SlideShare a Scribd company logo
1 of 18
Download to read offline
Competence Centre on Information Extraction
    and Image Understanding for Earth Observation




                           July 2009
                                                       using OTB



                      Télécom ParisTech
                Marie Liénou - Marine Campedel
                                                 Image Semantic Coding




1
and Image Understanding for Earth Observation
                                                SUMMARY
 Competence Centre on Information Extraction




                                                     COC
                                                                  Notion of
                                                               semantic Coding
                                                                                      A promising
                                                                                       approach

                                                                Semantic Coding
                                                                   OTB tool

                                                                                  Development of
                                                      Conclusion and               an OTB tool
                                                       perspectives


                                                                                                    2
and Image Understanding for Earth Observation
                                                COC = COmpetence Centre…
 Competence Centre on Information Extraction



                                                 Tripartite agreement between CNES – DLR and Télécom
                                                  ParisTech
                                                 Signed in June 2005

                                                 Goal : joint action on image understanding
                                                    SAR/Optical, HR and VHR, temporal series
                                                    Feature extraction, modeling, indexing, compression,
                                                     (interactive) classification, interpretation, knowledge
                                                     representation, reasoning, …
                                                 Means
                                                    ~ 4 new phds / year
                                                    ~10 permanent researchers partially involved
                                                    financial support for specific actions (studentships, engineers,
                                                     post-docs)

                                                                                                                        3
and Image Understanding for Earth Observation
                                                Image Semantic coding
 Competence Centre on Information Extraction


                                                         Semantic                        Coding



                                                         Meaning                    Compression
                                                       Understanding              Reduce data while
                                                       Interpretation           ensuring informational
                                                       Image to text?                  content

                                                           [Barnard et al., 2003 ; Jeon et al., 2003]
                                                           [Li et Bretschneider, 2006]

                                                      Goal: find an image representation able to
                                                      capture the contained semantics
                                                      Idea: use text indexing approach + active learning

                                                                                                           4
and Image Understanding for Earth Observation
                                                Image Semantic coding
                                                            Visual interaction
 Competence Centre on Information Extraction



                                                            Manual annotation




                                                             Feature
                                                             extraction
                                                                                                  Where is
                                                           Quantization                          semantics?
                                                                                 Automatic
                                                                                 annotation
                                                           « visual words »
                                                                                              Active learning


                                                               Indexing                            Mining
                                                                                                                5
and Image Understanding for Earth Observation
                                                Image Semantic coding vs KIM
 Competence Centre on Information Extraction




                                                  « Design and evaluation of HMC for Image Information Mining »
                                                  Daschiel and Datcu
                                                  IEEE transaction on multimedia, vol 7, no6, dec. 2005



                                                                                                                  6
and Image Understanding for Earth Observation
                                                A promising approach
 Competence Centre on Information Extraction



                                                 Feature extraction
                                                    Segmentation, arbitrar regions
                                                    “Classical” signature: color, texture, shape descriptors
                                                    Experiments: intensity mean and variance in each spectral band


                                                 Quantization
                                                    K-Means: each estimated cluster corresponds to one “visual word”
                                                    K estimated using MDL (Minimum Description Length) descriptor


                                                 Bag-of-words signature for semantics identification
                                                    Count visual words on image regions which will be annotated
                                                    Normalize (tf-idf)


                                                 Exploitation using machine learning (SVM, LDA)
                                                                                                                   7
and Image Understanding for Earth Observation
                                                A promising approach
 Competence Centre on Information Extraction



                                                 Marie Lienou PhD work (march 2009)
                                                 Tested on several VHR (multispectral) images
                                                 Compared to other classification approachs (GMM, SVM)

                                                                        Visual word
                                                                        production
                                                Feature                                                    Classification
                                                             Quantization                Count words
                                                extraction                                                 SVM, LDA
                                                                                                                            Annotations

