| Summary: | 博士 === 國防大學中正理工學院 === 國防科學研究所 === 89 === The research of image steganography becomes an interesting topic consequently. There are many applications like caption data embedding, secret message delivering, copyright protection, watermarking, etc. And the strategies fall into two major categories: information hiding and digital watermarking. Information hiding is a technique of hiding secret concealed in digital media. Using this scheme, we can confuse the intruders and prevent the detection of hidden secret. Watermarking is a technique to identify the rightful owner of multimedia data. When the ownership of data is in question, the watermark can be extracted to prove the copyright.
In this dissertation, a novel adaptive image steganograpy approach based on the human visual system model and fuzzy logic method is proposed. The human visual system model is utilized to guarantee the suitable strength of embedded secret. The fuzzy logic method has been employed to obtain the local characters of image. The adaptive strength can meet two important requirements: transparency and robustness.
Two main fuzzy logic methods are involved in the proposed adaptive image steganography: fuzzy inference system and fuzzy c-means algorithm. The estimation of image visual characters is resolved by these fuzzy logic methods. Applying the human visual system model on the image steganography scheme, we can depend on the imperceptible and perceptible areas of our image and then decide to increase and decrease the amount of hiding data. The degree of visual transparency from the image modified by our approaches can be constrained, and the degree of secret robustness can be increased at the same time.
|
|---|
