Contrast-dependent saturation adjustment for outdoor image enhancement

Area of Science:

• Computer Vision
• Image Processing
• Digital Photography

Background:

• Outdoor images in adverse weather conditions often exhibit reduced contrast and color saturation due to light scattering and attenuation.
• Existing image enhancement techniques frequently lack generalizability, performing well on limited image sets but failing on diverse datasets.

Purpose of the Study:

• To develop a robust image enhancement method for improving the quality of outdoor images captured in poor weather.
• To address the limitations of current methods by proposing a technique that enhances both intensity contrast and color saturation adaptively.

Main Methods:

• A novel content-adaptive histogram equalization method, inspired by the human visual system, is proposed to enhance image intensity contrast.
• A contrast-dependent saturation adjustment prior is introduced to refine color saturation based on the improved intensity contrast.

Main Results:

• The proposed method demonstrated superior performance in enhancing image visibility and preserving naturalness compared to eight state-of-the-art enhancement and haze removal techniques.
• Experimental results validated the method's effectiveness across various types of challenging outdoor images.

Conclusions:

• The developed image enhancement technique offers a significant improvement for outdoor photography in adverse conditions.
• The combination of content-adaptive contrast enhancement and contrast-dependent saturation adjustment provides a more effective and natural-looking image quality improvement.