Nabin Ghoshal

Masking based Data Hiding and Image Authentication Technique (MDHIAT)

In this paper a masking based data hiding and image authentication technique (MDHIAT) has been proposed by embedding a message/image into a colour image. Bits from authenticating message/image are embedded in single bit position under each byte of the source image by choosing a standard 3 times 3 mask in row major order. Point of insertion of a bit is obtained by computing modulus operation of the position of the image byte within the mask and an integer s whose value lies between 1 and 8. A message digest MD-5 has also been generated from authentication message/image and inserted into the source image in same manner to impart additional security. Experimental results show a good fidelity in embedded image when the value of s is less than equal to 6. Results are compared with the most popular steganographic algorithm S-Tools in terms of histogram, noise and, standard deviation analysis where proposed MDHIAT shows better performance in compared to S-Tools.

Image Authentication by Hiding Large Volume of Data and Secure Message Transmission Technique using Mask (IAHLVDSMTTM)

The paper presents an image authentication and secures message transmission technique by embedding message/image into color images. Authentication is done by embedding message/image by choosing image blocks of size 3 × 3 called mask from the source image in row major order. The position of insertion is chosen within the mask according to the formula k % s and (k + 1) % s + 1 where k and s are any number between 0 to 7 and 2 to 7 respectively. The dimension of authenticating image followed by MD-5 key and then the content of authenticating message/image are also embedded. This is followed by an XOR operation of the embedded image with another self generated MD-5 key obtained from the source image. The decoding is done by applying the reverse algorithm. The result has been tested with the aid of Histogram analysis, noise analysis and standard deviation computation of the source image with the embedded image and has been compared with popular existing steganographic algorithms like S-Tools where the proposed IAHLVDDSMTTM is capable to hide large volume of data than S-Tools and shows better performance.

A framework for block based image authentication (FBIA)

This study deals with devising and implementing a novel technique for image authentication and secured message transmission based on hiding large volume of data into gray scale images. Image authentication is done by embedding message / image in spatial domain by choosing image blocks of size 3 × 3 from the source image in row major order. Three bits of authenticating message /image/message-digest are fabricated within each source image byte of each image block where the position is chosen randomly using a hash function b7b6b5 ⨁k where b7, b6, b5 are higher order three bits of each source image byte and k is an integer number varies from 0 to 7, and the process is repeated for the whole image. The dimension of authenticating image followed by message digest (MD) and the content of authenticating message/image are also embedded. To enhance the security a layer has also been fabricated by XOR operation of the embedded image with a self generated MD obtained from the source image. Reverse process is applied for decoding. The results have been tested through histogram analysis, noise analysis and standard deviation computation of the source image with the embedded image and compared with popular existing steganographic algorithms like S-Tools. It is seen that the proposed FBIA is capable to embed more information than S-Tools with better fidelity.

Diffusion and Encryption of Digital Image Using Genetic Algorithm

The security level of any symmetric key algorithm is directly proportional to its execution time. The algorithm will be more secured if the number of iteration is increase. Soft computing methods are trying to reduce time complexity without compromise the security level. In this paper, an image has been encrypted by number of steps. At first, a key set has been developed based on 16 arbitrary characters and a large number. The image has been diffused in the next step. The key set and diffused image have been computed by genetic algorithm. Here, ring crossover and order changing mutation operator is used in genetic algorithm. The encrypted image is constructed based on logical operation between the diffused image and key. The effectiveness of the algorithm has been tested by number of statistical tests. Finally, a comparative study has been made between our proposed algorithm and some other algorithms. It has been observed that the proposed algorithm has given better result.

Steganography Based Visual Cryptography (SBVC)

A digital signature performs the function of conventional handwritten signatures for authentication of documents, data integrity and non-repudiation. Most of the conventional digital signature schemes are based on complex mathematical computations to generate the keys and verify signatures. In late nineteen century comparatively less computation based scheme known as visual cryptography (VC) is introduced. VC has high security and requires comparatively less complex computations. In this paper, we consider a new digital signature scheme, based on the concept of Visual Secret Sharing associated to XOR-based non-expansion visual cryptography systems. Our proposed scheme is towards enhancement of security of born digital and digitized document by generating visual shares and embedding them within a cover image using steganographic techniques rather than complicated mathematical computations. The proposed scheme could be applied for ownership protection, copy control, authentication of digital media etc.

