The main motivation of this project is present an approach to obtain an estimate of data hiding in different transforms for compressed still images. Though most methods reported in literature use DCT and wavelet transform decomposition for data embedding, the choice of transform is not always obvious. We compare the achievable data hiding capabilities for different decompositions like DCT, DFT, Hadamard and Hartley – sub-band transforms. By means of implementation and optimization of the said algorithms we show that the magnitude of DFT decomposition performs best among the ones compared. The images are first subjected to one of the transforms; data is added after forming frequency bands, namely low frequency, mid frequency and high frequency bands .The low frequency bands are very noisy due to the high energy content of the image, whereas, high frequency components are very vulnerable to processing, as most compressors would discard them at low bit rates. We show the addition of message signal in a suitable transform domain rather than spatial domain can significantly increase the channel capacity. Our focus is on the mathematical models, fundamental principles, and code design techniques that are applicable to data hiding. Such codes are also called watermarking codes; they can be used in a variety of applications, including copyright protection for digital media, content authentication, media forensics, data binding, and covert communications.
3.4. System Specifications
The system specifications have been categorized into hardware and software specifications.
This includes the specifications of the computer for the study
1. Intel Pentium III processor.
2. 850 MHz.
3. 512 MB of RAM.
1. Image format: JPEG(.jpg)
2. Types of images: Monochrome of size 256 × 256
3. Method of Image Encryption : Transform Domain Encoding
4. Transforms used in the encoding algorithms:
a. Discrete Cosine Transform (DCT)
b. Discrete Fourier Transform (DFT)
c. Hadamard Transform
d. Hartley Transform