Research On High Capacity Data Hiding Methods Of Digital Image

Internet is used as a leading object for simple and effective communication because of advancements technology and science.Users access the internet and maintain their information by the help of steganography and watermarking.It is of great practical and theoretical value to realize high-capacity information hiding.Such as that is conducive to transforming information and protecting intellectual property.Digital image is one of the commonly used multimedia carriers.After more than twenty years' study,researchers have made a great progress in steganalysis,however,the research of multi-style secret information and the contradiction between high capacity and high quality need to be further resolved.To realize high capacity data hiding in digital image,high capacity reversible data hiding in encrypted image and high capacity and robust watermark,the main works are as follows.1.Research on high capacity reversible data hiding in encrypted images(RDHEI)by bit stream preprocessing.Based on the analysis of the encrypted images usually lose redundancy and the receiver cannot realize the image reconstruction completely,a method of high capacity RDHEI by bit stream preprocessing is proposed.The owner of the cover image retains the redundant space of the original image through using Arnold transform and Sudoku transform.In the embedding phase,we propose preprocessing the information by DES algorithm,and the binary bit strings are compressed by a factor of 1.5.The compressed information is embedded into the encrypted image by most significant bit(MSB)algorithm.The receiver extracts the information by xor operator,and reconstructs the original image with no error.The experimental results prove that the proposed method not only realizes RDHEI,but can also completely reconstruct the cover image.Our study is a widely used research on RDHEI.2.Research on high capacity RDHEI based on scanned document.Multimedia information has redundancy,which affects high capacity data hiding.By analyzing the scanned document and the above algorithm,a method of high capacity RDHEI based on scanned document is proposed.The cover image is encrypted by the owner.Before the data hiding phase,the scanned document is preprocessed by half-tone,quad tree and DES algorithm,and the bit stream is preprocessed by DES algorithm.The compressed information is embedded into the encrypted image by MSB algorithm.The receiver extracts the information by xor operator,and reconstructs the original image with error.For the scanned document,the content of the document is recovered according to the extracted secret information.The experimental results prove that the proposed method not only realizes high-capacity reversible data hiding in encrypted images,but also reconstructs the original image completely.3.Research on a LSB matching algorithm based on information preprocessing.Based on the analysis of the fidelity of the LSB matching algorithm decreases with the increase of the embedding capacity,a new mapping function is designed and a method of the LSB matching algorithm based on information preprocessing is proposed.The scanned documents are processed by half-tone and quad tree.The golden section ration is introduced,and more information is embedded into the least significant bit with small modify.The preprocessed scanned documents are reduced by an average of 8.11 times.In the hiding stage,a mapping function based on the golden ratio is introduced to realize the embedding of secret information.According to the mapping function,the secret information is quickly extracted by the receiver.The experimental results prove that the proposed method not only reduces image distortion,but also increases the embedding capacity.4.Research on a reversible data hiding method for nonlinear substitution of secret information.Based on the analysis of the difference expansion algorithm with good invisibility and not high capacity data hiding,a reversible data hiding method for nonlinear substitution of secret information is proposed.The scanned document is processed by half-tone and quad tree,and the cover image is divided into no overlap sub-blocks sized 3×3.Pixels in even rows and even columns are initialized as embeddable pixels and others are reference pixels.The interpolated values and directional interpolation values are calculated.The differences of the interpolated pixels with the embeddable pixels and the reference pixels and threshold are compared respectively,to realize the data hiding or the adjustment of the histogram.The proposed method replaces the secret information non-linearly,so the embedding capacity of secret information is greatly increased,and the image quality is well guaranteed.The experimental results prove that this method has a good promotion prospect in the communication field of high-capacity secret information.5.Research on high-capacity and robust image watermarking algorithm.In this paper,we propose a high-capacity and robust method based on discrete wavelet transforms(DWT)and optimal discrete cosine transforms(DCT).We present two approaches,they are:DWT-ODCT(high-capacity and robust image watermarking algorithm based on DWT and optimal DCT),and P-DWT-ODCT(high-capacity and robust image watermarking algorithm based on DWT and optimal DCT by watermark preprocessing).The watermark image is preprocessed by halftone and quad tree techniques,and the position information of the content is extracted as the actual embedded value.The cover image is processed by DWT,and then the high frequency sub-band is divided into blocks,sized by 8×8.The blocks were processed by an optimal DCT psychovisual threshold.Embedding process in this paper utilizes certain frequency regions of DCT so the insertion of watermark bits will cause the least image distortion.The information was extracted by comparing certain coefficients based on some rules.According to the value of information,the watermark image can be reconstructed.The experimental results prove that our algorithm achieves higher capacity and robustness than the existing schemes.The watermark extraction produces high image quality after cutting,noise,filter and JPEG compression attacks.The above solutions to the embed capacity and image quality problems in the data hiding of digital image are proposed in this paper.These solutions can realize high-capacity and high-quality image information hiding under the two commonly used secret information forms of data stream and scanned documents.Analysis and experimental results showed the effectiveness of the proposed scheme.