Analysis of watermarking framework for color image through a neural network-based approach

TLDR: The approaches of the embedding and the de-embedding in case of learning algorithm of the aforementioned neural network through individual training data set are considered in the present research to carry out a series of experiments with different scenarios for the purpose of verifying the effectiveness of the proposed approach.

Abstract: In the research presented here, the general idea of watermarking framework is analyzed to deal with color image under a set of attacks through a neural network-based approach. It is realized in the area of transformation, especially with a focus on contourlet transform to address the proposed technique, as long as the bands of the suitable coefficients are accurately chosen. In summary, there is the logo information that is embedded in the edge of color image, while the Zenzo edge detector is correspondingly realized to handle the approach. In fact, the edge of the second subband is acquired, and subsequently, the capability of the above-referenced edge is calculated. A number of techniques are discussed to cope with the above-captioned watermarking framework through the new integration of contourlet transform in association with the multilayer perceptron to extract the logo information, appropriately. The approaches of the embedding and the de-embedding in case of learning algorithm of the aforementioned neural network through individual training data set are considered in the present research to carry out a series of experiments with different scenarios for the purpose of verifying the effectiveness of the proposed approach, obviously.