Image Enhancement Thanks to Negative Grey Levels in the Logarithmic Image Processing Framework

Area of Science:

• Image Processing
• Computer Vision
• Human Visual System Modeling

Background:

• Image enhancement is crucial in image processing, often focusing on perceived quality.
• Existing methods lack strong physical justification and human visual system modeling.

Purpose of the Study:

• To propose a physically justified image enhancement approach modeling the human visual system.
• To extend the dynamic range of low-light images using the Logarithmic Image Processing (LIP) framework.

Main Methods:

• Utilizing the Logarithmic Image Processing (LIP) framework.
• Introducing negative grey levels as light intensifiers within the LIP model.
• Extending dynamic range for low-light images in real-time.

Main Results:

• Successful extension of low-light image dynamic range to the full grey scale.
• Real-time image enhancement at camera speed.
• Method demonstrated to be reversible and generalizable to color and reflection images.

Conclusions:

• The proposed LIP-based approach offers a physically justified and effective method for image enhancement.
• The technique successfully models the human visual system and enhances low-light images in real-time.
• The reversibility and generalizability of the method open avenues for diverse applications in image processing.