利用模數運算及多階段藏密提高資訊隱藏量之研究

資訊隱藏技術用意在保護訊息傳遞的安全性，主要先將訊息進行加密產生機密訊息後，再將機密訊息藏入多媒體之中；而近年不斷有研究者提出更高藏密量之各種新技術，但在過往技術中，如區塊分割、相鄰像素差異值等方法，在進行運算時，均因還原或者取密需求，難免出現其計算用之對照點(參考點)無法藏密的缺憾。為解決此一問題，並且有效提高藏密量，本論文提出以影像之相鄰像素值差異甚小以及多次運算後像素互補的特性，成功利用簡單的模數運算搭配預測誤差編碼技術，創造出更高的藏密量並維持不錯的影像品質。實驗結果顯示，本論文提供的龐大藏密空間，相當適合用於大量資料傳遞時的保護，更具有許多延展與發展性。

關鍵字

資訊隱藏；高藏密量；預測編碼

並列摘要

Information hiding technique is intended to protect the security of messaging. Generally, secure confidential information transmission is via encrypting and hiding. In recent years, a variety of new technologies have been proposed in order to achieve better hiding capacity. However, techniques with reference points involved, such as Block Segmentation and Pixel-Value Differencing methods, usually face the problem of limited capacity because those points cannot be used to hide secret message. To improve this problem, we propose an approach based on modulus and multi-stage hiding approaches to increase the hiding capacity, meanwhile, keep high image quality. Due to the use of pixel value correction technique and the characteristic of a natural image that the pixel values of neighboring pixel are closed, we are able to achieve the goal. According to our experimental results, the capacity is effectively increased and high image quality is also retained. The proposed approach is suitable for those applications that need a great amount of hiding capacity.