05-05-2011, 04:43 PM
abstract
Article history:Received 1 February 2007Received in revised form 14 November 2008Accepted 8 December 2008Available online 24 December 2008Keywords
ata hidingImage compressionIndex codingSearch-order-coding algorithmVector quantizationThe VQ-based data hiding technique has not received much attention compared to various spatial domainbaseddata hiding techniques in digital images. Consequently, a new data hiding scheme, applied in the VQcompresseddomain of cover images, is introduced in this article. To provide more hiding capacity for secretdata and to keep an acceptable bit rate for the compressed cover images, the search-order-coding (SOC)algorithm was implemented to compress the VQ indices of the cover images in the process of data hiding.During the process of data hiding, the proposed scheme embeds secret data into the compressed VQ indicesof the cover image adaptively, adjusting the bit rate according to the size of the secret data and thecompressed cover image. In addition, the hiding process induces no extra coding distortion. Experimentsshow that the receiver can efficiently receive both the secret data and the compressed cover imagesimultaneously with an acceptable bit rate. Simulation results also demonstrate that our proposed schemeoutperforms earlier proposed methods.© 2008 Elsevier B.V. All rights reserved.
1. Introduction
Along with recent progress in the areas of computer hardware andsoftware, the internet has become the most popular channel fortransmitting various forms of digital media. Since the environment ofthe internet is open, the protection of digital information transmittedon the network has become an important research topic in recentyears. Data hiding is a common technique for data protection. Itinvolves secretly embedding significant data into various forms ofdigital media such as text, audio, image and video [1–3]. With therapid growth of network communication, data hiding techniques havebeen widely utilized in the applications of copyright protection,fingerprinting and secret communication [4–7].The purpose of data hiding techniques is different from traditionalcryptography [8,9] andwatermarking techniques [10–12]. Cryptographyencrypts meaningful messages into meaningless data, while watermarkingtechniques are utilized to protect the intellectual copyright ofdigital media. The data hiding technique hides the presence of secretinformation using the cover media as camouflage, and is considered anextension of traditional cryptography. Hiding data in an image involvesembedding a large amount of secret data into a cover image withminimal perceptible degradation of the image quality. However, there isalways a tradeoff between the hiding capacity for secret data and thedistortion of the cover image since more hidden data always result inmore degradation of visual quality in the cover image. Moreover, whenthe data hiding technique is implemented on compressed images, thehiding capacity and the visual quality of the cover images are even morerestricted. Furthermore, the bit rate of the compressed cover imagebecomes another major consideration in data hiding applications.Hiding secret data should not cause an apparent increase in the bitrate of the compressed cover images.In the past few years, several vector quantization (VQ)-baseddata hiding techniques have been proposed in the literature. Du et al.proposed a linguistic data hiding scheme to adaptively hide secretdata into VQ-compressed cover images according to the size ofsecret data [5]. In Du et al.'s scheme, the secret data were hidden inthe compressed cover images using the following two phases. In thefirst phase, all the codewords in the codebook were rearranged intoexclusive groups according to their similarity. In the second phase,during the VQ encoding procedure, the best matched codeword ofthe block being encoded was replaced by another codeword fromthesame group, chosen based on the secret bits to be hidden. Moredetails of this scheme can be found in [5]. To improve theperformance of [5], Shie et al. proposed another VQ-based datahiding technique that takes advantages of the prediction property ofthe side-match VQ (SMVQ) state codebook [6]. The major idea of thistechnique is to hide secret data in VQ-compressed codes of coverimages based on a modified SMVQ encoding process, takingadvantage of the concept of prediction. This technique providesbetter visual quality for the cover images and has lower computationalcomplexity than that of [5]. Chang et al. proposed a new datahiding scheme in which a limited amount of binary data can behidden in the VQ-compressed codes of the cover image [7]. Theauthors claimed that their scheme is the first one that technicallyand directly hides secret data in the VQ-compressed
Download full report
http://www.cs.pu.edu.tw/~ychu/class982/D...images.pdf
