Deep learning methods and advancements in digital image forensics ; Méthodes et avancement d’apprentissage profond en criminalistique des images
Uloženo v:
| Název: | Deep learning methods and advancements in digital image forensics ; Méthodes et avancement d’apprentissage profond en criminalistique des images |
|---|---|
| Autoři: | Berthet, Alexandre |
| Přispěvatelé: | Eurecom Sophia Antipolis, Sorbonne Université, Jean-Luc Dugelay |
| Zdroj: | https://theses.hal.science/tel-03859790 ; Computer Aided Engineering. Sorbonne Université, 2022. English. ⟨NNT : 2022SORUS252⟩. |
| Informace o vydavateli: | CCSD |
| Rok vydání: | 2022 |
| Témata: | AI-based compression, Camera recognition, Digital image forensics, Contre-Criminalistique, Protocoles d'Évaluation, Compression basée sur l'Intelligence Artificielle, Reconnaissance de Caméras, Criminalistique des Images, [INFO.INFO-IA]Computer Science [cs]/Computer Aided Engineering |
| Popis: | The volume of digital visual data is increasing dramatically year after year. At the same time, image editing has become easier and more precise. Malicious modifications are therefore more accessible. Image forensics provides solutions to ensure the authenticity of digital visual data. Recognition of the source camera and detection of falsified images are among the main tasks. At first, the solutions were classical methods based on the artifacts produced during the creation of a digital image. Then, as in other areas of image processing, the methods moved to deep learning. First, we present a state-of-the-art survey of deep learning methods for image forensics. Our state-of-the-art survey highlights the need to apply pre-processing modules to extract artifacts hidden by image content. We also highlight the problems concerning image recognition evaluation protocols. Furthermore, we address counter-forensics and present compression based on artificial intelligence, which could be considered as an attack. In a second step, this thesis details three progressive evaluation protocols that address camera recognition problems. The final protocol, which is more reliable and reproducible, highlights the impossibility of state-of-the-art methods to recognize cameras in a challenging context. In a third step, we study the impact of compression based on artificial intelligence on two tasks analyzing compression artifacts: tamper detection and social network recognition. The performances obtained show on the one hand that this compression must be taken into account as an attack, but that it leads to a more important decrease than other manipulations for an equivalent image degradation. ; Le volume de données visuelles numériques augmente considérablement d'année en années. En parallèle, l’édition d'images est devenue plus facile et plus précise. Les modifications malveillantes sont donc plus accessibles. La criminalistique des images fournit des solutions pour garantir l’authenticité des données visuelles numériques. Tout ... |
| Druh dokumentu: | doctoral or postdoctoral thesis |
| Jazyk: | English |
| Relation: | NNT: 2022SORUS252 |
| Dostupnost: | https://theses.hal.science/tel-03859790 https://theses.hal.science/tel-03859790v1/document https://theses.hal.science/tel-03859790v1/file/BERTHET_Alexandre_these_2022.pdf |
| Rights: | info:eu-repo/semantics/OpenAccess |
| Přístupové číslo: | edsbas.6F1A5BFD |
| Databáze: | BASE |
Nájsť tento článok vo Web of Science | FullText | Text: Availability: 0 CustomLinks: – Url: https://theses.hal.science/tel-03859790# Name: EDS - BASE (s4221598) Category: fullText Text: View record from BASE – Url: https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=EBSCO&SrcAuth=EBSCO&DestApp=WOS&ServiceName=TransferToWoS&DestLinkType=GeneralSearchSummary&Func=Links&author=Berthet%20A Name: ISI Category: fullText Text: Nájsť tento článok vo Web of Science Icon: https://imagesrvr.epnet.com/ls/20docs.gif MouseOverText: Nájsť tento článok vo Web of Science |
|---|---|
| Header | DbId: edsbas DbLabel: BASE An: edsbas.6F1A5BFD RelevancyScore: 847 AccessLevel: 3 PubType: Dissertation/ Thesis PubTypeId: dissertation PreciseRelevancyScore: 847.000732421875 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Deep learning methods and advancements in digital image forensics ; Méthodes et avancement d’apprentissage profond en criminalistique des images – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Berthet%2C+Alexandre%22">Berthet, Alexandre</searchLink> – Name: Author Label: Contributors Group: Au Data: Eurecom Sophia Antipolis<br />Sorbonne Université<br />Jean-Luc Dugelay – Name: TitleSource Label: Source Group: Src Data: <i>https://theses.hal.science/tel-03859790 ; Computer Aided Engineering. Sorbonne Université, 2022. English. ⟨NNT : 2022SORUS252⟩</i>. – Name: Publisher Label: Publisher Information Group: PubInfo Data: CCSD – Name: DatePubCY Label: Publication Year Group: Date Data: 2022 – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22AI-based+compression%22">AI-based compression</searchLink><br /><searchLink