| Summary: | 碩士 === 國立彰化師範大學 === 資訊工程學系 === 94 === Information hiding is a technique that embeds secret data in a cover data such as text, image, audio, video, etc. By this way, the grabber would not notice and the secret data can transmit to destination safely. The main point of information hiding is to decrease the destruction of the cover data and to increase the embedding capacity.
In this thesis, we propose two embedding methods. The first method improve the steganography that refer some neighbor pixel values to hid secret data. In this improve method, we not only improve the capacity, but also increase the stego-image quality. The target pixel may lie in a rough area when the difference between the mean of neighbor pixels and the target is larg. The target pixel may lie in a smooth area when the difference between the mean of neighbor pixels and the target is small. Although it fits to the human vision sensitivity (HVS) property by using neighbor pixels to compute how many bits can be embed in, it can not get the area property accurately. Hence in the second method, we propose a method that combines the HVS to embed secret data. By HVS, we can get the area property where the target pixel lie in more accurately, and we can compute fitter pixel and embeding bits.
By experimental results, our method can improve the capacity and the stego-image quality effectively. By using HVS, we can provide more imperceptible in human eye.
|
|---|
