Image Fusion Based On Multi-resolution Wavelet Transform

Image fusion is combing images that get from two or more sensors at the same time or different time ,then images are processed by some rules ,the processed images can be used in future research. Multi-focus image fusion is fusing several images which are got by each target focused separately and shot repeat. The purpose is getting a clear image in which each target is distinct. It is widely used in these areas such as digital camera , machine vision and object recognition and so on.The existing image fusion techniques and wavelet transform techniques are analysis firstly, then the principles of multi-focus image fusion based on wavelet transform and the rules of fusion are researched, taking into account the result of the traditional rules based on regional energy in fusing the explicit and implicit changes in the gray area of the image are not ideal, the algorithm based on the regional energy of multi-scale adaptive is presented. Finally, the superiority of this method is proved by experiment.The rules widely used in low-frequency area are weighted average, the simulation found that the index getting from principal component analysis is higher than weighted average .For the high frequency region , the multi-focus image fusion algorithm based on the coefficient matrix is proposed, which first of all, the source images are carried out by the discrete wavelet frame transform, in accordance with the discrete wavelet frame transform coefficients to calculate each image pixel clarity indicators, then the clarity of indicators of the image pixels are used to calculate the coefficient matrix of the image information, through this information coefficient matrix to determine how to choose a pixel to accomplish our ultimate fusion images in the wavelet domain Finally, through experiments to prove the superiority of the proposed algorithm. Large comparison of fused images which are got by the proposed algorithm and old methods are done according to some image evaluation standards, and then the corresponding conclusions are drawn.