Visibility Restoration: A Systematic Review and Meta-Analysis

Area of Science:

• Computer Vision
• Image Processing
• Optical Engineering

Background:

• Image acquisition is susceptible to environmental factors, impacting applications.
• Visibility restoration is essential for high-level computer vision and photography tasks.
• Poor weather conditions significantly degrade image quality.

Purpose of the Study:

• To systematically review and analyze visibility restoration algorithms.
• To focus on algorithms effective in adverse weather conditions.
• To propose a novel framework for haze visibility restoration.

Main Methods:

• Introduction to optical image formation principles.
• Comprehensive description and comparative evaluation of existing algorithms.
• Development of a general framework using haze-relevant features and maximum likelihood estimates.

Main Results:

• A systematic review and meta-analysis of visibility restoration techniques.
• Identification of current challenges in the field.
• A proposed framework for enhancing visibility in hazy conditions.

Conclusions:

• Visibility restoration is critical for robust image analysis.
• The proposed framework offers a promising approach for hazy weather.
• Further research is needed to address existing difficulties and advance the field.