CASIA OpenIR  > 毕业生  > 博士学位论文
Alternative TitleResearch on Semantic Knowledge of Remote Sensing Image Understanding
Thesis Advisor杨一平
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline计算机应用技术
Keyword遥感图像理解 概念知识树 遥感图像理解知识体系 语义知识 Image UnderstandIng In Remote sensIng Concept Knowledge Tree Knowledge System On Remote Sensing Image Understanding Semantic Knowledge
Abstract现代空天飞行技术和传感器技术的高速发展,为人类提供了极为丰富的遥感图像数据。为了高效地从遥感图像中提取有效信息,利用计算机系统解译遥感图像,即进行遥感图像理解,已成为遥感信息获取技术研究的热点。然而已有图像理解技术和系统在实用性上与人们预期的效果还有相当的距离。人工判读遥感图像的过程与常规图像理解算法的一个重要区别在于人类具有综合应用视觉经验知识与视觉背景知识的能力。本论文就此开展相关研究与系统设计。 在分析遥感图像理解领域知识内容的基础上,通过构建遥感图像理解领域知识体系,为遥感图像理解各个过程提供知识指导和支撑,以提高遥感图像理解系统的准确性和实用性。本文主要研究工作和创新点包括: [1]遥感图像领域知识表示体系的研究 本文采用概念知识树作为遥感图像领域知识的表达模型。本文在已有研究成果的基础上,对该表达体系的应用做了进一步的研究,如相似度计算和知识推理。 [2]遥感图像理解知识体系的构建 在对遥感图像理解领域知识分析和归纳的基础上,采用概念知识树知识表达模型对遥感图像领域知识进行语义化描述,构建遥感图像理解知识体系。通过对知识的语义化描述,使计算机能够以符号推理的方式完成诸如算法调用,图像特征转换,图像管理等相关操作。语义化描述不仅为知识提供了统一的描述和应用的接口,也为知识的管理和更新提供便利。 [3] 遥感图像理解知识体系的应用 在遥感图像理解知识体系的基础上,以遥感图像理解处理流程为主线,利用遥感图像理解处理流程与知识体系中不同类型知识之间的相互联系,为遥感图像处理流程提供知识指导和支撑。采用流程推理和算法推理相结合的层次推理方式,辅助系统自主地根据当前处理的状态,规划出合理的处理流程序列,并选择恰当的知识和信息完成相应处理任务。
Other AbstractWith the rapid development of modern aerospace flying and sensor technology, a great amount of remote sensing images have been available for people. In order to extract the information from remote sensing images efficiently, image understanding in remote sensing has drawn increasing attention from researcher. However, there is a considerable gap between the practicality of existing technology and that of expected. People have the good ability to integrate visual experience and background knowledge, which differentiates traditional image understanding algorithms from manual image interpretation. According to this, we carry out research on architecture design and relevant technologies. A knowledge system on remote sensing images understanding is proposed in the paper based on the analysis of domain knowledge in remote sensing understanding. It can provide knowledge guidance and support for the process of remote sensing image understanding and improve the accuracy and practicability. The main contributions and novelties are summarized as follows. [1] The Research on Knowledge Representation in Remote Sensing Image The concept knowledge tree based on the Concept is adopted as the core model for knowledge representation and formalization. Based on the precious work , we make the further studies on its application, such as semantic similarity computing, and knowledge reasoning. [2] Construction of Knowledge System on Remote Sensing Understanding Analyzing the domain knowledge in the field of remote sensing, the knowldge system on remote sensing understanding is built with the concept knowledge tree as its core model of knowledge representation and formalization. Besides, by representing the knowledge semantically, we can implement many operations with symbolic reasoning, such as the algorithm invoking, feature transformation, and image management. Semantic representaion can not only provide the unified the interface in knowledge representation and application, but also offer convenience in knowledge management. [3] Application of Knowledge System on Remote Sensing Understanding Taking the process of remote sensing image understanding as the clue, a layering reasoning method is adopted to provide knowledge guidance and support for the process of remote sensing image understanding, based on the correlation between process and different type of knowledge in knowledge system on remote sensing understanding. It can help the computer to ...
Other Identifier200618014629075
Document Type学位论文
Recommended Citation
GB/T 7714
于海涛. 遥感图像理解中语义知识应用研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2009.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20061801462907(2264KB) 暂不开放CC BY-NC-SA
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[于海涛]'s Articles
Baidu academic
Similar articles in Baidu academic
[于海涛]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[于海涛]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.