Robust Fuzzy corner detector

TL;DR

This paper introduces a fuzzy logic-based corner detection algorithm that effectively handles imprecise image data, demonstrating robustness and improved performance over traditional methods under uncertain conditions.

Contribution

The paper proposes a novel fuzzy reasoning-based corner detection method that enhances robustness in noisy and uncertain imaging environments.

Findings

Outperforms conventional corner detectors in noisy conditions

Demonstrates robustness across various benchmark images

Effective handling of uncertainties in real-world images

Abstract

Reliable corner detection is an important task in determining the shape of different regions within an image. Real-life image data are always imprecise due to inherent uncertainties that may arise from the imaging process such as defocusing, illumination changes, noise, etc. Therefore, the localization and detection of corners has become a difficult task to accomplish under such imperfect situations. On the other hand, Fuzzy systems are well known for their efficient handling of impreciseness and incompleteness, which make them inherently suitable for modelling corner properties by means of a rule-based fuzzy system. The paper presents a corner detection algorithm which employs such fuzzy reasoning. The robustness of the proposed algorithm is compared to well-known conventional corner detectors and its performance is also tested over a number of benchmark images to illustrate the efficiency of the algorithm under uncertainty.