
 
The Weights Detection of Multi-Criteria by using Solver
The Weights Detection of Multi-Criteria by using Solver The Weights Detection of Multi-Criteria by using Solver
The Weights Detection of Multi-Criteria by using Solver IJECEIAES
 
Migration strategies for object oriented system to component based system
Migration strategies for object oriented system to component based systemMigration strategies for object oriented system to component based system
Migration strategies for object oriented system to component based systemijfcstjournal
 
QuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software EcosystemsQuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software EcosystemsGESSI UPC
 
Empirical Model of Supervised Learning Approach for Opinion Mining
Empirical Model of Supervised Learning Approach for Opinion MiningEmpirical Model of Supervised Learning Approach for Opinion Mining
Empirical Model of Supervised Learning Approach for Opinion MiningIRJET Journal
 
Mc0084 software project management & quality
Mc0084 software project management & qualityMc0084 software project management & quality
Mc0084 software project management & qualitysmumbahelp
 
1 Saint Leo University GBA 334 Applied Decision.docx
 1 Saint Leo University  GBA 334  Applied Decision.docx 1 Saint Leo University  GBA 334  Applied Decision.docx
1 Saint Leo University GBA 334 Applied Decision.docxaryan532920
 
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...SEAA 2022
 
Sentiment Analysis using Naïve Bayes, CNN, SVM
Sentiment Analysis using Naïve Bayes, CNN, SVMSentiment Analysis using Naïve Bayes, CNN, SVM
Sentiment Analysis using Naïve Bayes, CNN, SVMIRJET Journal
 
Comparison Of Methodologies
Comparison Of MethodologiesComparison Of Methodologies
Comparison Of Methodologiesguestc990b6
 

Similaire à Unsupervised ensemble of experts (EoE) framework for automatic binarization of document images (20)

Testing survey by_directions
Testing survey by_directionsTesting survey by_directions
Testing survey by_directions
 
2cee Master Cocomo20071
2cee Master Cocomo200712cee Master Cocomo20071
2cee Master Cocomo20071
 
UNIT V TESTING.pptx
UNIT V TESTING.pptxUNIT V TESTING.pptx
UNIT V TESTING.pptx
 
Fulcher and Davidson Unit a5
Fulcher and Davidson Unit a5Fulcher and Davidson Unit a5
Fulcher and Davidson Unit a5
 
Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...
 
Hydraulics Team Full-Technical Lab Report
Hydraulics Team Full-Technical Lab ReportHydraulics Team Full-Technical Lab Report
Hydraulics Team Full-Technical Lab Report
 
Object Oriented Design
Object Oriented DesignObject Oriented Design
Object Oriented Design
 
Répondre à la question automatique avec le web
Répondre à la question automatique avec le webRépondre à la question automatique avec le web
Répondre à la question automatique avec le web
 
Gap Survey, Assessment and Analysis for DevSecOps
Gap Survey, Assessment and Analysis for DevSecOpsGap Survey, Assessment and Analysis for DevSecOps
Gap Survey, Assessment and Analysis for DevSecOps
 
The Weights Detection of Multi-Criteria by using Solver
The Weights Detection of Multi-Criteria by using Solver The Weights Detection of Multi-Criteria by using Solver
The Weights Detection of Multi-Criteria by using Solver
 
Migration strategies for object oriented system to component based system
Migration strategies for object oriented system to component based systemMigration strategies for object oriented system to component based system
Migration strategies for object oriented system to component based system
 
QuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software EcosystemsQuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software Ecosystems
 
Empirical Model of Supervised Learning Approach for Opinion Mining
Empirical Model of Supervised Learning Approach for Opinion MiningEmpirical Model of Supervised Learning Approach for Opinion Mining
Empirical Model of Supervised Learning Approach for Opinion Mining
 
Mc0084 software project management & quality
Mc0084 software project management & qualityMc0084 software project management & quality
Mc0084 software project management & quality
 
1 Saint Leo University GBA 334 Applied Decision.docx
 1 Saint Leo University  GBA 334  Applied Decision.docx 1 Saint Leo University  GBA 334  Applied Decision.docx
1 Saint Leo University GBA 334 Applied Decision.docx
 
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...
Have Java Production Methods Co-Evolved With Test Methods Properly?: A Fine-G...
 
