Image-matching enhancement using a polarized intensity-hue-saturation fusion method

Area of Science:

• Computer Vision
• Image Processing
• Remote Sensing

Background:

• Atmospheric conditions like haze degrade outdoor scene image quality, hindering image matching.
• Traditional Scale-Invariant Feature Transform (SIFT) is ineffective under hazy conditions.
• Edge information is crucial for improving image matching accuracy.

Purpose of the Study:

• To propose a novel fusion method for enhancing image matching in hazy environments.
• To leverage polarization and depth information for improved keypoint localization.
• To address the limitations of existing image-matching schemes in adverse atmospheric conditions.

Main Methods:

• A new polarized intensity-hue-saturation fusion method is introduced.
• The method fuses a polarized haze-removed image with estimated depth information using the S1 Stokes component.
• An instant dehazing technique utilizing polarized images estimates the haze-removed image and depth map.

Main Results:

• The proposed fusion method generates an image with high spatial details.
• Experimental results demonstrate superior performance compared to existing image-matching schemes.
• The conventional SIFT matching method is significantly improved by the proposed approach.

Conclusions:

• The polarized intensity-hue-saturation fusion method effectively enhances image matching in hazy conditions.
• Utilizing polarization and depth information improves keypoint localization and overall matching accuracy.
• The proposed method offers a robust solution for image matching in challenging atmospheric environments.