Hardware implementation and power analysis of HWT for medical imaging

Processing large images requires a large amount of memory, power hungry and computationally intensive which causes a high demand for an efficient algorithm implementation to overcome the problems. This paper discusses an efficient hardware implementation of the Haar wavelet transform (HWT) core on reconfigurable platforms. The proposed HWT implementation can be explored for automatic detection and segmentation of tumour in medical images. The HWT algorithm has been demonstrated for the segmentation of phantom data and its hardware implementation has been carried out using Handel C on different field programmable gate arrays (FPGAs) devices. Results obtained for the implementation performance in terms of area and power have been evaluated and compared with other existing systems for HWT computation.