SOME INVESTIGATIONS ON STEGANOGRAPHY TECHNIQUES

Book excerpt: Steganography is the art of hiding and transmitting data through apparently innocuous carriers in an effort to conceal the existence of the secret data. The Least Significant Bit (LSB) steganography that replaces the least significant bits of the host medium is a widely used technique with low computational complexity and high insertion capacity. Although it has good perceptual transparency, it is vulnerable to steganalysis which is based on statistical analysis. Many other steganography algorithms have been developed such as Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT) and Spread Spectrum Embedding. But the insertion capacities for all the above methods were not satisfied. Therefore, developing new steganography algorithms against statistical analysis seems to be the prime requirement in steganography. The LSB insertion method is the most common and easiest method for embedding messages in an image with high capacity. However, it is detectable by statistical analysis such as RS and Chisquare analysis. Hence, researchers are still in look out for steganography techniques with enhanced insertion capacity of secret data along with greater security and which can resist attacks. In this work, in order to enhance the embedding capacity of secret data four techniques for secret communication have been proposed. They are classified into two categories. In first category, cryptography is used along with steganography to enhance the security, while in second category only steganography is used. In the first category, two improved LSB substitution techniques have been proposed. The first technique is known as Zigzag Modulo Substitution Method in which embedding locations are Sequence based. The second technique is known as Random Modulo Substitution Method using Random Technique in LSB Steganography and user key based LSB substitution steganography for RGB images where in, RSA algorithm is used for encryption. The techniques under the first category are exclusively LSB array based. The first LSB array based technique embeds message bits into LSB arrays of cover image by using zigzag scanning. On the other hand the Random Modulo Substitution Method embeds secret data into the different locations of cover image by using pseudo random index generator. Moreover, both these LSB array based techniques use RSA algorithm to enhance security. Histogram and Statistical analysis performed on the stego image proved that the proposed techniques can effectively resist steganalysis. Comparison of the statistical parameters like Root Mean Square (RMS), Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Matrix (SSIM) for the proposed techniques with cover image and stego image was carried out and analyzed. The Second category includes pixel value modification method and pixel value differencing method in which the embedding decision for a target pixel is taken by random technique. Data hiding by using pixel value modification with modulus function in color images guarantees that no pixel value will exceed the range 0 to 255 in stego image. In the existing PVD embedding methods, only one secret digit was embedded for two consecutive pixels, but the proposed method embeds one secret digit in only one pixel. Proposed method on color images gives more capacity and security than the PVD methods. It also provides better visual quality of stego image. Moreover, proposed method extracts the hidden secret message efficiently without using the range tables. In existing steganography algorithms like Pixel Value Differencing (PVD) methods, the secret data are embedded into the differences of adjacent pixels. This pair wise modification mechanism in cover image increases the histogram distortion.