Tanmay Bhattacharya

A Session Based Multiple Image Hiding Technique using DWT and DCT

This work proposes Steganographic technique for hiding multiple images in a color image based on DWT and DCT. The cover image is decomposed into three separate color planes namely R, G and B. Individual planes are decomposed into subbands using DWT. DCT is applied in HH component of each plane. Secret images are dispersed among the selected DCT coefficients using a pseudo random sequence and a Session key. Secret images are extracted using the session key and the size of the images from the planer decomposed stego image. In this approach the stego image generated is of acceptable level of imperceptibility and distortion compared to the cover image and the overall security is high.

A Novel Fruit Recognition Technique using Multiple Features and Artificial Neural Network

based Fruit recognition technique that has been proposed several times in past revealed in maximum research works either the color and shape or shape and texture or texture and color features. Different fruit images may have same color or identical shape or similar type texture but all three features are rarely identical at the same time for two different types of fruit. So the classification of fruit considering the three features at the same time increases the efficiency and accuracy of the algorithm. The proposed method represents fruit recognition expanding multiple feature based analysis that includes texture, color and shape. To recognize the texture of a fruit the Log Gabor filter has been used, mean hue has been calculated for color and shape has been analyzed by counting perimeter and area pixel. In the study of fruit recognition texture analysis using Log Gabor has been rarely used. The Projected fruit recognition technique is used to extract the above mention three features and Artificial Neural Network is used for classification. The system has achieved more than 90 % accuracy using successful implementation of the proposed algorithm.

An Image Authentication Technique by Handwritten Signature Verification using DWT and ANN

Image authentication is important in content delivery to preserve originality as well as integrity of data. Using handwritten signature we can authenticate a person accurately. This paper proposes an efficient image authentication technique by hiding handwritten signature image in selected DWT sub-band of the image. At the receiver end signature image is extracted and verified with template signature using Artificial Neural Network and hence image authentication is achieved. General Terms Image Authentication, Handwritten Signature verification

Stationary wavelet transform based audio authentication technique

Present work proposes a Stationary Wavelet Transformation based audio authentication technique. Audio stream is converted into two dimensional square matrix forms. 2-D signal is decomposed into four sub bands using Stationary Wavelet Transformation. HH sub band is further decomposed using Stationary Wavelet Transformation. Authenticating image is embedded within the generated HH2 sub band. During embedding, authenticating image is dispersed within the selected sub and using a random sequence and a session key. Authenticating image is extracted using the session key and the size of the image. In this approach, the generated modified audio is of acceptable level of imperceptibility and distortion compared to the original audio.

A Session based Spatial Domain Multiple Image Hiding Technique using Variable Bit Replacement and Multiple Passwords

For hiding secret information in images, there exists a large variety of Steganography techniques. The least significant bit (LSB) insertion method is the most common and easiest method for embedding messages in an image in spatial domain but it has some limitations such as it is easier to understand using steganalysis, this limitation has been overcame in this work. This work intends to embed multiple images in a single cover image. Three images can be embedded within a single cover image in its three different color channels (RGB) and they are embedded in between least significant bit (LSB) and moderately significant bit (MdSB). Three different passwords which are later converted into keys are used for three different color channels. Three different keys of 64 elements are generated from three different passwords. The key consists of the position values of a colored channel of the cover image where the information of a grey image is embedded using variable bit replacement (VBR) technique. Three different channels are combined to form a stego image and later grey images are retrieved from the stego image similarly. This transition of password to key and also the dependency on key imposes greater security. Another important aspect is that the same stego image bears different meaning for different receiver.

A Novel Highly Secured Session Based Data Encryption Technique Using Robust Fingerprint Based Authentication

Fingerprint based authentication does not give a foolproof security because fingerprint image can be scanned and can be used later on for the purpose of authentication. In this paper we proposed another level of security by using Session based password in addition to the features of the fingerprint for authentication. The proposed approach shows significant promise and potential for improvement, compared with the other conventional authentication techniques with regard to efficiency of the process.

A General Bit Level Data Encryption Technique using Helical & Session Based Columnar Transpositions

This paper gives a new approach for encrypting different types of files like text, image, audio, video etc. The file is considered as a binary string. It is first broken into blocks of equal sizes. Each block is then represented as a square matrix. Helical transposition is applied first, and then columnar transposition is done on the basis of session key. Session key is a 24 bit sequence which is used to generate decimal number sequence and accordingly columnar transposition is done. For decryption Session key and anti-helical algorithm both are required.

Analyzing Social Trend Towards Girl Child in India: A Machine Intelligence-Based Approach

In India, feticide or sex-selective abortion is the main reason for the sharp drop in the Child Sex Ratio (CSR). To remove this type of sick mindset from the society and to empower women by proper education, the Government of India has launched a social consciousness program under the banner “Beti Bachao Beti Padhao (BBBP)” when translated stand as “save girl child, educate her”, in the year 2015. It has been observed that, nowadays, researchers are using the power of Natural Language Processing (NLP) to analyze the opinion or suggestions posted on social media. In the proposed work, the opinion expressed by various people on Twitter for a certain period of time has been taken up for analysis. Machine intelligence has been used to classify the opinion. In the current work, instead of binary classification, a multi-class sentiment classification has been introduced. To achieve the objective of the proposed work, four popular classifiers of machine intelligence domain have been used. Accuracy and Kappa statistics has been employed to measure the performance of the proposed model.

A Semi-Fragile Blind Digital Watermarking Technique for Medical Image File Authentication using Stationary Wavelet Transformation

Electronically stored health information enhances resource sharing, help us to reduce the number of errors, speed up clinical communication and assist doctors in diagnosis and treatment. Medical image watermarking is a special type of watermarking technique where the watermarked medical images should not differ perceptually from the original images, because the diagnosis must not be affected due to the presence of watermark. This paper presents a new blind watermarking technique for medical image using Stationary Wavelet Transformation (SWT). In the proposed method multiple watermarks containing patient’s information are embedded and original image is not needed for the extraction of those watermarks. General Terms Watermarking, Medical Image, SWT