Two-Exposure Image Fusion Based on Optimized Adaptive Gamma Correction.

Area of Science:

• Computer Vision
• Image Processing
• Digital Imaging

Background:

• High-dynamic-range (HDR) images capture greater detail than conventional images by encompassing a wider intensity range.
• Image fusion techniques, particularly multi-scale exposure fusion (MEF), combine multiple exposures to create HDR images.
• Existing MEF algorithms typically require images with small exposure value (EV) intervals, limiting their applicability.

Purpose of the Study:

• To address the challenge of fusing two-exposure images, especially those captured simultaneously with short and long exposures using spatially multiplexed exposure technology.
• To enhance the quality of fused images by generating a superior intermediate virtual image for the fusion process.
• To improve contrast, saturation, and overall well-exposedness in the final HDR image.

Main Methods:

• Development of an Optimized Adaptive Gamma Correction (OAGC) algorithm to generate an enhanced virtual image.
• Fusion of input images (potentially underexposed or overexposed) with the OAGC-generated virtual image.
• Evaluation against state-of-the-art image fusion methods.

Main Results:

• The proposed OAGC method successfully generates a virtual image that improves the fusion process.
• The fusion technique effectively handles input images with extreme exposure values (underexposed or overexposed), outperforming existing methods.
• Experimental results demonstrate superior performance in generating high-quality fused images compared to other state-of-the-art approaches.

Conclusions:

• The OAGC-based approach offers a robust solution for two-exposure image fusion, particularly for challenging exposure conditions.
• This method significantly enhances the detail and quality of HDR images derived from limited input exposures.
• The technique provides a valuable advancement for applications requiring high-quality image fusion.