fieldCode="DE" term="%22Camera+recognition%22">Camera recognition</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+image+forensics%22">Digital image forensics</searchLink><br /><searchLink fieldCode="DE" term="%22Contre-Criminalistique%22">Contre-Criminalistique</searchLink><br /><searchLink fieldCode="DE" term="%22Protocoles+d'Évaluation%22">Protocoles d'Évaluation</searchLink><br /><searchLink fieldCode="DE" term="%22Compression+basée+sur+l'Intelligence+Artificielle%22">Compression basée sur l'Intelligence Artificielle</searchLink><br /><searchLink fieldCode="DE" term="%22Reconnaissance+de+Caméras%22">Reconnaissance de Caméras</searchLink><br /><searchLink fieldCode="DE" term="%22Criminalistique+des+Images%22">Criminalistique des Images</searchLink><br /><searchLink fieldCode="DE" term="%22[INFO%2EINFO-IA]Computer+Science+[cs]%2FComputer+Aided+Engineering%22">[INFO.INFO-IA]Computer Science [cs]/Computer Aided Engineering</searchLink> – Name: Abstract Label: Description Group: Ab Data: The volume of digital visual data is increasing dramatically year after year. At the same time, image editing has become easier and more precise. Malicious modifications are therefore more accessible. Image forensics provides solutions to ensure the authenticity of digital visual data. Recognition of the source camera and detection of falsified images are among the main tasks. At first, the solutions were classical methods based on the artifacts produced during the creation of a digital image. Then, as in other areas of image processing, the methods moved to deep learning. First, we present a state-of-the-art survey of deep learning methods for image forensics. Our state-of-the-art survey highlights the need to apply pre-processing modules to extract artifacts hidden by image content. We also highlight the problems concerning image recognition evaluation protocols. Furthermore, we address counter-forensics and present compression based on artificial intelligence, which could be considered as an attack. In a second step, this thesis details three progressive evaluation protocols that address camera recognition problems. The final protocol, which is more reliable and reproducible, highlights the impossibility of state-of-the-art methods to recognize cameras in a challenging context. In a third step, we study the impact of compression based on artificial intelligence on two tasks analyzing compression artifacts: tamper detection and social network recognition. The performances obtained show on the one hand that this compression must be taken into account as an attack, but that it leads to a more important decrease than other manipulations for an equivalent image degradation. ; Le volume de données visuelles numériques augmente considérablement d'année en années. En parallèle, l’édition d'images est devenue plus facile et plus précise. Les modifications malveillantes sont donc plus accessibles. La criminalistique des images fournit des solutions pour garantir l’authenticité des données visuelles numériques. Tout ... – Name: TypeDocument Label: Document Type Group: TypDoc Data: doctoral or postdoctoral thesis – Name: Language Label: Language Group: Lang Data: English – Name: NoteTitleSource Label: Relation Group: SrcInfo Data: NNT: 2022SORUS252 – Name: URL Label: Availability Group: URL Data: https://theses.hal.science/tel-03859790<br />https://theses.hal.science/tel-03859790v1/document<br />https://theses.hal.science/tel-03859790v1/file/BERTHET_Alexandre_these_2022.pdf – Name: Copyright Label: Rights Group: Cpyrght Data: info:eu-repo/semantics/OpenAccess – Name: AN Label: Accession Number Group: ID Data: edsbas.6F1A5BFD |
| PLink | https://erproxy.cvtisr.sk/sfx/access?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.6F1A5BFD |
| RecordInfo | BibRecord: BibEntity: Languages: – Text: English Subjects: – SubjectFull: AI-based compression Type: general – SubjectFull: Camera recognition Type: general – SubjectFull: Digital image forensics Type: general – SubjectFull: Contre-Criminalistique Type: general – SubjectFull: Protocoles d'Évaluation Type: general – SubjectFull: Compression basée sur l'Intelligence Artificielle Type: general – SubjectFull: Reconnaissance de Caméras Type: general – SubjectFull: Criminalistique des Images Type: general – SubjectFull: [INFO.INFO-IA]Computer Science [cs]/Computer Aided Engineering Type: general Titles: – TitleFull: Deep learning methods and advancements in digital image forensics ; Méthodes et avancement d’apprentissage profond en criminalistique des images Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Berthet, Alexandre – PersonEntity: Name: NameFull: Eurecom Sophia Antipolis – PersonEntity: Name: NameFull: Sorbonne Université – PersonEntity: Name: NameFull: Jean-Luc Dugelay IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2022 Identifiers: – Type: issn-locals Value: edsbas – Type: issn-locals Value: edsbas.oa Titles: – TitleFull: https://theses.hal.science/tel-03859790 ; Computer Aided Engineering. Sorbonne Université, 2022. English. ⟨NNT : 2022SORUS252⟩ Type: main |
| ResultId | 1 |