Font Size: a A A

The Pivotal Data Processing Of Using And Transformation Of Outskirts' Soil

Posted on:2009-12-25Degree:MasterType:Thesis
Country:ChinaCandidate:J Y QuanFull Text:PDF
GTID:2199360272960924Subject:Geodesy and Survey Engineering
Abstract/Summary:PDF Full Text Request
The land's utilization and change takes one of present global change research core questions, had reflected the most close relationship between the nature and the humanities .Now the remote sensing technique already became the region even whole world criterion the land utilization change research effective method, use the remote sensing images to carry on the land utilization change analysis to be possible to use in the remote sensing phantom fully the rich terrain feature information, realizes and carries on the different time land utilization change to the land type classification the comparison.In the remote sensing image classification's system, the traditional BP neural network classification, has the training speed slowly, not to be easy to restrain best and so on shortcomings, has the structure from the organization neural network simply practical, the training speed to be quick, astringency good and so on merits, can carry on the remote sensing land classification well.Traditional application land change's computer software processing is complex, needs to compile certain procedures, the work is troublesome, although now has softwares and so on ERDAS, ENVI, but more or less has the shortcoming or insufficient. MATLAB the high performance numerical calculus and the visualization integrates in together, and has provided the built-in function which massive may transfer directly, contains toolbox like wavelet theory which, neural network toolbox many functions vary and so on.Through to Qingdao and Tai'an land utilization change's analysis, uses the MATLAB imagery processing software and based on the wavelet fusion theory, carries on processing and so on image pretreatment, image fusion, image classification from the organization neural network classification to two city different time remote sensing image, promulgates the suburb land utilization change the tendency, to establish the city data model to provide the data foundation.This article obtains the Shandong Province foundation geography information and the digitized technology key laboratory opening fund financing projects "based on the neuron network city land utilization change simulation theory and the method" (SD060805) support.
Keywords/Search Tags:Remote sensing image, Land utilization change, MATLAB, SOFM
PDF Full Text Request
Related items