Multi-sensor Image Fusion Algorithm

Image fusion technology, which owns many good characters has been widely researched and applied in many fields in recent years. This paper is focusing on the study of image fusion method for same sensors and different sensors. The following content is the main work of the paper.Firstly, some general methods for the same sensors are simulated to analyze the advantages and disadvantages of them;Secondly, fusion methods based on multiresolution analysis have been discussed in this paper, and a fusion framework which divides the fusion process into four modules which are measure module, match module, decision module and fusion module is shown. In addition, some typical fusion methods are simulated.Thirdly, a new false color image fusion algorithm based on HSI color space for different sensor is presented. Comparing to other false color methods, our algorithm can not only reserve better space resolving rate, but also present important information of a special sensor. Experiment result shows the performance of our algorithm is satisfied.Fourthly, based on analyzing different application fields, several image fusion evaluation algorithms have been created. These algorithms can deal with the fusion image evaluation problems of both same sensors and different sensors. What' s more, based on those fusion and evaluation algorithms mentioned above, we bring forward a new image fusion framework, which can be based to construct a practical fusion system.