ENHANCING IMAGE QUALITY: DECISION-BASED FILTERING TECHNIQUES FOR IMPULSE NOISE REMOVAL

Abstract

Images corrupted by impulse noise are commonly encountered in practical scenarios. Such noise arises from various sources including channel-decoded damages, signal degradation during transmission in communication channels, noise from video sensors, and other system partitions. Prior to the advent of advanced image processing tools like Photoshop and other digital restoration software, photo restoration was primarily carried out manually by restoration experts using techniques such as airbrushing directly on the damaged photo. This research proposes a novel decision-based filtering technique that integrates K-means clustering with Principal Component Analysis (PCA) to efficiently reduce unwanted noise, resulting in improved image quality. The proposed filter has been shown to provide superior performance when compared to traditional filtering techniques.