Real-Time Implementation of Order-Statistics Based Directional Filters

TL;DR

This paper presents two methods to accelerate order-statistics based directional filters for color images, achieving significant computational efficiency improvements while maintaining accuracy, thus enabling their use in real-time applications.

Contribution

The paper introduces novel techniques to speed up computationally intensive directional filters based on order-statistics, making them suitable for real-time image processing.

Findings

Substantial reduction in computation time for directional filters.

Maintained filtering accuracy despite speed improvements.

Effective on diverse color images.

Abstract

Vector filters based on order-statistics have proved successful in removing impulsive noise from color images while preserving edges and fine image details. Among these filters, the ones that involve the cosine distance function (directional filters) have particularly high computational requirements, which limits their use in time critical applications. In this paper, we introduce two methods to speed up these filters. Experiments on a diverse set of color images show that the proposed methods provide substantial computational gains without significant loss of accuracy.