Kaggle's WISE 2014 challenge
Kaggle's WISE 2014 challenge Kaggle's WISE 2014 challenge
Kaggle's WISE 2014 challenge
 
Sentiment Analysis using Naïve Bayes, CNN, SVM
Sentiment Analysis using Naïve Bayes, CNN, SVMSentiment Analysis using Naïve Bayes, CNN, SVM
Sentiment Analysis using Naïve Bayes, CNN, SVM
 
De carlo rizk 2010 icelw
De carlo rizk 2010 icelwDe carlo rizk 2010 icelw
De carlo rizk 2010 icelw
 
Comparison Of Methodologies
Comparison Of MethodologiesComparison Of Methodologies
Comparison Of Methodologies
 

Plus de Reza Farrahi Moghaddam, PhD, BEng

40 Gbps Access for Metro networks: Implications in terms of Sustainability an...
40 Gbps Access for Metro networks: Implications in terms of Sustainability an...40 Gbps Access for Metro networks: Implications in terms of Sustainability an...
40 Gbps Access for Metro networks: Implications in terms of Sustainability an...Reza Farrahi Moghaddam, PhD, BEng
 
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...Reza Farrahi Moghaddam, PhD, BEng
 
Challenges and complexities in application of LCA approaches in the case of I...
Challenges and complexities in application of LCA approaches in the case of I...Challenges and complexities in application of LCA approaches in the case of I...
Challenges and complexities in application of LCA approaches in the case of I...Reza Farrahi Moghaddam, PhD, BEng
 
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...Reza Farrahi Moghaddam, PhD, BEng
 
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...Reza Farrahi Moghaddam, PhD, BEng
 
TIC pour un développement durable de la société (ICT for Sustainable Developm...
TIC pour un développement durable de la société (ICT for Sustainable Developm...TIC pour un développement durable de la société (ICT for Sustainable Developm...
TIC pour un développement durable de la société (ICT for Sustainable Developm...Reza Farrahi Moghaddam, PhD, BEng
 
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...Reza Farrahi Moghaddam, PhD, BEng
 
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing
Cognitive Behavior Analysis framework for Fault Prediction in Cloud ComputingCognitive Behavior Analysis framework for Fault Prediction in Cloud Computing
Cognitive Behavior Analysis framework for Fault Prediction in Cloud ComputingReza Farrahi Moghaddam, PhD, BEng
 

Plus de Reza Farrahi Moghaddam, PhD, BEng (11)

40 Gbps Access for Metro networks: Implications in terms of Sustainability an...
40 Gbps Access for Metro networks: Implications in terms of Sustainability an...40 Gbps Access for Metro networks: Implications in terms of Sustainability an...
40 Gbps Access for Metro networks: Implications in terms of Sustainability an...
 
Sustainability: Actors, Behavior, and Transparency
Sustainability: Actors, Behavior, and TransparencySustainability: Actors, Behavior, and Transparency
Sustainability: Actors, Behavior, and Transparency
 
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...
A Sustainable Future: Potentials of our Tools (ICT and Energy) and Responsibi...
 
Challenges and complexities in application of LCA approaches in the case of I...
Challenges and complexities in application of LCA approaches in the case of I...Challenges and complexities in application of LCA approaches in the case of I...
Challenges and complexities in application of LCA approaches in the case of I...
 
Sustainability Analysis of Broadband wireless access (BWA)
Sustainability Analysis of Broadband wireless access (BWA)Sustainability Analysis of Broadband wireless access (BWA)
Sustainability Analysis of Broadband wireless access (BWA)
 
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...
Reza Farrahi Moghaddam's Progress Report. From the Perspective of the Axe of ...
 
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...
Reza Farrahi Moghaddam’s Progress Report within the Perspective of the GSTC P...
 
TIC pour un développement durable de la société (ICT for Sustainable Developm...
TIC pour un développement durable de la société (ICT for Sustainable Developm...TIC pour un développement durable de la société (ICT for Sustainable Developm...
TIC pour un développement durable de la société (ICT for Sustainable Developm...
 