                                                Feature                     Classification
                                                                                                           Majority rule
                                                extraction                  GMM, SVM
                                                                                             Low level
                                                                                             annotations

                                                 Recognition accuracy demonstrated for “semantically complex”
                                                  classes Ex: “urban area”
                                                 LDA = fast + does not need negative examples
                                                                                                                                   8
and Image Understanding for Earth Observation
                                                OTB tool: cocSemanticCoding
 Competence Centre on Information Extraction


                                                 Feature extraction
                                                    Vectorial image with as many components as feature dimension
                                                    Exploitation of OTB extractors at each pixel
                                                 Quantization
                                                    Use of K-Means filter
                                                 Bag-of-words signature
                                                    Count visual words on image regions which will be annotated
                                                    Normalization (tf-idf)


                                                 Learning from manual annotation
                                                      Fluid interface facilities
                                                      Learn LDA from only target samples
                                                      Learn SVM from target samples and counter examples
                                                      Classify the whole image
                                                      Iterate
                                                                                                                   9
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation
                                                     OTB tool: cocSemanticCoding




10
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation
                                                     OTB tool: cocSemanticCoding




11
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation
                                                     OTB tool: cocSemanticCoding




12
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation




13
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation




14
Competence Centre on Information Extraction
     and Image Understanding for Earth Observation




15
and Image Understanding for Earth Observation
 Competence Centre on Information Extraction




                                                Learning and classification tools
                                                LDA on occurrence data
                                                SVM on TFiDF data (features)
                                                Both results can be obtained with same labeling for comparison
                                                Difficulty for the user : compute features adapted to the underlying
                                                semantics


                                                                                                                       16
and Image Understanding for Earth Observation
                                                Conclusion
 Competence Centre on Information Extraction


                                                 OTB useful features
                                                    Vectorial image representation
                                                    Great diversity of available filters (extractors, classifiers)
                                                        New = LDA classifier + estimator
                                                    Visualization tools


                                                 cocSemanticCoding tool availability
                                                    www.tsi.enst.fr/~campedel/
                                                    will be updated


                                                 Necessity to valorize research results
                                                    Engineering process (C++ programming)
                                                    Not easy but OTB is a nice initiative to help researchers
                                                    In the future: centralize processing tools (in OTB) + easy their
                                                     exploitation (documentations, interfaces)
                                                                                                                      17
and Image Understanding for Earth Observation
                                                Perspectives
 Competence Centre on Information Extraction


                                                 Other COC tools should be integrated in
                                                  cocSemanticCoding
                                                      MDL to estimate of visual words number
                                                      new feature extractors (QMF-based texture descriptors)
                                                      Feature selection
                                                      Complete relevance feedback framework

                                                 New approaches for image interpretation
                                                    From semantics to knowledge?
                                                    Knowledge engineering: modeling (ontologies) + reasoning
                                                    Several works on characterizing relations between identified
                                                     concepts and/or image objects




                                                                                                                    18

More Related Content

What's hot

SCHEME OF WORK 2010
SCHEME OF WORK 2010SCHEME OF WORK 2010
SCHEME OF WORK 2010
SMS
 
Economic Attention Networks
Economic Attention NetworksEconomic Attention Networks
Economic Attention Networks
Matthew Ikle
 
Elettronica: Multimedia Information Processing in Smart Environments by Aless...
Elettronica: Multimedia Information Processing in Smart Environments by Aless...Elettronica: Multimedia Information Processing in Smart Environments by Aless...
Elettronica: Multimedia Information Processing in Smart Environments by Aless...
Codemotion
 
Fcv bio cv_simoncelli
Fcv bio cv_simoncelliFcv bio cv_simoncelli
Fcv bio cv_simoncelli
zukun
 
Probabilistic generative models for machine vision
Probabilistic generative models for machine visionProbabilistic generative models for machine vision
Probabilistic generative models for machine vision
zukun
 
