透過您的圖書館登入
IP:52.14.121.242
  • 期刊

利用影像內插與失真評估之資料隱藏演算法

Data Hiding Using Image Interpolation and Distortion Estimation Schemes

摘要


本篇論文針對灰階影像提出了一個具有高藏量、高偽裝影像品質,以及可回復之資料隱藏技術。我們的演算法是基於影像內插並使用混合式OPAP搭配LSB-matching所做的資訊隱藏技術。本研究利用平均每個像素的期望變動量分析,可以明確的預測在偽裝影像品質固定之下其所產生之最高藏量,以及在嵌入量固定之下其所產生之最高偽裝影像品質。我們發現Jung等人之演算法在某些情況下無法將秘密訊息完全正確取出,我們修正其演算法使秘密訊息能夠正確無誤取出。我們再另外提出一個資料隱藏方法與Jung等人之方法比較。於本論文中,我們不僅改正了Jung等人演算法之錯誤。我們另外提出的方法具有高藏量、高偽裝影像品質以及可回復之特性,僅需要極少的額外資訊來取出秘密訊息。此外,本方法也可抵擋RS偵測攻擊。

並列摘要


This paper proposes a data hiding algorithm that provides high capacity, high visual quality, and reversibility for gray level image. Inspired from the image interpolation proposed by Jung et al., we use hybrid OPAP and LSB-matching to achieve goal of the data hiding. When the visual quality of the stego image is fixed, our method can predict the maximal capacity through an analysis of the Expected Variation per Pixel (EVP) technique. On the other hand, the EVP technique can predict an optimal visual qualityof the stego image when we fix the capacity of the stego image. We discover that Jung et al.'s method is incapable of correctly extracting secret data in some conditions. We correct their algorithm so that it can extract message without encountering any errors. Experimental results verify that our method provides high capacity and high visual quality. In addition, our scheme supports the feature of reversibility, requiring very little extra information to extract the secret message. Furthermore, the proposed method can withstand the RS steganalysis attack.

延伸閱讀