Research On Mapping Coverless Image Information Hiding Technology

In the Internet age,cyberspace security is very important.As an important branch of Cyberspace Security Research,information hiding has attracted much attention.Information hiding refers to the technology of hiding secret information in public media information,which makes it difficult for people to detect its existence with intuitive vision and hearing.It plays an important role in military communications,intelligence agencies,trade secret transmission and other confidential communication occasions,as well as copyright protection,anti-counterfeiting authentication,image forensics and so on.However,when embedding secret information,traditional information hiding needs to modify the carrier,as long as the modification will leave traces,so it is difficult to resist steganalysis,so the idea of coverless information hiding is proposed.Coverless information hiding is not that there is no carrier,but that there is no need to modify the carrier when embedding a secret.Coverless information hiding technology has become a research hotspot with strong anti analysis ability.At present,there are mainly some problems,such as small hiding capacity,weak anti-interference ability and high complexity of image database construction.In order to improve the performance of coverless information hiding,this thesis first propose a low-frequency coefficient coding scheme based on discrete cosine transform(DCT)with higher coding coverage.Then two algorithms are proposed to apply the coding scheme to search coverless information hiding and texture synthesis coverless information hiding,and the prototype system is designed.The specific work is as follows:(1)A search mapping coverless information hiding method based on Arnold scrambling and DCT coding is proposed.This method first scrambles the picture with Arnold parameters a,b and N,then divides the picture into 8 * 8 small blocks for DCT transformation,and then encodes the low-frequency coefficients after DCT transformation.Then replace the values of scrambling parameters a,b and N,continue the above operation,and build the index table corresponding to different parameters and coded values.When hiding the secret information,select the parameters corresponding to the same coding value as the secret information in the index table to build the candidate queue,then select the robust parameters in the candidate queue as the key,and finally send the key and picture to the receiver.In the experiment,this method is compared with the existing similar methods.The results show that this method improves the hiding capacity,has stronger anti JPEG compression ability,simple algorithm and strong application value.(2)A texture synthesis coverless information hiding method based on DCT coding is proposed.Firstly,a small source texture image is selected,and then the source texture image is divided into several candidate blocks.The candidate block is divided into core region and boundary region.The core region is transformed by DCT first,and then encoded to build the index table.When embedding secret information,select the corresponding small block in the index table according to the secret information and put it in the candidate queue.Then,in the candidate queue,the small block with the largest and smallest absolute value of DCT coefficient is selected as the best candidate block,that is,robustness screening.Then the block is pasted to the specified position according to the key,and finally the remaining part is synthesized by texture synthesis.The experimental results show that the dense image generated by this method not only has good visual effect,but also improves the hiding capacity and anti-interference ability,and has a good effect in practical application.(3)According to the proposed search mapping coverless information hiding algorithm based on Arnold scrambling and DCT coding and texture synthesis coverless information hiding algorithm based on DCT coding,the prototype systems are designed respectively.The system can not only hide and extract information,but also be easy to use.