Infrared and visible image fusion via octave Gaussian pyramid framework

Area of Science:

• Computer Vision
• Image Processing
• Signal Processing

Background:

• Image fusion combines data from multiple images for enhanced perception.
• Multiscale decomposition is a common image fusion technique.
• Existing methods may not fully capture all image information.

Purpose of the Study:

• To propose a novel image fusion framework based on the octave Gaussian pyramid principle.
• To improve information retrieval and detail preservation in fused images.
• To enhance the performance of image fusion for human and computer vision.

Main Methods:

• The proposed method utilizes an octave Gaussian pyramid for image decomposition.
• Images are decomposed into two scale spaces: octave and interval spaces.
• The framework generates multiple sets of detail and base layers, unlike traditional methods.

Main Results:

• The octave Gaussian pyramid framework retrieves more information than conventional multiscale decomposition.
• The method efficiently retains both high- and low-frequency information from original images.
• Qualitative and quantitative comparisons show superior visual effects and objective evaluation scores.

Conclusions:

• The proposed octave Gaussian pyramid-based image fusion framework offers significant improvements.
• It provides better visual quality and objective performance compared to existing methods.
• This approach enhances the integration of information for more informative composite images.