摘要
突发性地表异常是地表状态发生的突然变化或变异,具有随机性、发生演化快、发现难、影响大、损失重等特点。对突发性地表异常的及时探测对中国经济社会高质量发展具有战略意义,是国家的重大需求,也是地学和遥感科学面临的重大前沿科学挑战。为此,需要从卫星、载荷和应用三位一体的高度,对传统卫星遥感探测机制进行再造和创新。本文围绕突发性地表异常即时遥感,开展新型卫星体系和探测机制、数据传输和处理模式、异常发现和诊断机制等研究,构建以实时、智能、主动为特征的“通导遥”一体化的突发性地表异常即时遥感探测机制,形成以星间互联、星地互馈、星上处理、星链直传为核心的突发性地表异常即时遥感探测模式,为突发性地表异常即时遥感这一科技难题的突破提供了理论框架与方法支撑。
Abrupt earth’s surface anomalies are sudden changes/variations in the state of the earth’s surface,which exhibit characteristics including strong spatial-temporal randomness,suddenness of occurrence,rapidity of evolution,difficulty of detection,great impact and heavy losses.With the increase of various types of abrupt earth’s surface anomalies caused by natural and human factors in recent years,timely and accurate detection of abrupt earth’s surface anomalies is of strategic significance for the healthy and stable development of China’s society and economy.However,immediate remote sensing detection has long been a major frontier problem in remote sensing science that has yet to be overcome,whose breakthrough relies on theoretical innovation,mode reformation and mechanism reconstruction for traditional remote sensing methodologies in the whole chain of satellite revisit interval,inter-satellite interaction and synergy,satellite-ground transmission delay,anomaly detection mechanism,on-board intelligence paradigm,autonomous mission planning,etc.On this basis,by further integrating advanced satellite and artificial intelligence techniques,we can be able to bridge all these blockages of immediate remote sensing and finally achieve systematic breakthroughs.In this paper,characterized by immediate,intelligent and proactive,we propose an immediate remote sensing detection mechanism for abrupt earth’s surface anomalies on the basis of the integration of“communication-navigation-remote sensing”.The proposed mechanism includes a set of new methodological modes and process models centered around inter-satellite interaction,satellite-ground cross feed,on-board processing,and satellite chain direct transmission.Specifically,six immediate remote sensing detection modes have been presented based on the refactoring of traditional modes.The detection mode of single scene anomaly diagnostic adopts single view imagery produced from current observations to improve immediacy without waiting for satellite revisits or historical data reserves.The detection mode of uploading and mutual feedback transmits lightweight models,rules,data,and other resources to satellites ahead of observation time to avoid transmission delay.The detection mode of parallel onorbit fusion relies on the cooperative of satellites in a local cluster,where each satellite first uses its own single-satellite data to generate intermediate results in parallel without waiting for each other,and then completes the fusion at an optimal node within the cluster.The detection mode of predictive inferencing and passing utilizes causal associations between anomalous events and observed data to try to detect possible surface anomalies in advance,and the prediction can be passed to the next observer.The detection mode of inverse order processing breaks the conventional order of on-orbit image preprocessing by trying to determine the presence of anomalies directly from the unprocessed image data,and only localized data related to anomalies deserve to be processed if necessary.The detection mode of directly downward transmission always maintains available downlinks using communication and navigation satellites,and focuses on the instantaneous generation and compressed expression of content before transmission.The proposed mechanism also includes a system of on-board data intelligence theories and anomaly detection principles.Hopefully the proposed mechanism could provide scientific bases and development paths for the breakthrough of immediate remote sensing.
作者
王桥
张正
唐娉
WANG Qiao;ZHANG Zheng;TANG Ping(Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China;Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China)
出处
《遥感学报》
北大核心
2025年第6期1347-1364,共18页
NATIONAL REMOTE SENSING BULLETIN
基金
国家自然科学基金重大项目(编号:42192584,42192580)。
关键词
突发性
地表异常
即时遥感
探测机制
星群架构
数据智能
模式重构
abruptness
earth’s surface anomalies
immediate remote sensing
detection mechanism
satellite constellation architecture
data intelligence
pattern reconstruction
作者简介
第一作者:王桥,研究方向为环境遥感。E-mail:wangqiao@bnu.edu.cn。