Article history:Received 1 February 2007Received in revised form 14 November 2008Accepted 8 December 2008Available online 24 December 2008Keywords
ata hidingImage compressionIndex codingSearch-order-coding algorithmVector quantizationThe VQ-based data hiding technique has not received much attention compared to various spatial domainbaseddata hiding techniques in digital images. Consequently, a new data hiding scheme, applied in the VQcompresseddomain of cover images, is introduced in this article. To provide more hiding capacity for secretdata and to keep an acceptable bit rate for the compressed cover images, the search-order-coding (SOC)algorithm was implemented to compress the VQ indices of the cover images in the process of data hiding.During the process of data hiding, the proposed scheme embeds secret data into the compressed VQ indicesof the cover image adaptively, adjusting the bit rate according to the size of the secret data and thecompressed cover image. In addition, the hiding process induces no extra coding distortion. Experimentsshow that the receiver can efficiently receive both the secret data and the compressed cover imagesimultaneously with an acceptable bit rate. Simulation results also demonstrate that our proposed schemeoutperforms earlier proposed methods.© 2008 Elsevier B.V. All rights reserved.1. Introduction
Along with recent progress in the areas of computer hardware andsoftware, the internet has become the most popular channel fortransmitting various forms of digital media. Since the environment ofthe internet is open, the protection of digital information transmittedon the network has become an important research topic in recentyears. Data hiding is a common technique for data protection. Itinvolves secretly embedding significant data into various forms ofdigital media such as text, audio, image and video [1–3]. With therapid growth of network communication, data hiding techniques havebeen widely utilized in the applications of copyright protection,fingerprinting and secret communication [4–7].The purpose of data hiding techniques is different from traditionalcryptography [8,9] andwatermarking techniques [10–12]. Cryptographyencrypts meaningful messages into meaningless data, while watermarkingtechniques are utilized to protect the intellectual copyright ofdigital media. The data hiding technique hides the presence of secretinformation using the cover media as camouflage, and is considered anextension of traditional cryptography. Hiding data in an image involvesembedding a large amount of secret data into a cover image withminimal perceptible degradation of the image quality. However, there isalways a tradeoff between the hiding capacity for secret data and thedistortion of the cover image since more hidden data always result inmore degradation of visual quality in the cover image. Moreover, whenthe data hiding technique is implemented on compressed images, thehiding capacity and the visual quality of the cover images are even morerestricted. Furthermore, the bit rate of the compressed cover imagebecomes another major consideration in data hiding applications.Hiding secret data should not cause an apparent increase in the bitrate of the compressed cover images.In the past few years, several vector quantization (VQ)-baseddata hiding techniques have been proposed in the literature. Du et al.proposed a linguistic data hiding scheme to adaptively hide secretdata into VQ-compressed cover images according to the size ofsecret data [5]. In Du et al.'s scheme, the secret data were hidden inthe compressed cover images using the following two phases. In thefirst phase, all the codewords in the codebook were rearranged intoexclusive groups according to their similarity. In the second phase,during the VQ encoding procedure, the best matched codeword ofthe block being encoded was replaced by another codeword fromthesame group, chosen based on the secret bits to be hidden. Moredetails of this scheme can be found in [5]. To improve theperformance of [5], Shie et al. proposed another VQ-based datahiding technique that takes advantages of the prediction property ofthe side-match VQ (SMVQ) state codebook [6]. The major idea of thistechnique is to hide secret data in VQ-compressed codes of coverimages based on a modified SMVQ encoding process, takingadvantage of the concept of prediction. This technique providesbetter visual quality for the cover images and has lower computationalcomplexity than that of [5]. Chang et al. proposed a new datahiding scheme in which a limited amount of binary data can behidden in the VQ-compressed codes of the cover image [7]. Theauthors claimed that their scheme is the first one that technicallyand directly hides secret data in the VQ-compressed
Download full report
http://www.cs.pu.edu.tw/~ychu/class982/D...images.pdf