Conference of Computational Interdisciplinary Science

Infrared and visible images can transmit different information of the objects within an image. Important hidden objectives can be observed in infrared images. The visible images contain abundant detail of texture and high spatial resolution, being able to reveal clearly the information of the scene. Commonly, when fusing infrared and visible images there is a loss of texture, brightness and contrast information.
A fusion of images, which preserves relevant information, is quite useful for use in later stages.
In this work, an algorithm of fusion of visual and infrared images based on multi-scale top-hat transform is proposed. This algorithm provides significantly better results (in terms of metrics such as contrast, brightness, texture and image information) than state-of-the-art algorithms.