This paper proposes lossless, reversible, and combined data hiding schemes for ciphertext images encrypted by public-key cryptosystems with probabilistic homomorphic properties. In the lossless scheme, pixels are replaced new values to embed additional into several least significant bit planes of multilayer wet coding. Then, embedded can be directly extracted from domain, data-embedding operation does not affect decryption original plaintext image. reversible a preprocessing is employed...

Posting images on social network platforms is happening everywhere and every single second. Thus, the communication channels offered by various networks have a great potential for covert communication. However, transmitted through such will usually be JPEG compressed, which fails most of existing steganographic schemes. In this paper, we propose novel image steganography framework that robust channels. particular, first obtain channel compressed version (i.e., output) original image. Secret...

This letter proposes a batch steganographic method, which combines cover-selection and payload-allocation by distortion optimization. We first proved that with the value of payload increasing, first-order derivative single cover is monotonically increasing. Then, we deduced covers selected from given set should be equal if total corresponding minimal. Finally, an algorithm was designed to combine payload-allocation, so optimal stego-covers can obtained. Experiment results show...

This paper proposes a framework to improve the existing distortion functions designed for JPEG steganography, which results in better capability of countering steganalysis. Different from steganography approach that minimizes image distortion, we minimize feature caused by data embedding. Given image, construct reference close before compression. Guided both and minimization, state-of-the-art syndrome trellis coding embedding are improved distinguishing costs +1 versus -1 has three...

In the past decade, diversification of steganographic techniques has posed significant threats to information security, necessitating effective countermeasures. Current defenses, mainly reliant on steganalysis, struggle with detection accuracy. While “image vaccines” have been proposed, they often target specific methodologies. This paper introduces a universal vaccine enhance steganalysis Our symmetric approach integrates existing methods protect images before online dissemination using...

This paper investigates on the defense against steganography, and overall purpose of study is to design a satisfactory scheme in encrypted domain. Image vaccine steganography an effective technique discover utilization with extremely high detection accuracy. However, image owner provider are not same person usually. To meet requirements privacy protection simultaneously, this proposes for images. After encrypting entire data original using stream cipher, can be injected into without knowing...

A hybrid distortion function for JPEG steganography exploiting block fluctuation and quantization steps is proposed. To resist multidomain steganalysis, both spatial domain discrete cosine transformation (DCT) are involved in the proposed function. In domain, a value allotted each 8×8 according to fluctuation. DCT employed allot values coefficients block. The two elements, steps, combined together measure embedding risk. By employing syndrome trellis coding embed secret data, changes...

In recent decades, a large number of perceptual image hashing schemes have been designed to secure the authenticity and integrity digital images. However, feasible criterion evaluate performances has not developed yet. To this end, unified performance evaluation method for is proposed in paper. The contains six modules: robustness, discrimination, tampering detection, security, computational efficiency hash length. order relationship analysis (ORA) employed assign score proportion each...

Batch steganography is a technique that hides information into multiple covers. To achieve better performance on the security of data hiding, we propose novel strategy batch using generative network. In this method, approaches cover selection, payload allocation, and distortion evaluation are considered in round. We define quality metric to evaluate between image stego. When training generation function, an objective function containing two parts: entropy loss steganalytic loss. While used...

Content similarity is a representative property of natural images, for example, similar regions, which utilized by modern steganalysis. Existing JPEG steganographic methods mainly focus on the complexity content but ignore similarity. This article investigates to improve undetectability steganography. Specifically, DCT blocks and 64 parallel channels used design distortion function. Given image, initial embedding costs are assigned quantized coefficients using an appropriate algorithm among...

The advancement of internet has greatly facilitated development healthcare technology, particularly in the field remote diagnosis. However, during transmission medical images and electronic records through internet, there are risks such as theft, tampering, compression, which may result leakage or damage patients' information. A robust zero-watermarking scheme using DTCWT improved differential entropy is proposed to protect copyright security images. Firstly, low-frequency information image...

Existing cover selection methods for steganography cannot resist pooled steganalysis. This paper proposes a secure method which is able to steganalysis and single object meanwhile. To steganalysis, the maximum mean discrepancy (MMD) distance between stego set clear arbitrary image kept not larger than normal threshold during selection, where MMD two sets. Under this constraint, searching strategy designed select minimal steganographic distortion images within affordable computational...

This letter focus on a practical scenario of cover selection for steganography, in which part the available images sender have been processed to improve visual quality, e.g., contrast enhancement, image denoising. In this case, not only embedding distortion (caused by steganography), but also processing processing) should be considered when selecting image. We propose method combine two kinds together measure suitability steganography images. To calculate distortion, classifier is trained...

Existing cover selection methods for steganography mainly focus on embedding distortion of each image, but ignore the similarity between images. When images are similar, a number relevant samples provided to steganalysis, which is disadvantageous steganography. This paper proposes new method joint image and distortion. Due difference other processing tasks, e.g., reconstruction, recognition, we propose customized calculate based SVD (singular value decomposition). Importantly, small singular...

As generative models find broader applications in Internet of Things (IoT) image processing tasks, safeguarding the copyright these assumes increasing significance. Embedding watermarks on output images generated by such has been proposed some researchers as a means protecting intellectual property. However, prevailing methods for generating model inadvertently introduce significant high-frequency artifacts regions, compromising imperceptibility and security watermarking system. In pursuit...