Kbms knowledge
Kbms knowledgeKbms knowledge
Kbms knowledge
okeee
 
3 a cognitive heuristic model of community recognition final
3 a cognitive heuristic model of community recognition final3 a cognitive heuristic model of community recognition final
3 a cognitive heuristic model of community recognition final
Ale Cignetti
 

What's hot (20)

IntelliGO semantic similarity measure for Gene Ontology annotations
IntelliGO semantic similarity measure for Gene Ontology annotationsIntelliGO semantic similarity measure for Gene Ontology annotations
IntelliGO semantic similarity measure for Gene Ontology annotations
 
Towards Neuro–Information Science
Towards Neuro–Information ScienceTowards Neuro–Information Science
Towards Neuro–Information Science
 
SCHEME OF WORK 2010
SCHEME OF WORK 2010SCHEME OF WORK 2010
SCHEME OF WORK 2010
 
Economic Attention Networks
Economic Attention NetworksEconomic Attention Networks
Economic Attention Networks
 
A Study of Semantic Proximity between Archetype Terms based on SNOMED CT Rela...
A Study of Semantic Proximity between Archetype Terms based on SNOMED CT Rela...A Study of Semantic Proximity between Archetype Terms based on SNOMED CT Rela...
A Study of Semantic Proximity between Archetype Terms based on SNOMED CT Rela...
 
Elettronica: Multimedia Information Processing in Smart Environments by Aless...
Elettronica: Multimedia Information Processing in Smart Environments by Aless...Elettronica: Multimedia Information Processing in Smart Environments by Aless...
Elettronica: Multimedia Information Processing in Smart Environments by Aless...
 
Tutorial kcc-2011
Tutorial kcc-2011Tutorial kcc-2011
Tutorial kcc-2011
 
Workshop on sparse image and signal processing
Workshop on sparse image and signal processingWorkshop on sparse image and signal processing
Workshop on sparse image and signal processing
 
Fcv bio cv_simoncelli
Fcv bio cv_simoncelliFcv bio cv_simoncelli
Fcv bio cv_simoncelli
 
Evolution: It's a process
Evolution: It's a processEvolution: It's a process
Evolution: It's a process
 
Oop
OopOop
Oop
 
Patent valuation using MDMP methodology
Patent valuation using MDMP methodologyPatent valuation using MDMP methodology
Patent valuation using MDMP methodology
 
Probabilistic generative models for machine vision
Probabilistic generative models for machine visionProbabilistic generative models for machine vision
Probabilistic generative models for machine vision
 
A Cognitive Heuristic model for Local Community Recognition
A Cognitive Heuristic model for Local Community RecognitionA Cognitive Heuristic model for Local Community Recognition
A Cognitive Heuristic model for Local Community Recognition
 
Fundamentals of visual communication unit iii
Fundamentals of visual communication unit iiiFundamentals of visual communication unit iii
Fundamentals of visual communication unit iii
 
Maya
MayaMaya
Maya
 
Kbms knowledge
Kbms knowledgeKbms knowledge
Kbms knowledge
 
Meaning across Disciplines
Meaning across DisciplinesMeaning across Disciplines
Meaning across Disciplines
 
Hoip10 presentacion video-vigilancia_uam
Hoip10 presentacion video-vigilancia_uamHoip10 presentacion video-vigilancia_uam
Hoip10 presentacion video-vigilancia_uam
 
3 a cognitive heuristic model of community recognition final
3 a cognitive heuristic model of community recognition final3 a cognitive heuristic model of community recognition final
3 a cognitive heuristic model of community recognition final
 

Viewers also liked

путь клиента
путь клиентапуть клиента
путь клиента
MEB
 
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
JISC SSBR
 
World hunger bueno
World hunger buenoWorld hunger bueno
World hunger bueno
Svalent
 
