Image Fusion Based On Wavelet Transform

The main purpose of image fusion is to enhance reliability of the image through processing redundant data between many images and to enhance clarity of the image through processing supplementary information between many images. The image fusion has following merit: expanding system operating region, enhancing system reliability, gaining more highly effective expression form of information and improving the performance-to-price ratio of system. Image fusion technology occupies important places in the fields of remote sensing, medicine, natural resource exploration, marine resource management, terrain landform analysis and biology.The key point of this thesis is image fusion method based on the wavelet transformation. Involved content included the image fusion's concept, level, purpose, wavelet decomposition, criterion etc.. Wavelet transformation is applied in image fusion. For wavelet resolution ratio matrix after decomposition, its low and high frequency component are differentiated and different algorithms were used. It is also analyzed that different influence on image fusion effect due to different fusion algorithms. The algorithms includes: weighted average rule and importance rule which aim at low frequency component as well as area energy rule, area variance rule, everage grad rule and PCA rule which aim at igh frequency component. The theory importance rule and PCA rule is majorly introduced and practised. Adopting this fusing method leads to better fusion effect. This conclusion is verifyed under the appraisal targets enumerated in the thesis. Besides, the thesis has also compared and analysised effections on fusion due to different wavelet functions and decomposition levels.Image splicing technology may apply in satellite imagery as well as daily life pictures conformity. The thesis has conducted preliminary study on image splicing technology and SIFT matching method, applied my fusing method to image splicing technology and mended splicing aperture and chiaroscuro caused by simple splicing method.