A Novel design employing arbitrary tree structure for adaptive thrsholding discrete wavelet packet transformation

Abstract

Image compression is a very important technique in the realm of digital image processing. It acting as a key role in enabling effective storage area and data transmission rate of image data. Finite Impulse Response (FIR) filters have proven to be effective tools for image compression due to their linear phase response, which preserves image sharpness during the compression process. In this paper, we propose an innovative design of an FIR filter specifically tailored for effective image compression applications, leveraging the power of wavelet packet transform (WPT). Extensive simulations demonstrate that our proposed FIR filter outperforms conventional FIR filter designs and WPT-based compression methods in terms of both compression ratio and image quality. The filter effectively removes redundant information while preserving visual details, resulting in compressed images that are visually appealing and suitable for various applications. The filter coefficients are calculated serially, and the circuit complexity is decreased by rearranging the intermediate filter coefficients in addition to the memory elements. A 30% reduction in memory elements is required with the pipelined architecture that has been proposed. Additionally, the hardware implementation results demonstrate a 10% and 20% reduction in area and power requirements, respectively.