Dynamic partial reconfiguration of 2-D haar wavelet transform (HWT) for face recognition systems

摘要

This paper presents two novel architectures for twodimensional(2-D) Haar wavelet transform (HWT) oftransform block in face recognition systems. The proposedarchitectures comprises 2-D HWT with transpose-basedcomputation and dynamic partial reconfiguration (DPR)that have been synthesised using VHDL and implementedon Xilinx Virtex-5 FPGAs. To evaluate the proposedarchitecture, comparison for both configurations anda detailed performance analysis in terms of area, powerconsumption and maximum frequency are also addressedin this paper.