The Application Of Multi-source Image Fusion Algorithm In The Water Conservancy Image Processing

With the development of modern water utilities, especially the development of "Goldwater Project", some modern technology has been widely used, such as the rain real-time monitoring system、for small river basins planning and harnessing. For the quality of the equipment, weather conditions, time period and other factors, the low quality of the image is difficult to meet the work requirements. Meanwhile, with the development of the remote sensing technology in recent years, the single visible mode is instead of the multi-source sensor mode. The former can not be taken to reflect the characteristics of things, but the latter can be from many aspects. Because the sensors have different mechanisms, working different wavelength range, shooting in different environments and so on, the information we have obtained many and complex, and it is difficult to deal with them. It is worth to consider how to make more effective use of multi-source image to extract the useful information. Image fusion technology is a good solution to the above problem. Image fusion is the synthesis of a plurality of images acquired through a certain algorithm. The synthesized image has higher reliability, fuzzy fewer and better understandability. It is suitable for human visual and computer detection, classification, recognition, understanding. This dissertation mainly researches the method of image fusion:maximal or minimal pixel selection, aspects of the weighted method, IHS transform method, and wavelet transform method, and value its quality。On the basis of the study, it indicates the development direction of image fusion technology, which provides the new idea for image fusion be used in water conservancy.