Life cycle assessment (LCA) for ICT
Life cycle assessment (LCA) for ICTLife cycle assessment (LCA) for ICT
Life cycle assessment (LCA) for ICT
 
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing...
 
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing
Cognitive Behavior Analysis framework for Fault Prediction in Cloud ComputingCognitive Behavior Analysis framework for Fault Prediction in Cloud Computing
Cognitive Behavior Analysis framework for Fault Prediction in Cloud Computing
 

Dernier

DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsSergiu Bodiu
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfSeasiaInfotech2
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024The Digital Insurer
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfAddepto
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Manik S Magar
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationSlibray Presentation
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embeddingZilliz
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek SchlawackFwdays
 

Dernier (20)

DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platforms
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdf
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdf
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck Presentation
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embedding
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
 

Unsupervised ensemble of experts (EoE) framework for automatic binarization of document images

  • 1. Reza FARRAHI MOGHADDAM, Fereydoun FARRAHI MOGHADDAM and Mohamed CHERIET Synchromedia Laboratory, ETS, Montreal (QC), Canada H3C 1K3 imriss@ieee.org, rfarrahi@synchromedia.ca, ffarrahi@synchromedia.ca, mohamed.cheriet@etsmtl.ca ICDAR 2013, Washington, DC, USA, August 25th-28th, 2013
  • 2. Outline  Why Ensemble of Experts (EoE) framework?  EoE vs. Ensemble of Classifiers (EoC)  The big picture  Notations  Endorsements and the Endorsement Graph  The selection process  Calculation of the EoE result and its variations  Use cases  Conclusions and future prospects  Any questions!
  • 3. Why Ensemble of Experts (EoE) framework?  In recent years, a large number of binarization methods have been developed, but almost all suffer from varying performance, generalization and strength against different benchmarks.  There is, and will be, no winner approach in short (or even in long) term because of complexity of study subjects (document and manuscript images) and also because of new processing goals.  In this work, to leverage on all these methods of varying performance and interrelations, the ensemble of experts (EoE) framework is introduced, to efficiently combine their outputs toward an output of higher performance.  The EoE framework can also be applied to other decision making problems:  Medical image segmentation  Parliament setting  Opinion fraud detection  However, caution should be taken when working with smart experts, such as humans, because they could collectively adjust their behavior, having prior access to the rules of an EoE-based framework, to win the ensemble’s result.
  • 4. Ensemble of Experts vs. Ensemble of Classifiers EoE EoC  En ensemble.  It work on a “set” of problems not just one problem  Every member is “free” to devise its own approach to modeling and concluding its opinion on each problem.  It could be seen as an enabler toward featureless approaches.  Performance evaluation is not easy and straightforward.  En ensemble.  It (usually) works on one problem at a time.  Every member works on the “regularized” representations of problem, i.e., the feature vectors.  Performance comparison is more accurate and trustable because of regularization approach used.
  • 5. Basics of the EoE framework  The proposed EoE framework offers a new expert selection process from an ensemble, by introducing three concepts: confidentness, endorsement and schools of experts.  The EoE framework tries to combine the outputs of an ensemble of related and unrelated experts using consolidation and selection concepts toward an less-biased opinion.  Endorsement graph:  is defined based on the relations among the confidentness of the experts on their own opinions across the ensemble.  Two generic selection principles:  Consolidation of saturated opinions  Selection of schools of experts  For binarization methods, which lack the confidentness values, a confidentness map is defined.  After building the endorsement graph of the ensemble for an input document image based on the confidentness of the experts, the saturated opinions are consolidated, and then the schools of experts are identified by thresholding the consolidated endorsement graph.  The framework was successfully applied on the H-DIBCO’12 dataset. However, it is not limited only to handwritten documents.  A variation of the framework, in which no selection is made, is also introduced that combines the outputs of all experts using endorsement-dependent weights (called EwEoE).  Many aspects of the proposed framework could be improved.
