遥感图谱认知理论与计算

发布时间：2019-01-24 19:36

【摘要】：近年来,随着对地观测技术的迅猛发展,卫星遥感领域逐渐进入了大数据时代。针对当前遥感应用的需求和特点,开展与视觉认知相结合的高分辨率遥感认知理论与方法研究是可行且必要的。在此背景下,受地学信息图谱思想启发,本文对遥感认知领域的图谱问题进行了研究,系统地提出了遥感图谱认知理论与计算方法论,旨在规范高分辨率遥感信息提取流程,构建精细化、定量化、智能化、综合化相结合的遥感信息解译模型。整套方法体系由横向"自底向上的分层抽象"和纵向"自顶向下的知识迁移"2个方向上的认知计算组成,分别对应了"由谱聚图"、"图谱协同"和"认图知谱"3大图谱转化过程。论文涉及的概念、基本思想、关键技术及难点问题进行了重点分析,强调综合利用大数据、逐步融入知识来实现不同层次的遥感知,以期为数据源极大丰富条件下的遥感信息解译提供新的视角。
[Abstract]:In recent years, with the rapid development of Earth observation technology, the field of satellite remote sensing has gradually entered the era of big data. According to the demand and characteristics of remote sensing application, it is feasible and necessary to study the theory and method of high-resolution remote sensing cognition combined with visual cognition. Under this background, inspired by the idea of geoscience information atlas, this paper studies the map problem in the field of remote sensing cognition, and puts forward the cognitive theory and computational methodology of remote sensing map systematically, aiming at standardizing the extraction process of high resolution remote sensing information. A detailed, quantitative, intelligent and integrated remote sensing information interpretation model is constructed. The whole method system is composed of horizontal "bottom-up hierarchical abstraction" and vertical "top-down knowledge transfer" in the two directions of cognitive computation, corresponding to "spectral aggregation", respectively. The transformation process of three kinds of atlas was "Tupu Synergy" and "recognition Map knowledge Spectrum". The concepts, basic ideas, key technologies and difficult problems involved in the paper are emphatically analyzed, emphasizing the comprehensive use of big data and the gradual integration of knowledge to realize different levels of remote sensing knowledge. In order to provide a new perspective for the interpretation of remote sensing information under the condition of abundant data sources.
【作者单位】：中国科学院遥感与数字地球研究所遥感科学国家重点实验室;长安大学理学院数学与信息科学系;浙江工业大学计算机学院;
【基金】：国家自然科学基金项目(41271367) 国家高分辨率对地观测系统重大专项(03-Y30B06-9001-13/15-01) 广西科学研究与技术开发计划项目(14125008-1-6) 长安大学中央高校基本科研业务费专项资金基础研究项目(310812161010)
【分类号】：P237

【相似文献】