Digital Watermarking and Signing by Artificial Neural Networks
Tarik Hajji, Jaara El Miloud · American Journal of Intelligent Systems · 2014
The objective of this work is the description of a new approach based on artificial neural networks to signing the digital images using the principle of watermarking. The idea of this approach is to watermark the image to be protected by the compressed image previously generated by the artificial neural network. This approach provided, on the one hand, a very efficient approach to digital signature for protecting the integrity of the image and by the other hand; it provides an easy and immediate mechanism for verification. The role of artifices neural networks in this approach is to generate the compressed image that will be used in the sequel to the watermarking of the image by using an algorithm for optimal positioning and that does not change the visual appearance of the image and also to verify the signature. This work also includes a comparative study to select the structure and parameters of the artificial neural network are performing for the problem of the compression and decompression of digital images. Such as, it contains a description of an application that allows for the generation of an artificial neural network with a simple Meta model description.