Multisensor Video Fusion Based on Surfacelet Transform
Based on the three dimensional surfacelet transform, a novel multisensor video fusion algorithm is proposed, in which, multiple frames of input videos are regarded as special three dimensional signals and fused as a whole rather than frame by frame. Meanwhile, selection principles for different subband coefficients are discussed in detail. Experimental results demonstrate that the proposed algorithm outperforms the traditional individual-frame-based and motion-based methods in spatial-temporal information extraction as well as in temporal stability and consistency. In addition, it does not require the motion target detection pre-step and is easy to implement.
Q. Zhang et al., "Multisensor Video Fusion Based on Surfacelet Transform", Advanced Materials Research, Vols. 341-342, pp. 163-167, 2012