Vocab
VocabVocab
Vocab
twad
 
Activity 2. November 2010.Juvenile exploitation.Italy
Activity 2. November 2010.Juvenile exploitation.ItalyActivity 2. November 2010.Juvenile exploitation.Italy
Activity 2. November 2010.Juvenile exploitation.Italy
Carlos Ajamil Royo
 
English presentationiterlegis 130215
English presentationiterlegis 130215English presentationiterlegis 130215
English presentationiterlegis 130215
SilviaPN
 
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
Petros Michailidis
 
Compulink Core Presentation
Compulink Core PresentationCompulink Core Presentation
Compulink Core Presentation
Compulink
 
Bullfighting
BullfightingBullfighting
Bullfighting
Svalent
 
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
TNO
 
Flashback Scenes
Flashback ScenesFlashback Scenes
Flashback Scenes
kay91
 

Viewers also liked (20)

Valids Presentation (2010.07)
Valids Presentation (2010.07)Valids Presentation (2010.07)
Valids Presentation (2010.07)
 
путь клиента
путь клиентапуть клиента
путь клиента
 
Ict
IctIct
Ict
 
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
Jane Charlton Jisc Comms Inst Innov Project C M Jan2010
 
World hunger bueno
World hunger buenoWorld hunger bueno
World hunger bueno
 
What is the future of cloud security linked in
What is the future of cloud security linked inWhat is the future of cloud security linked in
What is the future of cloud security linked in
 
Vocab
VocabVocab
Vocab
 
Activity 2. November 2010.Juvenile exploitation.Italy
Activity 2. November 2010.Juvenile exploitation.ItalyActivity 2. November 2010.Juvenile exploitation.Italy
Activity 2. November 2010.Juvenile exploitation.Italy
 
English presentationiterlegis 130215
English presentationiterlegis 130215English presentationiterlegis 130215
English presentationiterlegis 130215
 
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
Kas saribelen questionnaire for 1 2-3-4-5 ing-2011
 
Φύλλο εικόνας Background
Φύλλο εικόνας BackgroundΦύλλο εικόνας Background
Φύλλο εικόνας Background
 
Μπουμπουλίνα
ΜπουμπουλίναΜπουμπουλίνα
Μπουμπουλίνα
 
Compulink Core Presentation
Compulink Core PresentationCompulink Core Presentation
Compulink Core Presentation
 
Bullfighting
BullfightingBullfighting
Bullfighting
 
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
Cocreatie voor en door het onderwijs - marketing management congres fontys ho...
 
101108 goldmedia web tv monitor 2010_english
101108 goldmedia web tv monitor 2010_english101108 goldmedia web tv monitor 2010_english
101108 goldmedia web tv monitor 2010_english
 
Slides for the open evening rp 27.05.10
Slides for the open evening rp 27.05.10Slides for the open evening rp 27.05.10
Slides for the open evening rp 27.05.10
 
Shashikumar_CV
Shashikumar_CVShashikumar_CV
Shashikumar_CV
 
Flashback Scenes
Flashback ScenesFlashback Scenes
Flashback Scenes
 
How To Help Employers Recognise Your Talent
How To Help Employers Recognise Your TalentHow To Help Employers Recognise Your Talent
How To Help Employers Recognise Your Talent
 

Similar to Image semantic coding using OTB

Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
SEO CAMP
 
Open Source for Enterprise Search: Breaking Down the Barriers to Information
Open Source for Enterprise Search: Breaking Down the Barriers to InformationOpen Source for Enterprise Search: Breaking Down the Barriers to Information
Open Source for Enterprise Search: Breaking Down the Barriers to Information
Lucidworks (Archived)
 
We Know It (Newsfromthefront 2010)
We Know It (Newsfromthefront 2010)We Know It (Newsfromthefront 2010)
We Know It (Newsfromthefront 2010)
STI International
 
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
imec.archive
 
Multilabel Image Annotation using Multimodal Analysis
Multilabel Image Annotation using Multimodal AnalysisMultilabel Image Annotation using Multimodal Analysis
Multilabel Image Annotation using Multimodal Analysis
ijtsrd
 

