Abstract

Filters are used to remove the different types of noises including salt and pepper, Gaussian, and random noises from image. Therefore, the VLSI oriented hardware implementation of filters plays the crucial role in real time applications. However, the conventional hardware-based filters are failed to reduce the look-up-table (LUT)s, path delays, and power consumption. Therefore, this work is focused on implementation of Gammatone Median Filter (GMF) using Data Comparator (DC) logic. Initially, the multiplexer selection logic-based data comparator is used to identify the high and low values from two numbers. Then, data comparator is repeated for multiple numbers of times for nine pixels combinations, which identifies the median value from nine pixels. The subjective and objective evaluation shows that the proposed GMF-DC resulted in superior performance in terms reduced noise, hardware metrics like LUTs, delay, and power consumption as compared to state of art approaches.

Description

.