Along with the development of the modern remote sensing technology, the acquired remote sensing image data gets more and more abundant, so the primary obstruction of the application of the remote sensing technology in the future is no longer the shortage of the image resource, but the capacity that how we can get more abundant, more useful and more credible information from the image resource. Multi-sensors remote sensing image fusion is an important apart of the information acquisition of the ground observation and also an important approach to resolve the problem of the mass remote sensing image data. The processing speed becomes a key point of an algorithm if can be in general use. In this paper we designed a high-pass filtering fusion algorithm of remote sensing image data in GPU (Graphics Processing Unit) using the programmability of GPU, which is a parallel vector processor. The result shows that the algorithm runs on a GPU is much faster than the CPU-based algorithm in the case of large data. And with the volumes of fusion images data getting bigger the advantage of the velocity on GPU is more obvious then on CPU.
展开▼
