A fast and efficient two-scale fusion of infrared and visible images using median filter and intensity transfer

We suggest a fast and efficient method for fusing infrared and visible images, which is based on a median filter and an intensity transfer. The median filter is used to construct a new two-scale image decomposition and obtain base and detail layers. Then a fused base layer is obtained by utilizing an optimization fusion rule based on intensity transfer, where the fusion is transformed into matrix operations. A fused detail layer is obtained by an improved maximum-selection rule which maximizes the details-appearance information with some image-enhancement effect. Our experimental results demonstrate that the method can provide a superior competitive fusion performance in terms of both subjective evaluation and objective metrics, when compared with seven standard state-of-the-art methods. Moreover, our method is quite suitable for real-time fusion applications.

Automated summary

We propose a fast and efficient method for fusing infrared and visible images based on median filter and intensity transfer. The method achieves superior fusion performance in terms of both subjective evaluation and objective metrics when compared with seven standard state-of-the-art methods, and is suitable for real-time fusion applications.