Biometrics, the word not only generates a spark of interest but it is also an important aspect to trace human expressions. Signature, an expression of this kind portrays a uniqueness which can be captured thus it becomes vital. Signature is a mark of authenticity which can be tampered by observing carefully. Dynamic Signature Recognition is one of the highly accurate biometric traits. Live signature of the person is captured, hence it is possible to have dynamic characteristics of signature for matching purpose. The signature captured by digitizer gives information about dynamic nature of signature and pressure applied while signing. This book is discussing use of Dynamic Signature Recognition using Hybrid wavelets. The technique is fast and gives good accuracy and tested in real time. Hybrid Wavelets are used for feature vector extraction, this is done by multiresolution analysis of the dynamic signatures. Wavelet energy based feature vector is generated and this is to be used for the matching of the signatures. The typical features set consists of x, y, z co-ordinates, Pressure, Azimuth and Altitude points. For performance evaluation we are using Conventional FAR-FRR analysis.

In this book a video signal processing algorithm is proposed which is Modified H.264/AVC as a solution for Bandwidth saving through LBR (Low Bit Rate) applications to ensure acceptable video quality. Based on this concept, where the intraframe pictures are hybrid wavelet-based coded and the interframe pictures are coded with a Modified H.264/AVC encoder and decoder is proposed. Use of Modified H.264/AVC algorithm shows that this proposed hybrid encoder and decoder not only retains the same quality of the video but also lowers down the bit rate of the video considerably. AES based encryption and decryption system is proposed for added security for transmission and reception. The proposed algorithm is characterized by their efficient compression ratio, good PSNR, MSE results and lower bit rate as well as its efficient implementation in MATLAB.

Biometric authentication techniques are in high demand for entrance monitoring and security systems. The techniques must be cheap, reliable and, simple. Handwritten signature verification satisfies these requirements. Signature Recognition is a very well known area in Biometrics. Signature of a person is one of the important biometric attribute, has been used for centuries as an authentication measure. In current era signatures are important in business, banking, legal application areas. With the tremendous developments in computer technology and advancements in programming platforms the field of biometrics has seen increments with leaps and bounds. In this book we have discussed an automatic off-line signature verification and forgery detection system based on clustering technique. This system uses the Vector Quantization, Walsh Coefficients, Geometric centers, Grid and Texture features as well as local and Global features of a static handwritten signature.

Morphology is the study of external features of an object. In image processing we use the morphological features based on the shape of the object, from this a template is created and used for matching purpose. We have applied this approach for the problem of handwritten signature recognition. The static signature recognition is simple and cost effective and as large number of commercial transactions are still done on paper, handwritten signatures are widely used for authorization of the documents. Hence the static signature recognition is an important aspect of document authorization using biometric authentication. Static signature is captured from the paper using image scanner, we have only static information such as shape, strokes, orientation available with us. We use techniques based on morphological operations such as erosion, dilation to generate the template and measure the authenticity of the signature.

Biometric Authentication Systems have become inevitable components of modern day security. In this book we have discussed fingerprint, palmprint, finger-knuckle print, face, iris, dynamic signature, keystroke dynamics and their multimodal implementations. In physiological category fingerprints, palmprints & finger-knuckle prints, face & iris are discussed. Their unimodal as well as multimodal implementations are presented. In case of behavioral biometrics dynamic signature & keystroke dynamics for biometric authentication have been investigated. Digital signal & image processing based techniques such as Kekre Wavelets, Kekre's Codebook Generation Algorithms and their variants are used for feature vector generation and recognition purpose. Multimodal systems consisting of the combinations of above mentioned biometric traits are presented. Multi-algorithmic, Multi-instance implementations are presented in this book. A new architecture called as hybrid multimodal system is also presented in this book. This book will be very much helpful for the people who are working in the biometric authentication domain.