Swanirbhar Majumder

Coded Fingerprinting Based Watermarking to Resist Collusion Attacks and Trace Colluders

The alteration, repackaging, and redistribution of multimedia content like image, video, audio, etc. poses a serious threat to both national security and commercial markets. Presently with the growing popularity of digital media the world is becoming smaller and smaller. That is mainly due to the internet connectivity and WWW phenomena. But the copyright protection of intellectual properties has become a necessity for prevention of illegal copying and verification of content integrity. Here wavelet transform plays the role of an efficient and robust tool due to its multi-resolution capability along with singular value decomposition (SVD) for watermarking. Both of these when implemented with coded fingerprinting not only resist collusion attacks but can also help tracing colluders. This work here by focuses implementation of a hybrid watermarking scheme using SVD and wavelet transform along with coded fingerprinting developed for protection of the intellectual property.

Digital Image Compression Using Neural Networks

Compression of data in any form is a large and active field as well as a big business. Image compression is a subset of this huge field of data compression, where we undertake the compression of image data specifically. Research in this field aims at reducing the number of bits needed to represent an image. Inter-pixel relationship is highly non-linear and unpredictive in the absence of a prior knowledge of the image itself. Thus Artificial Neural Networks has been used here for image compression by training the net using the image to be compressed. The ANN takes into account the psycho visual features, dependent mostly on the information contained in images. The algorithms, on application on the image data preserves most of the characteristics of the data while working in a lossy manner and maximize the compression performance. The results have been checked with and without the use of quantization, and without median filtering of the image. The ANN algorithm used here is mainly the back-propagation of multilayer perceptrons.

SVD and Error Control Coding Based Digital Image Watermarking

Digital media is a craze these days. With the present day internet connectivity the world has become smaller. The copyright protections of intellectual properties have become a necessity for prevention of illegal copying and content integrity verification. Newer data hiding techniques that satisfy the requirements of imperceptibility, robustness, capacity, or data hiding rate and security of the hidden data etc. are being developed. Therefore the preference to go for digital image watermarking, to show resiliency against various unintentional or deliberate attacks has increased. Watermarking is a method of data authentication by embedding a watermark in image characteristics with expectation. Here in this work, this is done using a mathematical tool called the singular value decomposition (SVD). Earlier SVD had been used individually, as a tool for watermarking of digital media, but this work incorporates error control coding to it, by a novel zig-zag scan operation on pixels done before coding, thereby increasing the robustness. This work here by focuses on using both of them together to provide a robust technique developed for protection of the intellectual property against the popular malicious attacks. The algorithm is later applied and tested on the standard checkmark techniques practically by attacking the watermarked image against standard simulated attacks and recovering the watermarked logo from it.

Back propagation artificial neural network (BPANN) based performance analysis of diesel engine using biodiesel

This work deals with the implementation of back propagation based artificial neural network (BPANN) model for prediction of the different diesel engine parameters, i.e., mean effective pressure, mechanical efficiency, fuel consumption, air-fuel ratio, and torque, to overcome the difficulties of practical experimentation and minimization of time and cost in this endeavor. The parameters provided as input to this model were engine speed, load and different biodiesel, and diesel fuel blends. The model has been trained with approximately 85% of the test results, and the rest were kept for prediction. The number of hidden layers and the number of neuron in each layer were varied to achieve the best predicting model. Once the model was ready, it had been used to predict the remaining data where the mean square error was as low as 6.51×10−4 and had a very low total R value. This work hereby shows that the BPANN based model can predict the performance of diesel engine fed with biodiesel and diesel fuel blends.

DWT and SVD based image watermarking scheme using noise visibility and contrast sensitivity

This paper presents a robust and imperceptible methodology of watermark embedding in the transform domain using Discrete Wavelet Transform (DWT). The robustness is brought about by hiding the watermark in the Eigen values after computing the Singular Value Decomposition (SVD) on low frequency sub-band after DWT. While for the imperceptibility the Contrast Sensitivity Function (CSF) has been employed here along with the Noise Visibility Function (NVF). Human beings are sensitive to visual contrasts so CSF is used to take care of that while NVF characterizes the local image properties of texture, edge and smoothness to determine the optimal watermark locations and strength at the wavelet sub-bands for watermark embedding. The algorithm has been tested on 16 metrics out of which 10 are for imperceptibility and 6 for robustness for some popular attacks and the result has been found encouraging.

