Steganography based on generative adversarial networks(GANs)has become a hot topic among researchers.Due to GANs being unsuitable for text fields with discrete characteristics,researchers have proposed GANbased stegan...Steganography based on generative adversarial networks(GANs)has become a hot topic among researchers.Due to GANs being unsuitable for text fields with discrete characteristics,researchers have proposed GANbased steganography methods that are less dependent on text.In this paper,we propose a new method of generative lyrics steganography based on GANs,called GAN-GLS.The proposed method uses the GAN model and the largescale lyrics corpus to construct and train a lyrics generator.In this method,the GAN uses a previously generated line of a lyric as the input sentence in order to generate the next line of the lyric.Using a strategy based on the penalty mechanism in training,the GAN model generates non-repetitive and diverse lyrics.The secret information is then processed according to the data characteristics of the generated lyrics in order to hide information.Unlike other text generation-based linguistic steganographic methods,our method changes the way that multiple generated candidate items are selected as the candidate groups in order to encode the conditional probability distribution.The experimental results demonstrate that our method can generate highquality lyrics as stego-texts.Moreover,compared with other similar methods,the proposed method achieves good performance in terms of imperceptibility,embedding rate,effectiveness,extraction success rate and security.展开更多
In the field of information hiding,text is less redundant,which leads to less space to hide information and challenging work for researchers.Based on the Markov chain model,this paper proposes an improved evaluation i...In the field of information hiding,text is less redundant,which leads to less space to hide information and challenging work for researchers.Based on the Markov chain model,this paper proposes an improved evaluation index and onebit embedding coverless text steganography method.In the steganography process,this method did not simply take the transition probability as the optimization basis of the steganography model,but combined it with the sentence length in the corresponding nodes in the model to gauge sentence quality.Based on this,only two optimal conjunctions of the current words are retained in the method to generate sentences of higher quality.Because the size of the training text dataset is generally large,this leads to higher complexity of the steganographic model;hence,fewer repetitions of the generated steganographic sentences occur.Different datasets and methods were selected to test the quality of the model.The results indicate that our method can achieve higher hiding capacity and has better concealment capability.展开更多
Utilizing OnLine Short Text (OLST) in social networking tools such as microblogs, instant messag- ing platforms, and short message service via smart phones has become a routine in daily life. OLST is ap- pealing for...Utilizing OnLine Short Text (OLST) in social networking tools such as microblogs, instant messag- ing platforms, and short message service via smart phones has become a routine in daily life. OLST is ap- pealing for personal covert communication because it can hide information in a very short carrier text, and this concealment is hard to detect due to the diversity of normal traffic. However, designing appropriate schemes confronts several challenges: they need to be provably secure, and their performance needs to maintain high efficiency and handy usability due to the short length of OLST messages. In this paper, we propose a family of customized schemes known as HiMix, HiCod, HiOpt, and HiPhs for text steganography in OLST. These schemes are evaluated in terms of their security and their performance with regard to two metrics that address the particular characteristics of OLST: hiding rate and hiding ease. All proposed schemes are proved to be at least computationally secure, and their performance in terms of hiding rate and hiding ease justifies their applicability in social networking tools that utilize OLST.展开更多
基金This work was supported in part by the National Natural Science Foundation of China under Grant 61872134,61672222,author Y.L.Liu,http://www.nsfc.gov.cn/in part by Science and Technology Development Center of the Ministry of Education under Grant 2019J01020,author Y.L.Liu,http://www.moe.gov.cn/+1 种基金in part by Science and Technology Project of Transport Department of Hunan Province under Grant 201935,author Y.L.Liu,http://jtt.hunan.gov.cn/Science and Technology Program of Changsha City under Grant kh200519,kq2004021,author Y.L.Liu,http://kjj.changsha.gov.cn/.
文摘Steganography based on generative adversarial networks(GANs)has become a hot topic among researchers.Due to GANs being unsuitable for text fields with discrete characteristics,researchers have proposed GANbased steganography methods that are less dependent on text.In this paper,we propose a new method of generative lyrics steganography based on GANs,called GAN-GLS.The proposed method uses the GAN model and the largescale lyrics corpus to construct and train a lyrics generator.In this method,the GAN uses a previously generated line of a lyric as the input sentence in order to generate the next line of the lyric.Using a strategy based on the penalty mechanism in training,the GAN model generates non-repetitive and diverse lyrics.The secret information is then processed according to the data characteristics of the generated lyrics in order to hide information.Unlike other text generation-based linguistic steganographic methods,our method changes the way that multiple generated candidate items are selected as the candidate groups in order to encode the conditional probability distribution.The experimental results demonstrate that our method can generate highquality lyrics as stego-texts.Moreover,compared with other similar methods,the proposed method achieves good performance in terms of imperceptibility,embedding rate,effectiveness,extraction success rate and security.
基金This work is supported under the National Key Research and Development Program of China(2018YFB1003205)in part under the 2019 Longyuan Youth Innovation and Entrepreneurship Talents(Team)Project(GanZuTongZi[2019]No.39)(No.23).
文摘In the field of information hiding,text is less redundant,which leads to less space to hide information and challenging work for researchers.Based on the Markov chain model,this paper proposes an improved evaluation index and onebit embedding coverless text steganography method.In the steganography process,this method did not simply take the transition probability as the optimization basis of the steganography model,but combined it with the sentence length in the corresponding nodes in the model to gauge sentence quality.Based on this,only two optimal conjunctions of the current words are retained in the method to generate sentences of higher quality.Because the size of the training text dataset is generally large,this leads to higher complexity of the steganographic model;hence,fewer repetitions of the generated steganographic sentences occur.Different datasets and methods were selected to test the quality of the model.The results indicate that our method can achieve higher hiding capacity and has better concealment capability.
基金Supported by the Open Research Fund from the Shandong Provincial Key Laboratory of Computer Networks (No. SDKLCN-201101)the Special Fund for Basic Scientific Research of Central Colleges, China University of Geosciences (Wuhan) (No. 110109)the National Natural Science Foundation of China (No. 61170217)
文摘Utilizing OnLine Short Text (OLST) in social networking tools such as microblogs, instant messag- ing platforms, and short message service via smart phones has become a routine in daily life. OLST is ap- pealing for personal covert communication because it can hide information in a very short carrier text, and this concealment is hard to detect due to the diversity of normal traffic. However, designing appropriate schemes confronts several challenges: they need to be provably secure, and their performance needs to maintain high efficiency and handy usability due to the short length of OLST messages. In this paper, we propose a family of customized schemes known as HiMix, HiCod, HiOpt, and HiPhs for text steganography in OLST. These schemes are evaluated in terms of their security and their performance with regard to two metrics that address the particular characteristics of OLST: hiding rate and hiding ease. All proposed schemes are proved to be at least computationally secure, and their performance in terms of hiding rate and hiding ease justifies their applicability in social networking tools that utilize OLST.