Self-recovery digital watermarking scheme robust against spatial and temporal attacks on compressed video
In recent decades the use of digital media has increased, and with it, the exposure of them to non desired manipulations that, although they may be unintentional, they may affect third parties. These manipulations in the area of digital watermarking are known as attacks, which can consist of intenti...
| Autor: | |
|---|---|
| Formato: | tesis de maestría |
| Estado: | Versión aceptada para publicación |
| Fecha de publicación: | 2020 |
| País: | México |
| Recursos: | Instituto Nacional de Astrofísica, Óptica y Electrónica |
| Repositorio: | Repositorio Institucional del INAOE |
| Idioma: | inglés |
| OAI Identifier: | oai:inaoe.repositorioinstitucional.mx:1009/2165 |
| Acesso em linha: | http://inaoe.repositorioinstitucional.mx/jspui/handle/1009/2165 |
| Access Level: | acceso abierto |
| Palavra-chave: | info:eu-repo/classification/Inspec/Watermarking info:eu-repo/classification/Inspec/Video encoding info:eu-repo/classification/Inspec/Self-recovery info:eu-repo/classification/Inspec/Authentication info:eu-repo/classification/Inspec/Transform domain info:eu-repo/classification/Inspec/Compressed Domain info:eu-repo/classification/cti/1 info:eu-repo/classification/cti/12 info:eu-repo/classification/cti/1203 info:eu-repo/classification/cti/330414 |
| Resumo: | In recent decades the use of digital media has increased, and with it, the exposure of them to non desired manipulations that, although they may be unintentional, they may affect third parties. These manipulations in the area of digital watermarking are known as attacks, which can consist of intentionally or unintentionally editing multimedia content such as images, videos or audio sequences. For video sequences, attacks may be performed both in the spatial and temporal domain. The first one includes changes in individual frames while the second includes manipulation of sub-sequences such as inserting or deleting frames. There is a necessity for digital content validation that, besides authenticating the data, they can recover the content that has been altered. On the one hand, passive techniques usually analyze content using machine learning or correlation methods to predict whether the content is authentic or not, and if it is manipulated, the probability that it is temporal or spatial. On the other hand, active techniques embed watermarks or digital signatures in the medium to detect which parts of it were edited, calling these active techniques selfauthentication. Digital video is typically compressed before being stored or transmitted. Active authentication techniques require redundancy to insert data, but compression removes most of this redundancy. The recovery techniques are part of the active methods, they work well spatially, but have problems with compression, which leads to the objective of this research, to design a robust self-recovery technique for compressed video that has a limited watermark payload size without compromising perceptual quality. |
|---|