任务感知的遥感信息服务动态组合方法

Remote sensing technology has been widely recognized for contributing to geospatial information efforts, due to the introduction of multi-platforms and multi-sensors.. But in many cases, raw remotely sensed images are not directly useful without further processing. Because the remotely sensed image processing demands large-scale, collaborative processing and massive storage capabilities, the effect and efficiency of remotely sensed image processing is far from the user's expectation. The emergence of Service Oriented Architecture (SOA) may make this challenge manageable. The SOA allows cooperation of data and process components among different organizational units and supports reusability and interoperability of components through the network to satisfy complex applications. Remotely sensed image processing services encapsulate all processing functions into services and combine them into a service chain to provide a value-added function. Automatic and intelligent service composition, if successful, can be of great value to the geospatial user community..It is a major challenge to select and chain appropriate remotely sensed image processing services for specific application task. Aiming at the correctness and quality of remotely sensed image processing service composition, a task-awared service dynamic composition method is proposed. the main work of this proposal can be summarized as follows: .1) the complexity in user demand, which makes intricate in semantics. Therefore, we propose a uniform semantics description model to establish the connection between user semantics and data services and processing services..2) Aiming at the low correctness in remotely sensed image processing service selection, a novel service selection approach is proposed..3) Aiming at the increasing diversity and real-time requirements of applications, the state-of-art remote sensing processing service composition approaches are facing more and more serious bottlenecks of effectiveness and stability. To break these bottlenecks, a dynamic services composition method is proposed to maximal reuse relative services chain.

随着多平台多传感器遥感技术的发展，遥感数据已经成为空间数据最主要的来源。但是由于缺乏有效的共享和增值应用手段，遥感数据快速获取与信息高效处理能力不足之间的鸿沟日益加剧。面向服务的思想为解决此问题提供了有潜力的途径，将遥感信息及其处理资源均作为服务对待，通过服务的发现与组合，实现分布式环境下的多源遥感信息及其处理资源的按需聚合与高效协同，从而满足更加复杂的应用需求。针对遥感信息处理服务组合的正确性与成功率难题，本申请提出任务感知的遥感信息处理服务动态组合方法，深入研究粗粒度、模糊的高层任务语义与细粒度、精确的底层服务语义之间的映射关系，主要包括：1）基于语义的遥感信息应用任务理解；2）顾及任务语义的遥感信息处理服务选择；3）自适应任务变化的遥感信息处理动态组合。

随着天空地一体化对地观测技术的快速发展，空间信息呈现出海量、高时效、多维度的时空大数据特征。智能化的空间信息服务成为地理信息服务时代的新特色，“以最方便快捷的方式、给最需要的人、提供最需要的服务”的智能化服务方式成为日益迫切的需求。然而，现有的数据服务方式主要依靠人工经验基于简单元数据的检索，耗时长、准确率低。如何面向遥感应用任务，快速准确地聚合有效数据并提供智能服务，已成为遥感信息应用的瓶颈问题和国际前沿。本项目系统研究任务感知的遥感信息智能服务方法，研究内容包括：1）基于语义关联的遥感应用任务理解方法；2）用户模型驱动的遥感信息智能服务方法；3）任务变化下的空间信息服务动态组合方法。.1）由于高层任务语义的抽象性、模糊性和隐含性，特别是针对任务和数据的关联关系描述过于简单，导致缺少计算机自主发现和推理知识的基础，是任务语义理解困难的根本所在。因此，本研究内容以任务-信息-服务的语义关联关系为中心，综合考虑遥感应用任务中隐含的空间特征和属性特征，提出了基于语义关联的遥感应用任务理解方法，设计了基于拓扑和度量关系相结合的关联函数，从时空语义和属性语义进行定量评价，为实现任务语义的准确理解奠定基础。.2）针对遥感信息服务的主动性和智能化问题，提出了用户模型驱动的遥感信息智能服务方法，通过用户模型记录用户的兴趣偏好，主动向用户推荐可能感兴趣的遥感信息及其服务，引导用户发现需要的结果，满足多样化、跨领域的遥感应用任务，实现从“人找服务”发展到“服务找人”的智能阶段。.3）针对遥感信息应用日益增长的多样性和实时性要求，提出了任务变化下的服务动态组合方法，针对用户提出的任务变化，对原有服务链进行重构，而非重新组合。将最小影响域原则引入到服务链重构问题中，采用回归搜索的算法实现服务链重构过程，遥感信息处理服务链重构过程的正确性和完整性通过流程约束和完整性约束保证。