應用改良式差異擴張法於彩色醫學影像之可逆資訊隱藏技術

Abstract

在本論文中我們提出了一個應用於彩色醫學影像的可逆資訊隱藏技術，而我們提出的可逆資訊隱藏技術包含了一個改良的差異擴張演算法和一個增加影像品質的嵌入規則。我們擴展了嵌入的條件以增加嵌入的範圍。在某些彩色的影像中，原先的差異擴張法可能會有嵌入資訊容量不足之問題，導致某些特定的資訊無法完全藏入，為了增進嵌入的容量，我們修改了原先差異擴張法的限制條件。此外為了進一步的增進影像的品質，我們定義了一個門檻值。

經由實驗的結果顯示，我們提出的改良方法能有效的增加嵌入的容量且使用我們的嵌入規則能在三個色彩平面中選擇其中一個色彩平面來嵌入資訊並獲得較佳的影像品質。由實驗結果獲得的峰值信噪比也顯示我們提出的可逆資訊隱藏技術更適合應用於彩色醫學影像。

In our research, we proposed a reversible data hiding scheme for color biomedical images, our proposed scheme consists of an improved difference expansion method and an embedding rule to elevate the image quality. We extended the region of embedding to increase capacity. In some color images, the original difference expansion method might not have enough capacity to embed specific information, so we modify the restrictions of original difference expansion method to increase the embedding capacity. In addition, we define a threshold value to enhance the image quality.

Experimental results show that our proposed method can increase the embedding capacity and our embedding rules can choose one of the three components in a color space for embedding in order to get the quality of embedded images. The results of Peak Signal-to-Ratio also demonstrated our proposed method is suitable for color biomedical images.