Font Size: a A A

Research On Extraction Of Oil And Gas Drilling Platforms With Remote Sensing In The South China Sea

Posted on:2016-07-21Degree:MasterType:Thesis
Country:ChinaCandidate:W Y ChengFull Text:PDF
GTID:2311330461958275Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Offshore oil and gas resources have gained more and more global attention as increasing energy demands and great development exploitation technologies.The South China Sea will be an important energy base in the future with rich resources,which relates to maritime rights,energy strategies and peaceful development.Currently,the exploitation of oil and gas resources in the South China Sea has become the focus conflict of strategic interests.Vietnam,Malaysia,the Philippines and other countries have expanded their drilling operation areas to China's coastal territory,which poses a serious threat to our energy security and sovereignty.Therefore,in view of the conflict between the intensifying dispute of the oil and gas resources and the deficient spatial locations of drilling platforms,this research aims at detecting the platforms based on the support of multi-source remote sensing images.The spatial locations of offshore platforms in the South China Sea have been automatically extracted using two strategies(time-series image strategy and multi-refinement strategy)and two features(position-invariance and size-invariance).It would serve a rapid,timely and accurate control purpose of grasping oil and gas resources exploitation conditions,as well as providing technical support for offshore resource exploitation,marine environment management and marine safety.The research included:(1)Based on the contextual features of offshore platforms from the DMSP/OLS nighttime light data,Gaussian filtering,Mean filtering and Threshold methods were utilized to detect the target regions of platforms.Accordingly,the related dataset was effectively collected with guidance.(2)Time-series image strategy and multi-refinement strategy were employed to Landsat-8 OLI images after radiation calibration and enhancement processing.Hence,false alarms such as vessels and clouds were excluded based on position-invariance and size-invariance of platforms locations.Additionally,Eliminations of onshore land masks and offshore false alarms finally achieved the automatic extraction of spatial locations of offshore platforms in the South China Sea.The omission error was 3.8%and the commission error was below 1.0%.(3)Further platform features such as ages,types,depth were also extracted with the support of time-series images,high resolution images and water depth data.Consequently,a full coverage with spatial locations and attributive character dataset has been successfully established.The results showed:(1)Totally 1075 offshore platforms were detected in the South China Sea.Among them,60 in China,125 in Vietnam,13 in the Philippines,306 in Thailand,341 in Malaysia,32 in Indonesia,and 129 in Brunei.Besides,33 platforms located in the Vietnam-Thailand-Cambodia disputed area,25 platforms located in Malaysia-Thailand Joint Development Area and 11 platforms located in Malaysia-Vietnam continental shelf Defined Area.(2)Meanwhile,a total of 60 offshore platforms from other countries include Vietnam and Malaysia located in China's coastline line by 2014.Vietnam has built 5 platforms since 1999 and Malaysia has built 55platforms since 1992.(3)The number of offshore platforms is increasing from 233 to 1075 since 1992.Offshore platforms in the South China Sea include:725 small platforms,171 large platforms,84 complex structured platforms and FPSO(Floating Production Storage and Offloading).It is 99%platforms located in shallow sea areas(<500 m)where were the major exploitation regions in the South China Sea.In other words,1037 platforms operating depth lower than 100 m while 31 platforms between 101 to 500 m.7 deepwater oil and gas drilling platforms were built since 2005.
Keywords/Search Tags:South China Sea, Oil and gas drilling platform, Automatic extraction, Multi-refinement strategy, Attribute feature
PDF Full Text Request
Related items