Discrete Fourier Transform based multimedia colour image authentication for wireless communication (DFTMCIAWC)

This paper presents a novel steganographic schemes based on Discrete Fourier Transformation (DFT) and demonstrates the multimedia colour image authentication process in frequency domain for wireless communication(DFTMCIAWC). Authentication is done through embedding secrete message/image into the transformed frequency components of the source image at message originating node. The DFT is applied on sub-image block called mask of size 2 × 2 in row major order where authenticating message/image bit is fabricated within the real frequency component of each source image byte except the first frequency component of each mask. In order to retain the quantum value positive and non fractional in spatial domain, a delicate readjustment phase is used in the first frequency component of each mask as a post embedding handler. Robustness is achieved through embedding secrete message/image into both positive and negatives frequency component of source image and invisibility is satisfied in spatial domain using delicate re-adjust phase. Inverse DFT (IDFT) is performed after embedding to transform embedded image from frequency domain to spatial domain and the embedded image is transmitted across the network. At the destination node authentication is done through extraction process of embedded image. Experimental results demonstrate that the proposed algorithm performs better than discrete cosine transformation and quaternion Fourier transformation based schemes, and provide security and originality of data in wireless domain.

Two Stage Color Image Steganography Using DCT (TSCIS-DCT)

In frequency domain steganography, use of color images for secret data hiding may prove to be a decisive innovation. The proposed concept uses two color images for hiding a color/gray authenticating message/image. The mathematical technique of Discrete Cosine Transform (DCT) is applied on each block of size 2x2 taken in row major order from three color planes (Red, Green & Blue) sequentially and from two carrier images alternatively. A single secret message/image bit is fabricated within the transformed real frequency component of each source image byte except the first frequency component of each mask. The first frequency component of each block is used for re-adjustment to maintain the quantum value positive, non-fractional in spatial domain and also to reduce the integrated noise due to embedding. The pseudorandom position of embedding and subsequent extraction is generated by a logical expression. Experimental results of this technique reveal more efficiency compared to other similar technique.

Controlled Data Hiding Technique for Color Image Authentication in Frequency Domain (CDHTCIAFD)

This paper presents a new steganographic technique which demonstrates the colour image authentication process in frequency domain based on the Discrete Fourier Transformation (DFT). Image authentication is done by hiding secrete message/image into the transformed frequency component of source image. The DFT is applied on sub-image block called mask of size 2 x 2 in row major order. Secrete message/image bit is fabricated within the transformed real frequency component of each source image byte except the first frequency component of each mask. After embedding, a delicate re-adjust phase is incorporated in the first frequency component of each mask, to keep the quantum value positive and non fractional in spatial domain. Robustness is achieved by hiding an authenticating or secrete message/image in the frequency component with positive and negatives both quantum values and invisibility is satisfied in spatial domain using delicate re-adjust phase. Inverse DFT (IDFT) is performed after embedding to transform embedded image in frequency domain to spatial domain. Experimental results conform that the proposed algorithm performs better than discrete cosine transformation and Quaternion Fourier Transformation based scheme.

Multiple-Image Encryption Using Genetic Algorithm

Multiple images have to be sent some time for different purposes. Those images encrypt individually before their sending through non-secure channel. This process is painful and also time-consuming. In this paper, multiple images are encrypted using genetic algorithm. Here, a set of crossover points and a set of values are used to diffuse and encrypt the images. These two sets are computed from a key sequence. The original images have been diffused by genetic algorithm. Bitwise XOR operation has been applied between key set and diffuse images to get encrypted images. The effectiveness of the algorithm has been tested by number of statistical tests like histogram analysis, correlation, and entropy test. A comparison is made between the proposed algorithm and other genetic-based encryption algorithm. Finally, it has been observed that the proposed algorithm is given better result in all tests with less execution time.

Lossless Audio Steganography in Spatial Domain (LASSD)

In our proposed work we put an effort to make the technique of standard Least Significant Bit (LSB) coding more secured to embed secret information within an audio file. The prime focus here is to transmit any secret message using an audio signal which can only be retrieved by the intended recipient, while keeping the original characteristics of the carrier audio signal unaltered. In order to achieve the objective, a hash function has been devised to generate pseudorandom positions for insertion and extraction of the secret data bits. In the process of embedding sample amplitude values are read from the source audio and secret data bits are embedded in each of the sample values sequentially at pseudorandom positions. In the process of extraction the reverse technique is applied. Experimental results, both objective and subjective, reveal enhanced performance in terms of imperceptibility and security of the proposed technique.