SVD and Neural Network Based Watermarking Scheme

Presently the WWW phenomenon has brought the world in to the personal computer. Digital media is thereby given high priority. But this has increased the frequency of security breach of intellectual properties. Therefore copyright protections and content integrity verification are highly recommended. What we need is newer data hiding techniques that must be imperceptible, robust, highly secured, etc. Digital image watermarking is one of the ways that is resilient to various attacks on the image based digital media where data authentication is done by embedding of a watermark in image characteristics. This work incorporates singular value decomposition (SVD) based image watermarking. Here unlike previous work done by researchers, error control coding (ECC) and artificial neural networks (ANN) for the authentication purposes have been used. ECC and ANN increase the robustness of the method against malicious attacks.

Estimation of arrhythmia episode using variational mode decomposition technique

Detection of life threatening arrhythmia like ventricular flutter (VFL) or ventricular tachycardia (VT) and atrial flutter (AFL) at the earlier stage may save life by defibrillation therapy. Different type of mechanism has been proposed earlier in time domain or by spectral analysis of ECG signal. In this work a frequency domain approach is proposed by spectral decomposition of ECG signal. Spectral decomposition of signal is done with the help of variational mode decomposition (VMD) technique. VMD model is used to obtain the required number of spectral mode of the test signal and their central mode of oscillation. This central frequency of decomposed signal and maximum phase change within a specified window are used to characterize ventricular tachycardia and atrial flutter and compared to normal rhythms by K-near neighbour (KNN) classification method. Accuracy of 98.6% is obtained for VT classification. The proposed method eliminates the requirement of detecting fiducial points of ECG signal as necessary in conventional classification methods.

Gene expression programming based age estimation using facial features

The core target of this paper is to estimate human age automatically through facial image analysis. In this research study we put forward a system constructed on the basis of Gene Expression Programming (GEP) to estimate human ages using face features. Gene expression programming (GEP) is a handy tool to find out functions. Due to prompt developments in machine vision and computer graphics, age estimation through faces have turn out to be most dominant issues now a days due to their widespread applications in real world, such as safety control, investigation monitoring, biometrics, scientific art, automated client relationship management and cosmetology. As it is difficult to estimate the actual age, our system is going to estimate the ages within certain ranges. Total age range is classified into four classifications which differentiate the individuals oldness in relation with age. Our proposed approach has been initialized with GEP and then developed and tested using MATLAB. A public data set, FG-NET was used to develop the system. The quality of the proposed system for image-based age estimation is shown by broad experiments on the available database of FG-NET. To assess the performance of our system, we have done a relative study based on various parameters of GEP and found significant results.

Spread Spectrum Embedding of Colluder Traceable Codeword in Multimedia

With digital multimedia distribution, intellectual property rights (IPR) are at a more risk than ever, due to the possibility of unlimited copying without fidelity loss. Encryption and copy protection mechanisms do not fully efficiently deal with the issue. Encryption usually protects the data only on the transport channel, and as soon as the data are decrypted for exhibit or playback, they can be copied. Copy protection mechanisms are difficult to realize in open systems as data are open to all. Thus, both encryption and copy protection offer only limited security. Watermarking has been proposed as a last line of defense in the protection of IPR. Watermarking methods embed information irremovably and invisibly into the host data. The watermark typically contains information about source and recipient of the distributed data. Thus, if pirated copies of the data are distributed, it can still be determined who owns the copyright and who the original, authorized recipient was. Thus, it allows tracing back illegally produced copies of the data. In the time of watermarking it is possible to assign some unique codeword to every copy to be distributed, by which it is possible to detect any alteration made to the watermarked copy. This paper presents spread spectrum watermarking of a specially designed colluder traceable codeword with help of combinatorial design in digital image.

Hybrid image watermarking scheme using SVD and PDFB based contourlet transform

Copyright protection of intellectual property is a very large and growing business. This is due to the easier availability of all multimedia data with just a click of a mouse, resulting to the increase of piracy. Watermarking is one of the popular methods of authentication to help protect intellectual property rights. Of the various forms of multimedia data image is one of the most popular variant. Here a method of digital image watermarking is presented using singular value decomposition (SVD) which one of the very popular mathematical computation tools. Along with it a lot of transforms have been used earlier like discrete wavelet transform (DWT), DCT, Fourier transform, etc. to provide hybrid techniques. But here an unconventional transform introduced in the middle of this decade, called contourlet transform (CT) has been used. This transform being similar to DWT fits well to provide a robust watermarking scheme along with the SVD. Here the combination of pyramidal and directional filter bank (PDFB) in CT has been used.