Inserção e extração de marca d'água em imagens digitais usando a transformada wavelet

• Autor(a) principal: Silva, Jaqueline Ferreira da [UNESP]
• Data de Publicação: 2014
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/123360; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/23-04-2015/000823589.pdf
• topic: Processamento de imagens; Criptografia de dados (Computação); Arquivos de computador; Medidas de segurança; Wavelets (Matematica); Image processing

The growing use of digital files through the media takes the user to employ security techniques against tampering attacks. The watermark is considered an additional information that can ensure security, detect and track changes in an image marked according to the cryptographic algorithm deployed. In general, digital signals such as images files are vulnerable to tampering attacks due to the way in which they circulate. Through digital watermarks, it is possible to obtain authenticity of an image. In cryptographic terms, the watermark embedded in the host image must be as compact as possible and its processing must be effective. After a theoretical survey on the implementation of the watermark images and, later, an analysis of the main methods of insertion, extraction and verification of the watermark, grounded by better performances, we present in this paper, a method of insertion, extraction and verification of the watermark. The proposed method, implemented in the wavelet domain, along with simple matrix manipulations used in cryptography aims to provide a robust watermark on various types of digital attacks. The method allows the inserting of the watermark without affecting the visual quality of the host image and after the attacks, extract it successfully, presenting a correlation between the watermark inserted and extracted close to optimum. The image processing programs via Wavelet transformed and methods of insertion, extraction and verification were developed in the MATLAB ® environment, which makes the manipulations in the host image and the watermark easier whereas the images are read as pixel matrix. Finally, we show how the very original results shown here suggest other types of attacks and increase the robustness of the watermark