A Parallel Image Fusion Algorithm Based on Wavelet Packet

The wavelet packet provides an accurate method for image fusion. However, the computation and time cost will increase with the size of image. Therefore it is difficult to achieve real-time fusion. After having analyzed the wavelet packet based image fusion algorithm on the single computer, according to the time complexity and the parallel character of the wavelet packet algorithm, we propose a wavelet packet based parallel image fusion algorithm under the situation of distributed storage. Based on the data local property of wavelet packet transform, we achieve the algorithm on the MPI (message passing interface) parallel computing platform, which is composed by Pentium PC and 1000 Mbps Ethernet. The efficiency of parallel computing is studied under different image size and different cluster size. The experimental result shows that the algorithm has good parallel computing property.

[1]  James Llinas,et al.  An introduction to multi-sensor data fusion , 1998, ISCAS '98. Proceedings of the 1998 IEEE International Symposium on Circuits and Systems (Cat. No.98CH36187).

[2]  Bicheng Li,et al.  A remote sensing image fusion method based on PCA transform and wavelet packet transform , 2003, International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003.

[3]  Pascal Vasseur,et al.  Introduction to multi-sensor data fusion , 2004 .

[4]  Zhiyun Gao,et al.  Multispectral image fusion using wavelet transform , 1996, Other Conferences.

[5]  R. Chellappa,et al.  SAR/IR sensor image fusion and real-time implementation , 1995, Conference Record of The Twenty-Ninth Asilomar Conference on Signals, Systems and Computers.

[6]  J. M. Hans du Buf,et al.  SPMD image processing on Beowulf clusters: directives and libraries , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[7]  L. Dadda.  Parallel algorithms and architectures for CPUs and dedicated processors: development and trends , 1995, Proceedings 1st International Conference on Algorithms and Architectures for Parallel Processing.

[8]  James Llinas,et al.  An introduction to multisensor data fusion , 1997, Proc. IEEE.