P000052R1
Multi-focus Image Fusion Based on Quaternion Wavelet Transform
Ming Yin (School of Mathematics, Hefei University of Technology)
Jiangmin Wu (School of Mathematics, Hefei University of Technology)
*Biao Chu (School of Mathematics, Hefei University of Technology)
For the characteristics of multi-focus image and the shortcomings of traditional wavelet in image fusion multiscale analysis, we propose multi-focus image fusion method based on quaternion wavelet transform (QWT) in this paper. First, the original images are processed by quaternion wavelet transform,the local region energy is adopted for the low frequency sub-bands, while a novel sum-modified-Laplacian(NSML) is employed for the high frequency sub-bands. Then the fused image is achieved by the inverse quaternion wavelet transform, the algorithm has a certain effectiveness and practicality. Experiments show that comparing with traditional methods, the fusion result of this method is better in subjective vision effect and objective evaluation index.