The strong demands for satellite images are increasing not only in professional fields, but also in the non-professionals.
But the online map services with up-to-date satellite images can serve few demands. One challenge is how to provide
online processing service, which need to handle real-time online data-intensive geospatial computation and visualization.
Under the background of the development of cloud computing technology, the problem can be figured out partly. The
other challenge is how to implement user-customized online processing without professional background and
knowledge. An online self-service processing system of ZY-3 Satellite images is designed to implement an on-demand
service mode in this paper. It will work with only some simple parameters being set up for the non-professionals without
having to care about the specific processing steps. And the professionals can assemble the basic processing services to a
service chain, which can work out a more complex processing and a better result. This intelligent self-service online
system for satellite images processing, which is called the prototype of satellite image cloud service in this paper, is
accelerated under the development of cloud computing technology and researches on data-intensive computing. To
realize the goal, the service mode and framework of the online self-service processing system of ZY-3 Satellite images
are figured out in this paper. The details of key technologies are also discussed, including user space virtualization
management, algorithm-level parallel image processing, image service chain construction, etc. And the experimental
system is built up as a prospective study of image cloud services.