Infrared-visual image sequence fusion algorithm with noise suppression

Xin Wang, Fu Liu, Wei You-Li, Huantian ZhouProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2013 · DOI

Abstract

Video sequence fusion has a high request on real-time. A new fusion method of infrared and visible video fusion is proposed, which has the characteristics of low computational complexity and noise suppression. Firstly the improved mixed frame difference method is used to achieve the separation of the infrared target areas and the background areas. Secondly the new fusion algorithm is proposed to fuse the target areas of the infrared and visible light sequence. By image smoothness operator , the source images are divided into two parts: the edge region and the smooth region. Different fusion strategies are adopted for the different regions, can highlight the image edges and texture details more accurately and remove redundant, as well as suppressing noise. Finally, the fused target areas are combined with the background area of the visible light sequence to form the final fused image, which can avoid the high background noise of infrared sequence. The experimental results show that the proposed method not only can suppress noise effectively ,but also can acquire good fusion effects as well as achieve the real time need.
BibTeX
@inproceedings{wang2013infrared,
  author = {Xin Wang and Fu Liu and Wei You-Li and Huantian Zhou},
  title = {{Infrared-visual image sequence fusion algorithm with noise suppression}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2013},
  doi = {10.1117/12.2035003},
}

Loading related content...

Figure