Retaining Image Feature Matching Performance Under Low Light Conditions

TL;DR

This paper explores how to improve feature matching in low light images by adjusting detection thresholds and applying image enhancement techniques, demonstrating that these methods can significantly enhance matching performance.

Contribution

It investigates the impact of threshold tuning and pre-processing with LLIE on feature matching in low light conditions, offering practical strategies for better image analysis.

Findings

Lowering feature detection thresholds improves matching in low light.

LLIE pre-processing enhances feature matching performance.

Traditional detectors remain effective with proper parameter tuning.

Abstract

Poor image quality in low light images may result in a reduced number of feature matching between images. In this paper, we investigate the performance of feature extraction algorithms in low light environments. To find an optimal setting to retain feature matching performance in low light images, we look into the effect of changing feature acceptance threshold for feature detector and adding pre-processing in the form of Low Light Image Enhancement (LLIE) prior to feature detection. We observe that even in low light images, feature matching using traditional hand-crafted feature detectors still performs reasonably well by lowering the threshold parameter. We also show that applying Low Light Image Enhancement (LLIE) algorithms can improve feature matching even more when paired with the right feature extraction algorithm.