IDGCP: Image Dehazing Based on Gamma Correction Prior.

Area of Science:

• Computer Vision
• Image Processing

Background:

• Hazy images degrade visual quality and hinder subsequent image analysis.
• Existing dehazing methods often suffer from high computational complexity or suboptimal restoration quality.

Purpose of the Study:

• To introduce a novel and efficient image dehazing method called IDGCP.
• To leverage a new gamma correction prior (GCP) for improved dehazing performance.

Main Methods:

• Preprocessing hazy images using the proposed gamma correction prior (GCP) for virtual transformation.
• Extracting depth ratio from original and transformed hazy images based on atmospheric scattering theory.
• Employing a global-wise strategy and vision indicator to recover scene albedo and restore the image.

Main Results:

• The IDGCP method achieves high-quality image restoration with a single unknown constant and no refining process.
• Demonstrated significant reductions in processing time and computational cost compared to other methods.
• Experimental validation confirmed the robustness of each IDGCP step.

Conclusions:

• IDGCP offers a superior image dehazing solution in terms of both restoration quality and implementation efficiency.
• The proposed gamma correction prior (GCP) is effective for image dehazing.
• The global-wise strategy simplifies the dehazing process and reduces computational overhead.