Research on Image Fusion Method Based on Wavelet Transform and NMF

The wavelet transform using in image fusion processing can get much more detail information of the source images in both airspace and frequency domains, but the high-frequency noise will be also saved to the fusion image. This paper brings out a new method based on NMF and wavelet transform for image fusion. Because of the good performance of NMF on character extraction, we can get more feature information and reduce the high-frequency noise to some extent. The result of image fusion is more faithful to the actual scene. Experiment shows that the method proposed in the paper has good performance on preserving the feature information for the test images.