Image steganalysis based on attention augmented convolution

Huang, Siyuan; Zhang, Minqing; Yan, Ke; Bi, Xinliang; Kong, Yongjun

doi:10.1007/s11042-021-11862-4

Cited by 10 publications

(3 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, several other researchers proposed methods to combine the CNNs with other approaches [16], [17] that preprocess the images used in the binary classification to improve the results in steganalysis tasks. However, the proposed methods exhibit certain drawbacks regarding classification accuracy due to several problems, which include the lack of quality dataset and extensive training dataset, the inconsistency of the feature learning process, and the use of low payload by steganography practitioners, showing the need for further enhancement to effectively mitigate the risk of undetected secret communication, which could potentially have detrimental implications for companies, governmental institutions, and the community in general.…”

Section: Introductionmentioning

confidence: 99%

Enhancing Secret Data Detection Using Convolutional Neural Networks With Fuzzy Edge Detection

Croix,

Ahmad,

Han

2023

IEEE Access

View full text Add to dashboard Cite

Progress in Deep Learning (DL), particularly Convolutional Neural Networks (CNNs), has significantly improved the accuracy of steganographic image detection. However, the applications of CNNs have several challenges, mainly due to insufficient dataset quality and quantity, the heightened imperceptibility of low payload capacities, and suboptimal feature learning procedures. This paper proposes an enhanced secret data detection approach with a CNN architecture that includes convolutional, depth-wise, separable, pooling, and spatial dropout layers. An improved fuzzy Prewitt approach is employed for pre-processing the images prior to being fed into the CNN. Experimental results show a significant outperformance over the state-of-the-art methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Enhancing Secret Data Detection Using Convolutional Neural Networks With Fuzzy Edge Detection

Croix,

Ahmad,

Han

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…One of the main characteristics is in the case of stego-images must look as like the cover images, in addition to the fact that stego-images must be considered to approve steganalysis techniques such as chi-square, RS and PVo, among others [11]- [14]. A steganography system is a quintuple 𝛿 = 𝐶, 𝑀, 𝐾, 𝐷 𝑘 , 𝐸 𝑘 , where 𝐶 is the set of all cover images, 𝑀 is the messages to hide, 𝐾 is the set of secret keys, 𝐸 𝑘 ∶ 𝐶 × 𝑀 × 𝐾 → 𝐶, and 𝐷 𝑘 : 𝐶 × 𝐾 → 𝑀 [15] are two functions, the first is the embedding and the second is the extraction function, such that 𝐷 𝑘 (𝐸 𝑘 (𝑐, 𝑚, 𝑘) 𝑘 = 𝑚 [16].…”

Section: Introductionmentioning

confidence: 99%

A comparative study of steganography using watermarking and modifications pixels versus least significant bit

Caballero,

Muñoz,

Ramos-Corchado

2023

IJECE

View full text Add to dashboard Cite

This article presents a steganography proposal based on embedding data expressed in base 10 by directly replacing the pixel values from images red, green blue (RGB) with a novel compression technique based on watermarks. The method considers a manipulation of the object to be embedded through a data compression triple process via LZ77 and base 64, watermark from low-quality images, embedded via discrete wavelet transformation-singular value decomposition (DWT-SVD), message embedded by watermark is recovered with data loss calculated, the watermark image and lost data is compressed again using LZ77 and base 64 to generate the final message. The final message is embedded in portable network graphic (PNG) images taken from the Microsoft common objects in context (COCO), ImageNet and uncompressed color image database (UCID) datasets, through a filtering process pixel of the images, where the selected pixels expressed in base 10, and the final message data is embedded by replacing units’ position of each pixel. In experimentation results an average of 40 dB in peak signal noise to ratio (PSNR) and 0.98 in the similarity structural index metric (SSIM) evaluation were obtained, and evasion steganalysis rates of up to 93% for stego-images, the data embedded average is 3.2 bpp.

show abstract

“…Various means are used to enhance the detection accuracy of deep learning for steganography. These means include transfer learning [14,15], residual architecture [16,17], absolute values [18], channel selection [19], diverse activation [20], attention augmentation [21], feature-guided adaptation [22], etc. Deep learning is a promising framework providing state-of-the-art performance for steganalysis; however, it is generally difficult to obtain all the signals about steganographic algorithms from an adversary.…”

Section: Introductionmentioning

confidence: 99%

A Generative Learning Steganalysis Network against the Problem of Training-Images-Shortage

et al. 2022

View full text Add to dashboard Cite

In recent years, several steganalysis neural networks have been proposed and achieved satisfactory performances. However, these deep learning methods all encounter the problem of Training-Images-Shortage (TIS). In most cases, it is difficult for steganalyses to obtain enough signals about steganography from a game opponent. In order to solve the problem of TIS for steganalysis, we propose a novel steganalysis method based on generative learning and deep residual convolutional neural networks. Comparative experiments show that the proposed architecture can achieve promising performance in response to spatial domain steganalysis despite a lack of training images.

show abstract

Image steganalysis based on attention augmented convolution

Cited by 10 publications

References 33 publications

Enhancing Secret Data Detection Using Convolutional Neural Networks With Fuzzy Edge Detection

Enhancing Secret Data Detection Using Convolutional Neural Networks With Fuzzy Edge Detection

A comparative study of steganography using watermarking and modifications pixels versus least significant bit

A Generative Learning Steganalysis Network against the Problem of Training-Images-Shortage

Contact Info

Product

Resources

About