A fusion method of infrared and visible images based on visual salience difference

Most infrared and visible image fusion methods use indiscriminate feature extraction strategies, which ignore the differences between multiple source images, and it is difficult to remove the interference of redundant information, such as noise according to visual habits, which has a negative impact on the fusion results. This paper proposes a fusion method of infrared and visible images based on visual saliency differences. Firstly, to highlight the target and suppress the background noise in the salient map, a visual attention system that conforms to the image fusion task is designed to accurately extract the salient visual features of the source image. Secondly, a salient feature adaptive fusion strategy is proposed based on image content and feature correlation. A saliency map is constructed based on visual salient features. Finally, based on the saliency map fusion strategy, the infrared and visible images are combined, and the texture details of weak activity areas are moderately enhanced. Experimental results show that the proposed method can effectively preserve the saliency of infrared targets, and the fusion results have uniform brightness distribution, clear edge outlines, and better visual effects.