Research On Image Steganography Algorithm Resisting Scaling Attack And Statistical Detection

The digital image steganography embeds the secret message into the cover image by a specific algorithm to achieve covert communication.Most of the current mainstream adaptive steganography algorithms embed secret message into the image texture complex area by combining the distortion function design and STCs coding to enhance its anti-detection.However,the images during the actual transmission,restricted by the bandwidth,software processing capacity,often suffer from compression,scaling and other processing operations.These operations will not only cause a decrease in image quality and loss of information,but will also result in the embedded secret message being unable to extract correctly.Therefore,it is imperative to study the image steganography algorithm against compression and scaling attacks on the basis of ensuring the resistance to detection.In this paper,three kinds of image steganography algorithms resisting scaling attack and statistic detection simultaneously are presented to resisting the scaling process in image transmission.The main work of the paper is as follows:1.To solve the problem that the existing adaptive steganography algorithm could not resist scaling attacks,a steganography method based on scaling invariant pixels resisting the scaling attack and statistical detection is proposed.This method is applied to the nearest-nearest interpolation method in image scaling processing.First,the principle of nearest-neighbor interpolation method is analyzed and summarized,which is using the resized pixel coordinates to find the original adjacent pixels,and obtaining the weights of the adjacent original pixels by the distance between resized pixel and the original adjacent pixels,finally calculating the resized pixel value by the linear combination of the original pixels and the weights and extracting the scaling invariant pixels.Then combine WOW,The S-UNIWARD and MiPOD distortion functions and STCs codes to achieve the minimum distortion embedding,which achieves the aim of anti-scaling attack and anti-statistic detection.Two kinds of typical characteristics of SPAM and maxSRM are used to carry out detection experiments.The results show that the method proposed has good antiscaling attack performance and anti-detection performance under various scaling factors and embedding rates.2.In view of the fact that the steganography method based on scaling invariant pixels is only robust to the nearest-neighbor interpolation scaling attack,an image adaptive steganography algorithm based on quantization index modulation is proposed for anti-scaling attack and antistatistical detection.For the spatial image,this paper uses the algorithm based on quantization index modulation to extract the embedded cover.Then construct the embedding distortion function of the new embedded cover using S-UNIWARD steganography,and uses the minimum distortion coding to realize the embedding of the secret messages.And finally,according to the embedding modification amplitude of secret messages in the new embedded cover,the quantitative index modulation algorithm is applied to realize the final embedding of secret messages in the original embedded cover.The experimental results show that the algorithm proposed is robust to the nearest neighbor interpolation,the bilinear interpolation and the bicubic interpolation attacks and average correct extraction rate of embedded messages is higher than the classical steganography algorithm S-UNIWARD.Also the algorithm has higher detection resistance than the original embedding algorithm based on quantization index modulation.3.In order to solve the problem that the steganography algorithm based on quantization index modulation has the low correct extraction rate of secret message and the lack of anti-detection,an adaptive steganography algorithm based on Zernike moment is proposed.For the spatial image,firstly,the amplitude of Zernike moment of normalized cover image is extracted,and the embedded cover of secret message is obtained using dither modulation algorithm.Then the embedded distortion of new embedded cover is designed by spatial domain steganography SUNIWARD,and the embedded amplitude of secret message is realized by minimizing distortion coding.Finally,according to the modified amplitude of secret message in new embedded cover,the embedded algorithm based on dither modulation is applied again to realize the final embedding of secret messages in the amplitude of Zernike moment,and the stego image is reconstructed.The experimental results show that the proposed algorithm can resist the scaling attack,and its antidetection is much higher than the steganography algorithm based on quantization index modulation.At last,the paper summarizes the whole work and points out the shortcomings of this work and the prospect of the next step.