Infrared and Visible Image Fusion Method Using Salience Detection and Convolutional Neural Network.

Area of Science:

• Computer Vision
• Image Processing
• Artificial Intelligence

Background:

• Infrared and visible image fusion aims to combine complementary information from different sensors.
• Existing methods often struggle with preserving salient features and suppressing noise effectively.

Purpose of the Study:

• To develop an advanced algorithm for infrared and visible image fusion.
• To improve the quality and perceptual detail of fused images.

Main Methods:

• Utilized global contrast-based significance detection for infrared image salient feature extraction.
• Designed a specialized loss function for infrared feature extraction and reconstruction, complemented by gradient loss for visible images.
• Employed a modified residual network for feature extraction and image reconstruction.

Main Results:

• Fused images exhibit enhanced sharpness and richer scene information.
• The fusion process results in outputs closely resembling high-quality visible images.
• Demonstrated superior performance compared to existing state-of-the-art fusion methods.

Conclusions:

• The proposed algorithm effectively integrates discriminatory features from infrared and visible images.
• The method achieves high-quality image fusion with improved visual perception and detail preservation.
• The model exhibits strong generalization capabilities across different sensor limitations.