遥感图谱认知理论与计算

doi:10.3724/SP.J.1047.2016.00578

Abstract

Abstract:

In recent years, with the rapid development of earth observation technologies, remote sensing using the satellites has gradually entered the era of big data. Facing the current demands and characteristics of remote sensing applications, it is feasible and necessary to explore the theories and methods of high-spatial-resolution remote sensing cognition with the cooperation of visual cognition. In this context, we are inspired by Geo-informatic-Tupu and intend to study the spatial-spectral cognition of remote sensing. This paper systematically presents the theory and calculation methodology for the spatial-spectral cognition of remote sensing, and expects to standardize the processes of remote sensing information extraction, as a result to further build a sophisticated, quantitative, intelligent and integrated model for the remote sensing information interpretation. The whole methodology contains two directions' cognitive calculation, namely horizontal "bottom-up hierarchical abstraction" and longitudinal "top-down knowledge transfer". These two steps are corresponded with three principal Spatial-Spectral transformation processes, which are summarized as "extracting spatial maps based on clustering pixels' spectrum", "coordinating spatial-spectral features" and "understanding attributes through the recognition of known diagram". Our study focuses on the analysis of the involved concepts, the basic idea, the key technologies and their existing difficulties, and emphasizes on the utilization of big data and gradually the application of integrated knowledge to achieve different levels of remote sensing cognition. Through these approaches, we expect to provide a new perspective for the remote sensing interpretation with the adoption of big data resources.

Key words: spatial-spectral cognition of remote sensing, extracting spatial maps based on clustering pixels' spectrum, coordinating spatial-spectral features, understanding attributes through the recognition of known diagram, bottom-up hierarchical abstraction, top-down knowledge transfer

LUO Jiancheng,WU Tianjun,XIA Liegang. The Theory and Calculation of Spatial-spectral Cognition of Remote Sensing[J].Journal of Geo-information Science, 2016, 18(5): 578-589.DOI:10.3724/SP.J.1047.2016.00578

