Research On Fusion Of Remote Sensing Information Of Infrared Image And Microwave Image

Along with the development of the sensor technology, images can be gained by more and more methods. Multi-source image information fusion is one of the important and effective way on using the multi-source data. Multi-source image information fusion is a new and important but not mature research area in remote sensing. And as a typical dissimilar image information fusion way, remote sensing information fusion of infrared image and microwave image is also be concerned as a very important way leading to fine image quality, fault tolerance and accurate target recognition. Based on the comprehensive review and summarization of previous articles and researching achievements, basic concepts, levels, models, structures, techniques and applications of the multi-source remotely sensed image information fusion are discussed. Pretreatment of infrared image and microwave image before fusion process are also discussed. Based on Bayes theory and Dempster-Shafer theory, a new fusion method for the target recognition of infrared image and microwave image is proposed. We named it Multi-levels information fusion arithmetic. The performance of the method is evaluated by fusing infrared image and passive microwave image, infrared image and SAR image. So it is an effective and robust fusion method for the target recognition of remote sensing. It is a guideline both to build an available image information fusion system and the research on fusion of remote sensing information of infrared images and microwave images.