  • 6. EoE Framework: The Big Picture EoE Framework is based on three concepts of Confidentness, Endorsement, and School of Expert 0. Assemble the Ensemble of Experts 1. Acquire the Set of Problems 2. Get the Opinions of experts on problems 3. Calculate the Confidentness of each expert on each problem 4. Calculate the Endorsement Graph among experts 5.1 Consolidate highly-similar experts (Reduce Bias) 5.2 Calculate the Schools of Experts (clusters of experts) 6. Calculate the EoE result by considering only members of the schools 7. Go back to step 1 to process a new set of problems
  • 7. Notations and application of the EoE Framework to document binarization Currently, the methods do not provide any estimation of their confidentness on individual pixels EoE framework notation Equivalent in document binarization 1 An expert A binarization method 2 An Ensemble of Experts A set of binarization methods (can be the same method with different parameters) 3 A problem Binarization of a pixel 4 A set of problems Binarization of an image as a set of pixels 5 Opinion of an expert on a problem Binarization value of a method on a pixel 6 Confidentness of an expert on its opinion <<To Be Defined>> 7 Endorsement (of expert A by expert B) Endorsement (of method A by method B) 8 Endorsement graph Endorsement graph
  • 8. Endorsement Graph Weights The relation among confidentness maps on all pixels is used to define the weight of corresponding edge on the endorsement graph Confidentness of a on pixel i masked by that of b Endorsement b  a
  • 9. EoE and EwEoE means The selection processEoE-adjusted mean output EwEoE-adjusted mean output “Regular” mean output
  • 10. An example of a highly-biased ensemble 84 experts using the Gb Sauvola method[1] 1. The Endorsement Matrix 2. Consolidated Endorsement Matrix 3. The selected experts 1. The Endorsement Graph 2. Consolidated Endorsement Graph 3. The selected experts (Graph) [1] Farrahi Moghaddam, Reza, and Mohamed Cheriet. "A multi-scale framework for adaptive binarization of degraded document images." Pattern Recognition 43, no. 6 (2010): 2186-2198. DOI: 10.1016/j.patcog.2009.12.024
  • 11. EoE Framework Performance (1): H-DIBCO’12 Ensemble on H-DIBCO’12 datasetOriginal Endorsement Graph of H-DIBCO’12 for H12 The performance Final Schools of Expert for H12
  • 12. EoE Framework Performance (2): Gb Sauvola ensemble (84 experts) on H-DIBCO’12 datasetOriginal Endorsement Graph of H-DIBCO’12 for H12 Final Schools of Expert for H12 EoE output for H05 “Regular” output for H05The performance
  • 13. EoE Framework Performance (3): Laplacian- energy[2] ensemble on H-DIBCO’12 dataset The performance H-DIBCO’12:H05 H-DIBCO’12:H09 H-DIBCO’12:H14 [2] Howe, Nicholas R. "Document binarization with automatic parameter tuning." International Journal on Document Analysis and Recognition (IJDAR) (2012): 1-12. DOI: 10.1007/s10032-012-0192-x
  • 14. Conclusions: The EoE framework Summary Future Prospects  The ensemble of experts (EoE) framework is introduced, to efficiently combine the opinion of experts methods on a set of problems.  It is based on  Confidentness  Endorsement  Schools of experts  The EoE framework:  combines the outputs of an ensemble of related and unrelated experts using consolidation and selection concepts toward reducing the bias of opinions.  Endorsement graph is defined based on the confidentness of the experts.  Two generic principles of the EoE framework:  Consolidation of saturated opinions  Selection of schools of experts  It has been applied to the H-DIBCO’12 database using various ensembles of experts: H-DIBCO’12 participants, Gb Sauvola, and Laplacian-energy.  Generalization to other applications in other decision making problems:  Medical image segmentation  Parliament setting  Opinion fraud detection  Improving the selection processes:  Especially the consolidation step  Adding another level of selection by selecting one school out of all the EoE schools  Improving the endorsement definition  Standardization of the confidentness value as the secondary output of an expert (a binarization method) in addition to its opinion value (binary output).
  • 15. Thank you; any questions! imriss@ieee.org, rfarrahi@synchromedia.ca Synchromedia Lab ETS NSERC http://arxiv.org/abs/1305.2949