Figures/Tables 12

Fig. 1

Fig. 2

Fig. 3

Tab. 1

Fig. 4

Tab. 2

Fig. 5

Tab. 3

Fig.6

Fig. 7

Fig.8

Tab. 4

References 46

[1]	周成虎. 遥感影像地学理解与分析[M].北京:科学出版社,1999.
	[ Zhou C H.Geo-understanding and analysis of remote sensing image[M]. Beijing: Science Press, 1999. ]
[2]	Crick F.The astonishing hypothesis: the scientific search for the soul[J]. The Journal of Nervous and Mental Disease, 1996,184(6):384. doi: 10.1001/jama.1994.03520070087050.
[3]	Hubel D H, Wiesel T N.Receptive fields, binocular interaction and functional architecture in the cat's visual cortex[J]. The Journal of Physiology, 1962,160(1):106. doi: 10.1113/jphysiol.1962.sp006837 pmid: 14449617
[4]	寿天德. 视觉信息处理的脑机制(第2版)[M].合肥:中国科学技术大学出版社,2010.
	[ Shou T D.The brain mechanism of visual information processing (the second edition)[M]. Hefei: University of Science & Technology China Press, 2010. ]
[5]	黄凯奇,谭铁牛.视觉认知计算模型综述[J].模式识别与人工智能,2013,26(10):951-958.
	[ Huang K Q, Tan T N.Review on computational model for vision[J]. Pattern Recognition and Artificial Intelligence, 2013,26(10):951-958. ]
[6]	Marr D.A computational investigation into the human representation and processing of visual information[M]. San Francisco, CA: W. H. Freeman and Company, 1982.
[7]	Konen C S, Kastner S.Two hierarchically organized neural systems for object information in human visual cortex[J]. Nature Neuroscience, 2008,11(2):224-231. doi: 10.1038/nn2036 pmid: 18193041
[8]	Karklin Y, Lewicki M S.Emergence of complex cell properties by learning to generalize in natural scenes[J]. Nature, 2009,457(7225):83-86. doi: 10.1038/nature07481 pmid: 19020501
[9]	Li N, Dicarlo J J.Unsupervised natural experience rapidly alters invariant object representation in visual cortex[J]. Science, 2008,321(5895):1502-1507. doi: 10.1126/science.1160028 pmid: 18787171
[10]	Hirabayashi T, Takeuchi D, Tamura K, et al.Microcircuits for hierarchical elaboration of object coding across primate temporal areas[J]. Science, 2013,341(6142):191-195. doi: 10.1126/science.1236927 pmid: 23846902
[11]	Riesenhuber M, Poggio T.Hierarchical models of object recognition in cortex[J]. Nature Neuroscience, 1999,2(11):1019-1025. pmid: 328800326597506077310065207529222322221052634381960969757863001613
[12]	Tanaka K.Inferotemporal cortex and object vision[J]. Annual Review of Neuroscience, 1996,19(1):109-139. doi: 10.1146/annurev.ne.19.030196.000545 pmid: 8833438
[13]	Hinton G E.Learning to represent visual input[J]. Philosophical Transactions of the Royal Society B: Biological Sciences, 2010,365(1537):177-184. doi: 10.1098/rstb.2009.0200 pmid: 20008395
[14]	Bengio Y, Lamblin P, Popovici D, et al.Greedy layer-wise training of deep networks[J]. Advances in Neural Information Processing Systems, 2007,19:153.
[15]	Bar M.The proactive brain: using analogies and associations to generate predictions[J]. Trends in Cognitive Sciences, 2007,11(7):280-289. doi: 10.1016/j.tics.2007.05.005
[16]	Engel A K, Fries P, Singer W.Dynamic predictions: oscillations and synchrony in top-down processing[J]. Nature Reviews Neuroscience, 2001,2(10):704-716. doi: 10.1038/35094565 pmid: 11584308
[17]	Melloni L, van Leeuwen S, Alink A, et al. Interaction between bottom-up saliency and top-down control: how saliency maps are created in the human brain[J]. Cerebral Cortex, 2012,22(12):2943-2952. doi: 10.1093/cercor/bhr384 pmid: 22250291
[18]	Navalpakkam V, Itti L.An integrated model of top-down and bottom-up attention for optimizing detection speed[C]. 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006:2049-2056.
[19]	Friston K.A theory of cortical responses[J]. Philosophical Transactions of the Royal Society B: Biological Sciences, 2005,360(1456):815-836. doi: 10.1098/rstb.2005.1622
[20]	Kveraga K, Ghuman A S, Bar M.Top-down predictions in the cognitive brain[J]. Brain and cognition, 2007,65(2):145-168. doi: 10.1016/j.bandc.2007.06.007
[21]	Iounousse J, Er-Raki S, Chehouani H.Using an unsupervised approach of probabilistic neural network (PNN) for land use classification from multitemporal satellite images[J]. Applied Soft Computing, 2015,30:1-13.
[22]	Demir B, Erturk S.Clustering-based extraction of border training patterns for accurate SVM classification of hyperspectral images[J]. IEEE Transactions on Geoscience and Remote Sensing Letters, 2009,6(4):840-844. doi: 10.1109/LGRS.2009.2026656
[23]	Knight A, Tindall D, Wilson B.A multitemporal multiple density slice method for wetland mapping across the state of Queensland, Australia[J]. International Journal of Remote Sensing, 2009,30(13):3365-3392. doi: 10.1080/01431160802562180
[24]	Giada S, De Groeve T, Ehrlich D, et al.Information extraction from very high resolution satellite imagery over Lukole refugee camp, Tanzania[J]. International Journal of Remote Sensing, 2003,24(22):4251-4266. doi: 10.1080/0143116021000035021
