Domains and Abstracts of Projects

Image Enhancement Using Transformation Techniques 

Image enhancement plays an important role in image processing and analysis. The existing system assumes that the sensations of color have a strong correlation with reflectance and the amount of visible light reaching observers depends in the product of reflectance and illumination. The main objective of our project is to propose a new image enhancement method to enhance the image as well as to preserve the naturalness of an image. Here, we propose a naturalness preserved enhancement method for non-uniform illumination images using transformation techniques.

Implementation of Adaptive FIR Filter for MTD in Radar Receiver 

This project presents the design of Adaptive Finite Impulse Response (FIR) filter based moving target detection in various clutter conditions in Radar Receiver. To reduce processing time as well as power, pipelined CORDIC architecture has been chosen for design of adaptive FIR filters in conjunction with very robust and popular LMS algorithm for filtering as well as weight update to suppress clutter of various intensity. The design of CORDIC in the circular rotation mode gives a high system throughput due to its pipelined architecture by reducing latency in each individual pipelined stage. Saving area on silicon substrate is essential to the design of any pipelined CORDIC. The area reduction in proposed design can be achieved through optimization in the number of micro rotations. The design has been synthesized and implemented on a Xilinx Virtex4 XC4VFX-12 FPGA device using 13.1 ISE design tool suite and results are shown and discussed. For better adaptation and performance of Adaptive Filters and to minimize quantization error, the numbers of iterations are also optimized.

Development of Efficient Algorithms for Personal Authentication based on Physiological Biometric Traits 

Face and Iris traits are considered to carry out the research work.

Fusion of Spatial and Transform domain features for face Recognition (FRSDL) is presented, the preprocessing is done by resizing the face images to required size of 128*128 to maintain the uniform dimensionality throughout the process. The Stationary Wavelet Transform (SWT) and Discrete Cosine Transform (DCT) are applied on the face images to generate absolute value of coefficients. The Local Ternary Pattern (LTP) is applied on SWT to generate the coefficients. The coefficients generated using SWT, DCT and LTP are concatenated to generate final feature vector set. Face Recognition using Bit-plane and Binary image compression (FRBBC) techniques for different databases such as combined face database, JAFFE, Indian male and Indian female face database are discussed. The MSB based iris recognition using multiple features descriptors is proposed. The left and right iris region is extracted from eye images using morphological operations. Binary split is performed to divide the eight binary bits of every pixel into four Least Significant Bits and four Most Significant Bits. The 2D-Discrete Wavelet Transform is applied on four MSB to extract the iris features. Then Independent Component Analysis is applied on approximate sub band to extract the significant details of iris. Iris recognition based on the multi-level decomposition of 2D-DWT and Bicubic interpolation is presented. The test face and iris features are compared with the database feature set using Euclidean Distance (ED) to determine FAR, FRR and EER. The techniques results low values of FAR, FRR and EER in the proposed system compared to other state of the art methods.