Similar to Image semantic coding using OTB (20)

Fundamentals of Information Architecture Workshop
Fundamentals of Information Architecture WorkshopFundamentals of Information Architecture Workshop
Fundamentals of Information Architecture Workshop
 
Taming digital traces for informal learning dhaval
Taming digital traces for informal learning  dhavalTaming digital traces for informal learning  dhaval
Taming digital traces for informal learning dhaval
 
Conferencia Web semantica Mihai Datcu
Conferencia  Web semantica Mihai DatcuConferencia  Web semantica Mihai Datcu
Conferencia Web semantica Mihai Datcu
 
Impact centre of_competence_for_workshop_ocr_rouen_march_2011[1]
Impact centre of_competence_for_workshop_ocr_rouen_march_2011[1]Impact centre of_competence_for_workshop_ocr_rouen_march_2011[1]
Impact centre of_competence_for_workshop_ocr_rouen_march_2011[1]
 
Towards a Better Understanding of Model-Free Semantic Concept Detection for A...
Towards a Better Understanding of Model-Free Semantic Concept Detection for A...Towards a Better Understanding of Model-Free Semantic Concept Detection for A...
Towards a Better Understanding of Model-Free Semantic Concept Detection for A...
 
Computing for Human Experience and Wellness
Computing for Human Experience and WellnessComputing for Human Experience and Wellness
Computing for Human Experience and Wellness
 
Towards Timely Efficient Semantic Reasoning for the Networked Society
Towards Timely Efficient Semantic Reasoning for the Networked SocietyTowards Timely Efficient Semantic Reasoning for the Networked Society
Towards Timely Efficient Semantic Reasoning for the Networked Society
 
Utilizing Semantics in the Production of iTV Shows (ESWC 2009)
Utilizing Semantics in the Production of iTV Shows (ESWC 2009)Utilizing Semantics in the Production of iTV Shows (ESWC 2009)
Utilizing Semantics in the Production of iTV Shows (ESWC 2009)
 
On Semantics in Onto-DIY
On Semantics in Onto-DIYOn Semantics in Onto-DIY
On Semantics in Onto-DIY
 
Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
Jean-Marc Lazard d'Exalead - Pioneering hypermedia - SEO Campus 2011
 
Open Source for Enterprise Search: Breaking Down the Barriers to Information
Open Source for Enterprise Search: Breaking Down the Barriers to InformationOpen Source for Enterprise Search: Breaking Down the Barriers to Information
Open Source for Enterprise Search: Breaking Down the Barriers to Information
 
Semantic Web powering Enterprise and Web Applications
Semantic Web powering Enterprise and Web ApplicationsSemantic Web powering Enterprise and Web Applications
Semantic Web powering Enterprise and Web Applications
 
We Know It (Newsfromthefront 2010)
We Know It (Newsfromthefront 2010)We Know It (Newsfromthefront 2010)
We Know It (Newsfromthefront 2010)
 
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
Acknowledge 06 Competentiemanagement In Het Acknowledge Platform Luk Vervenne...
 
Multilabel Image Annotation using Multimodal Analysis
Multilabel Image Annotation using Multimodal AnalysisMultilabel Image Annotation using Multimodal Analysis
Multilabel Image Annotation using Multimodal Analysis
 
Building Task-Aware Mobile Applications
Building Task-Aware Mobile ApplicationsBuilding Task-Aware Mobile Applications
Building Task-Aware Mobile Applications
 
Knowledge mobilization
Knowledge mobilization Knowledge mobilization
Knowledge mobilization
 
Albert Simard - Mobilizing Knowledge: Acquisition, Analysis, and Action
Albert Simard - Mobilizing Knowledge: Acquisition, Analysis, and ActionAlbert Simard - Mobilizing Knowledge: Acquisition, Analysis, and Action
Albert Simard - Mobilizing Knowledge: Acquisition, Analysis, and Action
 