[25]	Song M, Civco D, Hurd J.A competitive pixel-object approach for land cover classification[J]. International Journal of Remote Sensing, 2005,26(22):4981-4997.
[26]	Zhou Y, Qiu F.Fusion of high spatial resolution WorldView-2 imagery and LiDAR pseudo-waveform for object-based image analysis[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015,101:221-232. doi: 10.1016/j.isprsjprs.2014.12.013
[27]	Li X, Shao G.Object-based land-cover mapping with high resolution aerial photography at a county scale in midwestern USA[J]. Remote Sensing, 2014,6(11):11372-11390.
[28]	Voltersen M, Berger C, Hese S, et al.Object-based land cover mapping and comprehensive feature calculation for an automated derivation of urban structure types at block level[J]. Remote Sensing of Environment, 2014,154:192-201. doi: 10.1016/j.rse.2014.08.024
[29]	Blaschke T.Object based image analysis for remote sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2010,65(1):2-16. doi: 10.1016/j.isprsjprs.2009.06.004
[30]	宫鹏,黎夏,徐冰.高分辨率影像解译理论与应用方法中的一些研究问题[J].遥感学报,2006,10(1):1-5. doi: 10.11834/jrs.20060101
	[ Gong P, Li X, Xu B.Interpretation theory and application method development for information extraction from high resolution remotely sensed data[J]. Journal of Remote Sensing, 2006,10(1):1-5. ] doi: 10.11834/jrs.20060101
[31]	明冬萍,骆剑承,沈占锋,等.高分辨率遥感影像信息提取与目标识别技术研究[J]. 测绘科学, 2005,30(3):18-20. doi: 10.3771/j.issn.1009-2307.2005.03.004
	[ Ming D P, Luo J C, Shen Z F, et al.Research on information extraction and target recognition from high resolution remote sensing image[J]. Science of Surveying and Mapping, 2005,30(3):18-20. ] doi: 10.3771/j.issn.1009-2307.2005.03.004
[32]	Liang S.Quantitative remote sensing of land surfaces[M]. New Jersey: John Wiley & Sons, 2005.
[33]	Li X, Strahler A H.Geometric-optical bidirectional reflectance modeling of the discrete crown vegetation canopy: effect of crown shape and mutual shadowing[J]. IEEE Transactions on Geoscience and Remote Sensing, 1992,30(2):276-292. doi: 10.1109/36.134078
[34]	Jupp D L, Strahler A H.A hotspot model for leaf canopies[J]. Remote Sensing of Environment, 1991,38(3):193-210.
[35]	Nilson T, Kuusk A.A reflectance model for the homogeneous plant canopy and its inversion[J]. Remote Sensing of Environment, 1989,27(2):157-167. doi: 10.1016/0034-4257(89)90015-1
[36]	童庆禧,张兵,郑兰芬.高光谱遥感:原理,技术与应用[M].北京:高等教育出版社,2006.
	[ Tong Q X, Zhang B, Zhen L F.Hyperspectral remote sensing: principle, technology and application[M]. Beijing: Higher Education Press, 2006. ]
[37]	李小文,王绵地.植被光学遥感模型与植被结构参数化[M].北京:科学出版社,1995.
	[ Li X W, Wang J D.Optical remote sensing model of vegetation and structure parameterization of vegetation[M]. Beijing: Science Press, 1995. ]
[38]	李小文. 地球表面时空多变要素的定量遥感项目综述[J].地球科学进展,2006,21(8):771-780.
	[ Li X W.Review of the project of quantitative remote sensing of major factors for spatial-temporal heterogeneity on the land surface[J]. Advances in Earth Science, 2006,21(8):771-780. ]
[39]	陈述彭. 地学信息图谱探索研究[M].北京:科学出版社,2001.
	[ Chen S P.Exploratory research on geo-informatic Tupu[M]. Beijing: Science Press, 2001. ]
[40]	陈述彭,岳天祥,励惠国.地学信息图谱研究及其应用[J].地理研究,2000,19(4):337-343.
	[ Chen S P, Yue T X, Li H G.Studies on geo-informatic Tupu and its application[J]. Geographical Research, 2000,19(4):337-343. ]
[41]	齐清文,池天河.地学信息图谱的理论与方法研究[J].地理学报,2001,56(z1):8-18.
	[ Qi Q W, Chi T H.Research on the theory and method of geo-informatic Tupu[J]. Geographical Research, 2001,56(z1):8-18. ]
[42]	廖克. 地学信息图谱的探讨与展望[J].地球信息科学,2002,4(2):14-20.
	[ Liao K.The discussion and prospect for geo-informatic Tupu[J]. Geo-information Science, 2002,4(2):14-20. ]
[43]	骆剑承,周成虎,沈占锋,等.遥感信息图谱计算的理论方法研究[J].地球信息科学学报,2009,11(5):5664-5669. doi: 10.3969/j.issn.1560-8999.2009.05.017
	[ Luo J C, Zhou C H, Shen Z F, et al.Theoretic and methodological review on sensor information Tupu computation[J]. Journal of Geo-information Science, 2009,11(5):5664-5669. ] doi: 10.3969/j.issn.1560-8999.2009.05.017
[44]	李德仁,童庆禧,李荣兴,等.高分辨率对地观测的若干前沿科学问题[J].中国科学:地球科学,2012,42(6):805-813. doi: 10.1007/s11783-011-0280-z
	[ Li D R, Tong Q X, Li R X, et al.Some frontier science problems of high-resolution earth observation[J]. Science China: Earth Sciences, 2012,42(6):805-813. ] doi: 10.1007/s11783-011-0280-z
[45]	孙显,付琨,王宏琦.高分辨率遥感图像理解[M].北京:科学出版社,2011.
	[ Sun X, Fu K, Wang H Q.High resolution remote sensing image understanding[M]. Beijing: Science Press, 2011. ]
[46]	吴田军,骆剑承,夏列钢,等.迁移学习支持下的遥感影像对象级分类样本自动选择方法[J].测绘学报,2014,43(9):903-916. doi: 10.13485/j.cnki.11-2089.2014.0163
	[ Wu T J, Luo J C, Xia L G, et al.An automatic sample collection method for object-oriented classification of remotely sensed imageries based on transfer learning[J]. Acta Geodaetica et Cartographica Sinica, 2014,43(9):908-916. ] doi: 10.13485/j.cnki.11-2089.2014.0163

