Fusion of visual and infrared signals in visual sensor network for night vision

Abstract

Multisensory fusion has become an area of intense research activity in the past few years. The goal of this paper is to present a technique for fusing infrared and visible videos. In this technique we propose a fusion method that quickly fuses infrared and visible frames and gives a better performance. This is done by first decomposing the inputs using DWT and extracting two maps (resulted from Choose Max rule) from approximation sub frames and then fusing detail subframes according to these maps. After being compared to some of the popular fusion methods, the experimental results demonstrate that not only does this proposed method have a superior fusion performance, it can also be easily implemented in visual sensor networks in which speed and simplicity are of critical importance.