TCS Innovation Forum 2012 - Big Data Whodini
TCS Innovation Forum 2012 - Big Data WhodiniTCS Innovation Forum 2012 - Big Data Whodini
TCS Innovation Forum 2012 - Big Data Whodini
 
Sussex 210612
Sussex 210612Sussex 210612
Sussex 210612
 

More from melaneum

More from melaneum (9)

Implementing kohonen's som with missing data in OTB
Implementing kohonen's som with missing data in OTBImplementing kohonen's som with missing data in OTB
Implementing kohonen's som with missing data in OTB
 
Overview of the PolSARpro V4.0 software. The open source toolbox for polarime...
Overview of the PolSARpro V4.0 software. The open source toolbox for polarime...Overview of the PolSARpro V4.0 software. The open source toolbox for polarime...
Overview of the PolSARpro V4.0 software. The open source toolbox for polarime...
 
Toward a gui remote-sensing environment built over OTB
Toward a gui remote-sensing environment built over OTBToward a gui remote-sensing environment built over OTB
Toward a gui remote-sensing environment built over OTB
 
Urban area detection and segmentation using OTB
Urban area detection and segmentation using OTBUrban area detection and segmentation using OTB
Urban area detection and segmentation using OTB
 
The use of Orfeo Toolbox in the context of map updating
The use of Orfeo Toolbox in the context of map updatingThe use of Orfeo Toolbox in the context of map updating
The use of Orfeo Toolbox in the context of map updating
 
Assessment of interest points detection algorithms in OTB
Assessment of interest points detection algorithms in OTBAssessment of interest points detection algorithms in OTB
Assessment of interest points detection algorithms in OTB
 
Reference algorithm implementations in OTB: textbook cases
Reference algorithm implementations in OTB: textbook casesReference algorithm implementations in OTB: textbook cases
Reference algorithm implementations in OTB: textbook cases
 
Object counting in high resolution remote sensing images with OTB
Object counting in high resolution remote sensing images with OTBObject counting in high resolution remote sensing images with OTB
Object counting in high resolution remote sensing images with OTB
 
The Orfeo Toolbox remote sensing image processing software
The Orfeo Toolbox remote sensing image processing softwareThe Orfeo Toolbox remote sensing image processing software
The Orfeo Toolbox remote sensing image processing software
 

Recently uploaded

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
vu2urc
 

Recently uploaded (20)

