Arnab Kumar Maji

A Novel Image Steganographic Scheme Using \(8 \times 8\) Sudoku Puzzle

The word steganography has originated from two Greek words, namely, stegos and grahia meaning cover and writing, respectively. The concept behind steganography is that the transmitted message should be indiscernible to the regular eye. It is the art of concealing information within a medium. The main aim of steganography is to protect this concealed information. A message transmitted using a cipher text may incite suspicion, but an invisible message will not. Digital images constitute a flexible medium for carrying secret messages because slight modifications done to a cover image is hard to be distinguished by the human eyes. In this work, we have proposed an Image Steganographic Scheme using \(8{ \times }8\) Sudoku puzzle for secure data transmission. The originality of the work is that if any intruder changes any part of the secret message, then the proposed algorithm will be able to detect it.

A Novel Biometric Template Encryption Scheme Using Sudoku Puzzle

Identity theft is a growing concern in the digital era. As per the US Federal Trade Commission, millions of people got victimized in each year [1]. Traditional authentication methods such as passwords and identity documents are not sufficient to combat ID theft or ensure security. Such representations of identity can easily be forgotten, lost, guessed, stolen, or shared. On the contrary, biometric systems recognize individuals based on their anatomical traits (e.g., fingerprint, face, palm print, iris, and voice) or behavioral traits (e.g., signature, gait). As such traits are physically linked to the user, biometric recognition is a natural and more reliable mechanism for ensuring that only legitimate or authorized users are able to enter a facility, access a computer system, or cross international borders. Biometric systems also offer unique advantages such as deterrence against repudiation and the ability to detect whether an individual has multiple identity cards (e.g., passports) under different names. Thus, biometric systems impart higher levels of security when appropriately integrated into applications requiring user authentication. In this paper, an attempt has been made to secure the biometric data using sudoku puzzle.

A Comparative Study on Segmentation Techniques for Brain Tumor MRI

Segmentation of brain tumor from medical images is an interesting topic which is investigated by many researchers. It is important to locate tumor at an early stage so that it can be easily healed and can be used for further diagnosis. There are different imaging techniques which are used in segmentation of brain tumor. Among them, Magnetic Resonance Imaging (MRI) is most widely used radiological tool as it is radiation free in nature. For detecting the size, shape, and location of the tumor many segmentation algorithms were used. In this paper, an exhaustive study of brain tumor from MRI images has made by using different techniques. A comparative study and performance evaluation of different techniques based on certain performance metrics are also discussed in this paper.

Extraction and Analysis of Expression from a Captured Face

With the help of computer science in today’s world many advances have been made in the field of facial expression recognition. It is considered as an important measure of social interactions and there are many applications that can be helpful to human in many ways as facial expressions not only show the emotional status of an individual but also other mental activities as well. To develop such a system is a challenging task as it involves learning and understanding of expressions of human which varies from time to time. In this paper, a method on facial expression analysis has been proposed where face detection was done using RGB color model and feature classification using Gabor filter.

A Robust Video Steganographic Scheme Using Sudoku Puzzle for Secured Data Hiding

Video Steganography is a method for securing data in a video file and hence it prevents the chance of theft. The proposed scheme uses the Video Steganography technique on the encrypted data (using RSA and AES) for providing maximum security and stability. In this work, a secured Video Steganographic scheme is proposed using Sudoku puzzle. The novelties of the work imply that nobody could be able to see as well as modify data during transmission as the secret message is first encrypted into cipher text twice using RSA and AES cryptographic technique then that cipher text is embedded in video using a Sudoku reference matrix. So, our proposed work combines both Cryptographic and Steganographic technique to enhance the security in maximum level with maintained computational complexity.

An Indirect Addressing Image Steganographic Scheme Using 9 × 9 Sudoku Matrix

In recent world, everybody wants to transmit their valuable information securely without the knowledge of the intruder. One of the way-out is steganography. Using steganography, information can be transferred using cover media. Cover media can be image, video, sound, or text. Most popularly used cover media is a digital image because the slight modification of a cover image is difficult to distinguish by human eye. In our paper, the proposed method uses Sudoku puzzle to hide secret message within a cover image. The benefit of using Sudoku is if any intruder modifies the secret message then receiver will be able to detect this. In this proposed method, the length of the secret message is directly proportional to the resolution of the cover image. The proposed method has been developed to transmit the secret message more securely. Here, a single 9 × 9 Sudoku puzzle is used, and this Sudoku puzzle is embedded within the image by modifying the RGB value of pixels. Within each cell of the Sudoku puzzle we indirectly store one character of the secret data; hence, the intruder cannot fetch the secret message directly, to detect if intruder modifies the secret message or not, and associated information is stored which makes the proposed method more reliable than other existing methods. This proposed method is better than the existing methods because it uses RGB value, whereas the most of the methods used grayscale value.

Study and Analysis of Different Face Recognition Techniques Based on Graph

Human face recognition is very interesting as well as very challenging area of research. In our study of different existing face recognition schemes, we have seen that it has very important role in many applications. For this paper, we have studied and worked on the various steps of face recognition and analyzed a method to work on them. Where the face is detected from an input image and Gabor filter is applied. Then, two approaches are used (1) Delaunay triangulation and (2) Euclidean distance which are apply to the extracted feature points (which is referred as node set graph) and they are stored in the database. These store node set is used for checking the similarity with the input image. We have also compared both of the proposed approach in the system and we found that node set with Euclidean distance gives a better recognition rate.

Comparative Evaluation of Radial Basis Function Network Transfer Function for Filtering Speckle Noise in Synthetic Aperture Radar Images

Synthetic aperture radar (SAR) is imaging radar. It uses the movement of the antenna to synthesize a very large antenna and produces good resolution images of the mapped area. SAR images are polluted with speckle noise. Speckle noise inherently exists in all types of coherent imagery including SAR. It reduces the quality of the image and causes serious difficulty in automatic interpretation of the images. The radial basis function network (RBFN) is gaining importance in speckle noise filtering. RBFN is an artificial neural network (ANN) that has only one hidden layer known as the radial center. In this paper we have used Gaussian function, thin plate spline function, quadratic function, and inverse quadratic function as radial centers to design and implement different RBFNs. The performance of these RBFNs for filtering speckle noise is evaluated based on mean square error(MSE) achieved during training and peak signal-to-noise ratio (PSNR).

A Hybrid Face Recognition Scheme Using Contour and Gabor Wavelet

A face recognition system based on the hybrid approach is proposed in this paper. We have combined both the global and local features of a face using Gabor wavelet transform and a face contour. Gabor transform helps in extracting the features of a face, that is, the position of eyes, tip of the nose, and the corner of the mouth, whereas a face contour helps to extract the structure of a face. A combination of these techniques helps to extract the best features.

A Comprehensive Sudoku Instance Generator

Sudoku puzzles have become popular worldwide among many players of different intellectual levels. In this paper, we develop algorithms for creating numerous Sudoku instances of varying levels of difficulty. There are several ways to generate a Sudoku instance. The most popular one is to consider one solved Sudoku puzzle and remove some of the numbers from the cells based on the required difficulty level. Although all Sudoku puzzle creators assume that there is a single solution for a generated puzzle, we notice that a Sudoku puzzle may have multiple solutions. None of the instance generation techniques are able to find how many different solutions are present for a generated Sudoku instance. Here in this paper, we have devised one novel approach that can generate a Sudoku instance and check for its number of possible solutions, and then based on the number of solutions we further categorize the instance that has been generated. This approach is entirely novel and comprehensive for generating Sudoku instances.