Font Size: a A A

Road Information Extraction And Application Based On GF-1 Satellite Image

Posted on:2021-05-16Degree:MasterType:Thesis
Country:ChinaCandidate:W J LiFull Text:PDF
GTID:2392330623473549Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Domestic satellite images with high resolution have been widely used in agriculture,surveying and mapping,urban and rural construction,measurement statistics and other fields of production and life.As a link between regions,the development of roads determines the level of economic development and construction of a country.At the same time,roads are crucial to national security and military strategic layout.The popularity of high-resolution images provides powerful data support for extracting more types of road targets.Based on the Gaofen-1 satellite(GF-1)remote sensing image as the data source,this paper discusses the extraction of different roads under the different environment backgrounds,and analyzes the spatial and temporal characteristics of roads in Yajiang county based on the extraction results.The main contents and results are as follows:(1)The necessary preprocessing and characteristic band analysis are carried out for the images in the study area.The preprocessing mainly includes radiometric calibration,atmospheric correction,orthophoto correction and image fusion.According to the calculation results of Optimum index Factor(OIF),the best band combination of GF-1 images is 431.(2)In this paper,the types and characteristics of roads,especially the road types in Chengdu and Yajiang County are analyzed comprehensively,and the extracted road types in Chengdu and Yajiang County are determined;(3)This paper discusses the road segmentation in Chengdu and Yajiang County,obtains the best segmentation scale,analyzes the spectral characteristics,texture characteristics and geometric characteristics of the roads in Chengdu and Yajiang County,reveals the feature knowledge,establishes the recognition and extraction model based on the feature knowledge,and extracts the asphalt road,plain cement road,mountain cement road,bridge and tunnel by using the recognition model.The overall accuracy of Asphalt road,plain cement road,mountain cement road,bridge,tunnel mouth and toll station are 83.75%?84.43%?85.13%?94.32%?93.65%and83.93%,and the Kappa coefficients of them are 66.52%?68.86%?69.82%?88.54%?86.83 and 61.46%.Through comparative analysis,it can be concluded that the extraction effect of bridge and tunnel portal is the best,followed by cement road and asphalt road,and toll station extraction is the worst.Among them,the extraction effect of cement road is better than asphalt road's,and the extraction effect of mountain road is better than plain area's.(4)Combined with the Google image,the roads?data of two phases in Yajiang County have obtained by modifying and improving the data of roads,bridges,tunnels,etc.On this basis,the temporal and spatial characteristics of roads in Yajiang County have been analyzed,and the distribution characteristics of roads,bridges and tunnels in Yajiang County have been obtained.At the same time,it also has researched the dynamic changes of roads in recent five years and the relationship between the distribution of roads and topography in Yajiang County.The results show that the newly built roads in Yajiang County are mainly concentrated in pubarong Township,decha Township,milong Township and zhusang Township,and the road distribution in Yajiang County is mostly concentrated in the low latitude River Valley in the middle.(5)Under the background of less research on road accessibility on a small scale,studying the road accessibility of the towns in Yajiang County.The research results show:(1)The overall road density of Yajiang County is about 0.265 km/km~2,and most of the towns exceed the national average level of 0.2 km/km~2.The high-value areas of road density are mainly concentrated in the Central Valley and the eastern and western areas,while the low-value areas are concentrated in the north and south high-altitude areas.(2)The road connectivity of Yajiang County is about 1.54,which is close to the state standard value(1.6)of automobile special highway network and the overall road network connectivity is good.In each township,the lowest road connectivity is about 1.3,which indicates that the road network of each township in Yajiang County shows the distribution form of tree and each township has reached the level of village community access.(3)The road service index of Yajiang County is about 12.35%,and the overall road service scope is not large.From the perspective of spatial pattern,the road service index shows a high spatial distribution in the middle and low in the north and South.(4)The comprehensive road accessibility of Yajiang County and the spatial distribution pattern is generally consistent with the distribution of road service index,road connectivity and road density,showing a high distribution pattern in the middle and low distribution pattern in the north and south,of which Honglong township has the highest accessibility and Bayirong township has the lowest accessibility.The extraction method explored in this study has certain reference value in extracting various road types under different environments.The spatial and temporal characteristics of roads discovered by this research have great meaning for road planning,construction and management in Yajiang County,and the characteristics of road accessibility provide scientific theoretical basis for the comprehensive and coordinated development of each township.
Keywords/Search Tags:high remote sensing, object-oriented, roads information extraction, accessibility
PDF Full Text Request
Related items