طراحی و پیاده سازی سنجش از راه دور الگوریتم های تجسمی در حال ویرایش جریان کار و پردازش بستر های نرم افزاری

The visual workflow operate-mode can greatly improve the programming efficiency. Visual workflow technology which is brought into the design of remote sensing algorithms can not only control the processing of remote sensing algorithms flexibly, but also greatly improve the reusability of remote sensing algorithms. The principle of remote sensing algorithms visualization lies in the decomposition of remote sensing algorithms. By means of packaging the decomposed algorithms and showing them visually, the algorithm can be processing more efficient. In addition, through the visual platform, the decomposed algorithms can be restructured into new algorithm flow which is composed by a certain number of refined algorithms. Therefore, the composed algorithm can be accelerated through the parallel processing method. Currently, multi-core processors programming is popular. Multi-core processing has been becoming an emerging and high efficient computer technology. The platform will accelerate the processing of the algorithm flow further by using multi-core processing technology.

As the field of remote sensing image processing is becoming more widespread, number of remote sensing processing algorithms are also increasing rapidly. Many remote sensing image processing algorithms include public modules which have same functions, but when the public modules are needed ina new kind of algorithm, they are always written again. This work mode greatly reduces the programming efficiency. The remote sensing image processing algorithms' main characteristic is image processing, almost every algorithm is a series of image processing to generate a new image. According to this feature, we can encapsulate the image processing into a public module by designing the interface of input and output parameters, reconstructing the algorithms by workflow and providing the interfaces of workflow restructuring in the form of visualization. There will be the following advantages[1,2,3]: • Increased code reusability, and with the increasing of public modules, a written function can be directly used in the future • Visual workflow technology enables the algorithms to be shown in a clear way. Visual operation like dragging will also replace the bring code-editing work mode into the front, and greatly improve work efficiency • As the algorithm is refined, there will be a lot of parallel computing. The use of parallel computing technology will accelerate the overall speed of the algorithm and reduce computing time The platform analysis the remote sensing image processing thoroughly. According to which features packaged the atoms processing involved in algorithms, the platform abstracted the algorithms and developed a series of specifications describing the algorithms, making the platform analysis the I/O and running format and other related information of the encapsulated algorithms based on the specification. The platform allows users to customize algorithm flows in form of visual workflow, which implements the scheduling and computing of algorithms in parallel multi-core processing technology.

Through the research and application of visual workflow customization in the field of remote sensing algorithm, the customized algorithm flow can bring about huge advantages. Firstly, the reusability of algorithm increased greatly which improve the efficiency of research work; Secondly, the flexibility of the production process also significantly increased, algorithm flows can be customized reasonably according to data; Finally, the procedural algorithms bring about parallel processing which can effectively reduce the computing time and rational use of computing resources after experiment..