A new image data-hiding technique is proposed. The proposed approach modifies blocks of the image after projecting them onto certain directions. By quantizing the projected blocks to even and odd values, one can represent the hidden information properly. The proposed algorithm performs the modification progressively to ensure successful data extraction without the need for the original image at the receiver side. Two techniques are also presented for correcting scaling and rotation attacks. The first approach is an exhaustive search in nature, which is based on a training sequence that is inserted as part of the hidden information. The second approach uses wavelet maxima as image semantics for rotation and scaling estimation. Both algorithms have proved to be effective in correcting rotation and scaling distortion.

Progressive quantized projection approach to data hiding

The development of new technologies and communication networks creates new needs and stimulates the introduction of new functionalities particularly in high speed internet data paths. The drawbacks of traditional Discrete Wavelet Transform (DWT) in image compression were suitably estimated. This paper investigates the importance of new technique to compress X-Ray medical images and satellite extracted images. It also verified for other types of images also. The proposed method out performs the existing methods and the simulation results of the new proposed algorithm shows better performance. The values of PSNR and Compression ratios clearly show the robustness of the algorithm in compression of the images. The simulation results performed on MATLAB.

A robust approach for qualitative compression in JPEG 2000 standard

Digital watermarking has gained a lot of importance for digital media recently. It deals with hiding secret bits of information with a digital content as a cover. This is very useful for many applications like copyright control, authenticity of documents, captioning, broadcast monitoring etc. Various techniques for watermarking have been proposed by various researchers in the recent past. Many of these techniques have problems related to robustness, fidelity and payload capacity of watermarks. These techniques also exhibit the problems of trade-off between robustness and fidelity. Techniques based on Backpropagation neural network, Full counterpropagation neural network and Adaptive resonance theory(ART) have been developed and tested by the author for watermarking . The results have shown positive results in increasing robustness, fidelity and payload capacity of watermarks. The brief discussion of various techniques with their results are presented this paper which forms the basis of comparison between the these techniques with some of the conventional techniques whose results have also been shown.

Performance comparison of watermarking techniques of various domains

: Digital watermarking precisely means inserting secret bits of information in a digital cover This is analogous to the theory of a communication channel where the cover media may be regarded as the communication channel and the noise may be modeled as the digital watermark to be inserted Many research works aiming at optimizing robustness and fidelity of watermarking are cited in the literature Among such techniques, neural network ART based technique can be employed to create and reproduce watermark with reasonably high values of robustness and fidelity

https://www.jetir.org/papers/JETIR1802114.pdf

Digital Watermarking modelled as noise in communication channel and optimization of desirable characteristics using Adaptive Resonance Theory (ART)

The watermarking technique has been proposed as a method of hiding secret information in an image to protect the copyright of multimedia data. However, most previous work have focused on algorithms for embedding 1D watermarks or 2D binary digital watermarks. In this paper, a multiresolution-based method for embedding a gray-level digital watermark into an image is proposed. By taking advantage of multiresolution signal decomposition, both the watermark and the original image are decomposed into multiresolution representations with different structures, and then the decomposed watermarks of different resolutions are embedded into the corresponding resolution of the decomposed original image. For embedding conveniently, the gray-level digital watermark is decomposed into a series of binary digital images by means of a still-image decomposition technique. By embedding the watermark into the relationship between different layers within the transformed original image, the algorithms can extract the watermark without the original image. Experimental results show that the proposed technique results in an almost invisible difference between the watermarked image and the original image, and is robust to common image processing operations and JPEG lossy compression.

Sheng-He Sun

Papers

Multiresolution digital watermarking for still image