关键技术	常用实现算法	存在问题与难点	发展趋势
分割	基于边缘、区域（阈值、图论、能量泛函）的多种分割算法	一般的分割算法对地物复杂多变的遥感影像适用性较低,且数据量巨大的高分影像使分割效率大幅下降,如何提升分割的效率	发展复杂环境下高分辨率影像的多尺度快速聚合技术（多尺度：大数据综合处理）
分割	均值漂移、分水岭等多尺度分割算法	如何设置合适的尺度集来合理地表达地物的异尺度特征,实现成功抽取对象的目标^[44],即如何提升分割的精度
聚类、分类	Kmeans、ISODATA等非监督分类算法的聚类,基于SVM、神经网络、决策树、随机森林等监督分类算法的像元级分类或对象级分类	像元级分类造成的椒盐噪声的影响;对象级分类受分割算法问题的影响,存在对象分离不合理的问题;对象合并规则的设定
人工矢量编辑	目视勾画	矢量编辑工具的智能化程度,减少人工操作量

关键技术	常用实现算法	存在问题与难点	发展趋势
特征分析	光谱、空间、时间、地域等多源“图-谱”特征的提取与优选算法	地物特征的计算具有不确定性,造成后续的分类存在一定的错分率	① 构建“影像-结构-演化”紧致结构的多特征表达模型（多特征：异构时空特征表达） ② 建立高空间、高光谱与高时间分辨率大数据协同计算（多维度：多源数据协同处理）
特征分析	光谱、空间、时间、地域等多源“图-谱”特征的提取与优选算法	多时相获取的遥感数据是非平稳信号,地物特征具有不一致性,当协同时序特征时,需对数据进行有有效、可靠的滤波去噪、几何配准、辐射校正等一致性预处理
分类	SVM、神经网络、决策树、随机森林等监督分类算法以及自训练算法、生成模型、图论方法、多视角算法等半监督分类算法	地物外在特征不能描述其本质特征,典型特征难以确定,限制了分类的精细度（即地物可分性和分类精确性）^[45]
分类	SVM、神经网络、决策树、随机森林等监督分类算法以及自训练算法、生成模型、图论方法、多视角算法等半监督分类算法	训练样本的采集是费时费力的步骤,如何在少量样本或无样本条件下实现高精度的分类、提升分类的自动化与智能化程度^[46]

关键技术	常用实现算法	存在问题与难点	发展趋势
传统地物识别技术	分割、特征提取、分类等算法	同表1、表2的说明	① 建立波谱、视觉、环境和空间知识的逐步融合模型（多知识：遥感与GIS一体化） ② 发展“谱相”及“图形”间螺旋式认知的自适应计算模型（多模型：流程化、自动化）
迁移学习	实例迁移、特征迁移、参数迁移、关系知识迁移等算法	知识的形式化以及如何在不同时间、空间、尺度的先验知识中进行去伪存真以及与任务的关联
GIS空间分析	空间关系分析、叠置分析、网络分析、缓冲分析、地统计分析等算法	GIS数据与所需提取遥感信息的关联分析,以及各种GIS空间分析方法中存在的限制问题
语义推理	特征编码或表达（视觉词袋）、主题模型（概率潜语义分析pLSA、潜在狄利克雷分析LDA）、深度特征学习等算法	底层特征与高层语义间的鸿沟^[8]

来源	形态		表达	层次	计算	存储	迁移	应用
遥感知识	视觉知识		影像(对象)特征：色调、几何形状、纹理等波谱和和空间形态特征等	低	图像处理统计	对象表达特征库(表)	特征迁移	由谱聚图
遥感知识	参数知识		影像参数：成像时间、角度、传感器性能等	低	查询查阅	参数文档	参数迁移	由谱聚图
地域知识	波谱知识		地物波谱库、纯端元地物样本库等	中	采集测量	波谱库(表)	特征迁移	由图聚图、图谱协同
地域知识	环境知识		坡度、坡向等DEM高程相关信息;温度、湿度等土地资源信息;	中	采集测量经验总结	带有环境特征的栅格/矢量图斑、If… Then…规则	特征/关系知识迁移	由图聚图、图谱协同
解译知识	模型知识		物理量反演模型、专题指数计算模型、定理分析模型、分类体系等	高	实验分析模型解算	公式、文档	参数迁移	图谱协同
	物候知识		作物生长演变地学规律(季相变化物候特征)、景观格局演变规律等	高	时序分析	特征变化曲线	参数迁移
	空间知识	空间分布	地类的地理空间分布：土地利用/覆盖解译图的图斑、空间样本位置库	高	目视/机器解译	带有空间位置和地类属性的栅格/矢量图斑	关系知识迁移
	空间知识	空间关系	阴影等相邻、相交、包含、方向等空间关系/格局知识	高	GIS 空间分析	语义网络、拓扑	关系知识迁移

The Theory and Calculation of Spatial-spectral Cognition of Remote Sensing

RichHTML

PDF (PC)

Like

Knowledge

Cited

Abstract

Cite this article

share this article

Figures/Tables 12

References 46

Related Articles 1

Metrics

Comments

Recommended 0