透過您的圖書館登入
IP:18.191.202.45
  • 期刊

植基於像素差值與模數函數之新型灰階影像資料隱藏技術

A Novel Data Hiding Method for Grayscale Images Based on Pixel-Value Differencing and Modulus Function

摘要


本論文的主要研究目標是針對灰階影像研發出一個失真性非常少的新型資料隱藏技術,新方法主要是架構於像素差值法與模數函數運算的基礎上,來達到一個既符合人類視覺要求特性又可以有效減少偽裝影像的失真性。新方法主要是以二個連續像素為一個子影像單位來攜帶秘密資料,而每一個子影像的可嵌入量大小是由像素差值來決定的,在訊息嵌入的過程中,我們提出研發出一種新穎的訊息嵌入技術,可以讓秘密訊息在嵌入像素之後,其所產生的像素失真性降到最低。我們的新方法主要是採用了非十進制的數字系統,並且搭配模數運算的操作方式,可以有效地使用最少的像素變動量,即可以嵌入相對應的秘密訊息。此外,由於隱藏量是分成六等級的,因此,為了確保每一個像素組在藏入資料前、後能維持影像區塊等級的一致性,本文中也提出了隱藏量等級一致性校正的法,如此一來,接收方只要依照偽裝像素對的差值大小,即可正確判斷子影像區塊的隱藏量,故新方法具有盲擷取的特性,即不需要原始影像即可正確擷取的秘密訊息。最後,經由大量的實驗結果,証明了本文新方法確實可以有效地減少偽裝影像的失真性,並且在相同的隱藏量條件下,其所產生的偽裝影像品質與現存相同類型的技術相比較,其失真性確實可以減少地非常的多。

並列摘要


The purpose of this paper is to present a novel data hiding technique which is based on pixel-value differencing and modulus function and is capable of reducing the distortion of stego image for graylevel image. In the proposed scheme, we take two close pixels as a block to hide secret data where the payload of each block is determined by pixel-value differencing. In order to obtain a better stego-image quality, we hide more data into the blocks with the higher difference value than that the blocks with the smaller difference value. In data embedding procedure, we propose a novel technique which is capable of reducing the distortion of stego image with the minimal. In our new scheme, we adopt the non-decimal number system to represent the secret data, and utilize the modulus function to minimize the possible distortion after hiding data. Since there are six degrees of payload are used in our method, we present the revision method to ensure the consistency of degree for each block. Therefore, the proposed scheme is blind because the original image is unnecessary in the data extraction. The experimental results demonstrate that the proposed scheme is greatly reducing distortion and its performance is better than that of those existing schemes.

延伸閱讀