Iterative Learning Identification Method For Digital Image Hiding

As the rapid development of computer networks and digital communications technology,the digital multimedia products have been widely used. However, people can easily and quicklytransfer digital information. At the same time, the illegal theft and tampering with a variety ofmultimedia products at all times act in the event that these original creators of digital media orthe copyright owner’s interests are seriously infringed. Therefore, how the network environment,digital media, information security, intellectual property protection and other implementationissues become increasingly prominent. Information hiding as a new means of protectinginformation security, nowadays, it becomes the most important research issue of the network andinformation security. It also has broad application prospects in computer, communications,security and other areas.This paper focuses on a study of information hiding technology, combined with embeddedchaotic map and multiple hybrid algorithms, iterative learning identification proposed recoveryalgorithm. With a study of digital watermarking technology, digital watermark extractionproposed iterative learning identification algorithm. This major work and the results are asfollows:1. For the information hiding and digital watermarking field conditions made in detail.Discuss a general model of information hiding techniques, principles, characteristics,classification and performance evaluation. An overview of the principle of the digitalwatermarking technology, characteristics, classification, attack methods and performanceevaluation were given. In-depth discussion of the digital watermark embedding and extractionalgorithm, the spatial domain watermarking algorithm and transform domain watermarkingalgorithm.2. Discussed in detail the most commonly used two transform domain image hiding method, the DCT and DWT based image hiding method. Chaotic maps through the use of encryption tohide the image, the DCT and DWT based image hiding method can take advantage of humanvisual shielding characteristics, the invisibility of the hidden image. By a large number ofsimulation experiments, we verified the validity of these two methods in image hiding. Andsimulation results show that they have strong robustness.3. Focused on the general system of iterative learning identification method, and complexsystems Iterative Learning Identification Method. And they were used for the extraction ofhidden digital image recovery and digital image watermarking exraction.4. Study on an image hidden in a separate image into multiple hybrid method using imagewill be hidden image embedded into the carrier image the same time, the hidden image of achaotic encryption, iterative learning identification method to restore the hidden image.Simulation results show that the iterative learning identification algorithm is used to hideinformation recovery is effective. At the same time a large number of simulation experiments toverify the algorithm resist attack.5. Single image hiding technique of iterative learning identification method based onmultiple images hidden iterative learning identification method. Mixing parameters and pieces ofthe host image to hide an image the same time, the hidden image of the chaotic encryption, theiterative learning identification method to restore the hidden image. Simulation results show thatthe iterative learning identification algorithm used for the recovery of hidden information is validthrough extensive simulation to verify the anti-aggressive.6. The combination of chaotic maps and iterative learning identification method of digitalwatermarking, encryption and extraction. Use of multiple mixed-embedding algorithms to theoriginal watermark image is embedded into the carrier image, the iterative learning identificationmethod for complete extraction of the embedded watermark. Simulation results show that theiterative learning identification algorithm for digital watermark extraction is effective, and toverify the anti-aggressive.