摘要:The application of remote sensing data as a new source to update GISis increasingly popular in recent years.However,using GIS to improve classifica-tion of images is neglected in most of current approaches.Although attempts havebeen made in using texture and context information,the effects have limitation。The paper proposes a new method to integrate GIS so that the shape information,which is frequently used in visual interpretation,can be easily employed to improvethe performance of classification。The confusion of classification can be significantlyclarified after shape information is involved。
关键词:Shape information;Geographical information system;remote sensing;classification
摘要:This paperpresents an estimating method of Foliage Area Volume De-nsity(FAVD)and Leaf-Area Index(LAI)of tree crown by combining computedtomography and vegetation gap probability model.The theory and method of Multi-viewing data acquisition,crown 3D shape reconstruction,and FAVD estimation areintroduced.The measurement tree architectural data are used in validation。
摘要:Using different sort and time remote sensing image,a lot of hydrologyand environment information and the variation law can be get and used not only toselect the best location of bridge crossing in new railway line,but also to analysethe cause of flood disaster and put forwared the dredging measure in old railwaybridge crossing。
摘要:This paper presents a knowledge-based classification process of an itera-tive quadtree splitting and sharing a common knowledge data-base,This methodprovides two functions:(1)the coordination of parallel operated processes;(2)theincorporation of various types of knowledge into the different levels of decisionmaking。
摘要:With the development of remote sensing technology,the conventionalmanual approach for image interpretation can not meet the increasing needs of more andmore information。The incorporation of artificial intelligence into remote sensing,such as an building for the expert system of automatic recognition which can greatlyenhance the efficiency of image interpretation,is an important approach to sovlingthis problem。The authors take the expert system as the guide,used the digitalimage processing and pattern recognition technique as the methods of evidence ac-quisition,and combined the quantitative with the qualitative abstraction so that wecould build an expert system for automatic recognition of fault structure。This ex-pert system is only an attempt to develop an approach to the automatic recognitionof remote sensing images instead of conventional way.The basic structure of thisexpert system as shown in Fig.1 includes the following parts:1.Knowledge-base:It consists of facts and expert’s knowledge associated withrecongnition of fault image. The knowledge is presented in the form of rules。2.Inference engine:It utilizes the knowledge base, and its reasoning strategyis step forward。3.Knowledge base management system:It manages the knowledge base by automa-tically organizing, propagating,modifying and deleting the stored knowledge。4.User interface:It fulfils the communication between user and the expert system。5.Automatic recognition system:It includes several subsystems such as abstractionand examination of linear body and triangle plane of fault structure,lake and thedistribution of all the linear signs。This expert system is completed on IBM-PC with C language。Whose user interfaceis friendly,there by leading to its practical applications。
摘要:A triangular hierarchical data structure has been proposed as the basisfor a global geographical information system,In this paper we briefly review onesuch scheme based on recursive subdivision of an octahedron,and conversion algo-rithms to and from latitude/longitude。Schemes for representing point,line and areaobjects on the earth’s surface are described.We present algorithms for identifyingtriangle neighbors,region filling and object dilation。
摘要:On the basis of ecological characteristics,the pine leaves damaged bypine moth were classified on five levels and their reflectance and chlorophyll con-tents were measured and analyzed in this paper.The results indicate that damageis becoming more serious,①the chlorophy11 content is down,②the reflectance at 550nm,the difference between the infrared shoulder reflec-tance and the lowest reflectance in red range,and the largest value of the firstderivative in red reflectance edge,all come down,③the reflectance at 630nm goes up,④blue shift of the red edge spectra and red shift of the chlorophyll reflec-tance peak are obvious。With the stepwize discrimination analysis,it was analyzed and was proved thatthe characteristic parameters of the fine spectra of the damaged pine leaves havestronger capability than the single-reflectance parameters in the three bands ofgreen,red and infrared for discrimination and classification of the damage,Thispaper also gave out a discrimination model for the damage forecast with the remotesensing approach。
摘要:Multicategory recoginition is very important to practicality of remotely-sensed image classification。This paper has presented a mixed classification methodintegrating multi-layer neuron network and unsupervised classification algorithm,Atthe first step,a multi-layer neuron network is used and the result serves as inputfor the unsupervised classification at the second step.The number of patterns thatcan be recognized is increased from 10 to 30.Applying this algorithm to SPOTremotely-sensed image recognition shows it can adapt the requiry of multicategoryrecognition.
摘要:This paper presents the approaches and principles for building Fuzzyanalysis model of soil erosion intensity in hills and gullies of loess regions.Withthe theories of Fuzzy Subset,the author builds the attribute function of soil ero-sion intensity,and designs the computing table for soil erosion intensity Using thecomputing table,the author obtains the indexes of soil erosion intensity and gradesand classsifies the intensity of soil erosion with the principles of Fuzzy compatibi-lity degree.With the model the author makes the test researches on soil erosion inAnsai County,north of Shanxi Province。The test results show that the model hasa high prediction precision, and can objectively reflect the dynamic states of soilerosion intensity in loess regions。