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 

Image semantic coding using OTB

  • 1. Competence Centre on Information Extraction and Image Understanding for Earth Observation July 2009 using OTB Télécom ParisTech Marie Liénou - Marine Campedel Image Semantic Coding 1
  • 2. and Image Understanding for Earth Observation SUMMARY Competence Centre on Information Extraction COC Notion of semantic Coding A promising approach Semantic Coding OTB tool Development of Conclusion and an OTB tool perspectives 2
  • 3. and Image Understanding for Earth Observation COC = COmpetence Centre… Competence Centre on Information Extraction  Tripartite agreement between CNES – DLR and Télécom ParisTech  Signed in June 2005  Goal : joint action on image understanding  SAR/Optical, HR and VHR, temporal series  Feature extraction, modeling, indexing, compression, (interactive) classification, interpretation, knowledge representation, reasoning, …  Means  ~ 4 new phds / year  ~10 permanent researchers partially involved  financial support for specific actions (studentships, engineers, post-docs) 3
  • 4. and Image Understanding for Earth Observation Image Semantic coding Competence Centre on Information Extraction Semantic Coding Meaning Compression Understanding Reduce data while Interpretation ensuring informational Image to text? content [Barnard et al., 2003 ; Jeon et al., 2003] [Li et Bretschneider, 2006] Goal: find an image representation able to capture the contained semantics Idea: use text indexing approach + active learning 4
  • 5. and Image Understanding for Earth Observation Image Semantic coding Visual interaction Competence Centre on Information Extraction Manual annotation Feature extraction Where is Quantization semantics? Automatic annotation « visual words » Active learning Indexing Mining 5
  • 6. and Image Understanding for Earth Observation Image Semantic coding vs KIM Competence Centre on Information Extraction « Design and evaluation of HMC for Image Information Mining » Daschiel and Datcu IEEE transaction on multimedia, vol 7, no6, dec. 2005 6
  • 7. and Image Understanding for Earth Observation A promising approach Competence Centre on Information Extraction  Feature extraction  Segmentation, arbitrar regions  “Classical” signature: color, texture, shape descriptors  Experiments: intensity mean and variance in each spectral band  Quantization  K-Means: each estimated cluster corresponds to one “visual word”  K estimated using MDL (Minimum Description Length) descriptor  Bag-of-words signature for semantics identification  Count visual words on image regions which will be annotated  Normalize (tf-idf)  Exploitation using machine learning (SVM, LDA) 7
  • 8. and Image Understanding for Earth Observation A promising approach Competence Centre on Information Extraction  Marie Lienou PhD work (march 2009)  Tested on several VHR (multispectral) images  Compared to other classification approachs (GMM, SVM) Visual word production Feature Classification Quantization Count words extraction SVM, LDA Annotations Feature Classification Majority rule extraction GMM, SVM Low level annotations  Recognition accuracy demonstrated for “semantically complex” classes Ex: “urban area”  LDA = fast + does not need negative examples 8
  • 9. and Image Understanding for Earth Observation OTB tool: cocSemanticCoding Competence Centre on Information Extraction  Feature extraction  Vectorial image with as many components as feature dimension  Exploitation of OTB extractors at each pixel  Quantization  Use of K-Means filter  Bag-of-words signature  Count visual words on image regions which will be annotated  Normalization (tf-idf)  Learning from manual annotation  Fluid interface facilities  Learn LDA from only target samples  Learn SVM from target samples and counter examples  Classify the whole image  Iterate 9
  • 10. Competence Centre on Information Extraction and Image Understanding for Earth Observation OTB tool: cocSemanticCoding 10
  • 11. Competence Centre on Information Extraction and Image Understanding for Earth Observation OTB tool: cocSemanticCoding 11
  • 12. Competence Centre on Information Extraction and Image Understanding for Earth Observation OTB tool: cocSemanticCoding 12
  • 13. Competence Centre on Information Extraction and Image Understanding for Earth Observation 13
  • 14. Competence Centre on Information Extraction and Image Understanding for Earth Observation 14
  • 15. Competence Centre on Information Extraction and Image Understanding for Earth Observation 15
  • 16. and Image Understanding for Earth Observation Competence Centre on Information Extraction Learning and classification tools LDA on occurrence data SVM on TFiDF data (features) Both results can be obtained with same labeling for comparison Difficulty for the user : compute features adapted to the underlying semantics 16
  • 17. and Image Understanding for Earth Observation Conclusion Competence Centre on Information Extraction  OTB useful features  Vectorial image representation  Great diversity of available filters (extractors, classifiers)  New = LDA classifier + estimator  Visualization tools  cocSemanticCoding tool availability  www.tsi.enst.fr/~campedel/  will be updated  Necessity to valorize research results  Engineering process (C++ programming)  Not easy but OTB is a nice initiative to help researchers  In the future: centralize processing tools (in OTB) + easy their exploitation (documentations, interfaces) 17
  • 18. and Image Understanding for Earth Observation Perspectives Competence Centre on Information Extraction  Other COC tools should be integrated in cocSemanticCoding  MDL to estimate of visual words number  new feature extractors (QMF-based texture descriptors)  Feature selection  Complete relevance feedback framework  New approaches for image interpretation  From semantics to knowledge?  Knowledge engineering: modeling (ontologies) + reasoning  Several works on characterizing relations between identified concepts and/or image objects 18