Font Size: a A A

Methods Of Radar Remote Sensing Archaeology In Typical Sites Of Han-Wei Dynasty In He-luo Region

Posted on:2019-01-28Degree:DoctorType:Dissertation
Country:ChinaCandidate:A H JiangFull Text:PDF
GTID:1362330578972565Subject:Geodesy and Survey Engineering
Abstract/Summary:PDF Full Text Request
As typical heritage sites in the He-Luo Region,the Han-Wei Capital City and the Hangu Pass of Han Dynasty are important sites of a world heritage project named "Silk Road:the Routes Networks of Chang'an-Tianshan Corridor".Besides,their archaeological value becomes even more significant with the implementation of China's "One Belt and One Road" strategy.Multi-mode and multi-resolution radar remote sensing provides new data sources,methods and technological supports for archeology.This dissertation focuses on some problems existing in the archaeological exploration using radar remote sensing in the He-Luo Region,such as large vegetation cover,long phenological phase,singular scale of remain identification and high omission ratio.Finally some novel methods are proposed including a two-step classification method based on time series crop marks,a multi-scale remains identification method based on micro-topography marks and an instantaneous attributes integration(IAI)method aimed to identify the interfaces of deeply buried cultural layers.The main research contents are as follows:(1)The working mechanism of radar remote sensing archeology was deeply studied starting from the characteristics of radar remote sensing.The similarities and differences between synthetic aperture radar and ground penetrating radar in the archaeological field are systematically studied from two aspects:the image quality parameters and the response of typical features to the radar beam.Then the satellite-ground dual-scale observation strategy was produced to complete the integration and cross-validation of spaceborne synthetic aperture radar and ground penetrating radar remote sensing data in archaeological work.(2)Currently the radar remote sensing archaeological research based on crop marks mainly focuses on single imagery,long-band(L,C-band)SAR data.In the study,the backscatter coefficient values of the multi-temporal X-band COSMO-SkyMed data were used and analyzed,and a "two-step classification method" based on temporal crop marks was proposed.This article uses the Han-Wei Capital City Heritage as the research area,20 multi-temporal X-band COSMO-SkyMed images as data sources,and calculates the backscattering coefficients,numerical differences,and their corresponding mean difference of the time series of the surface winter wheat samples in the multi-temporal SAR images of relic area/non-relic area.We set the first step time series crop marks screening threshold and calculate the mean variance,and Intersection point values of the filtered backscatter coefficients.By determining the time series crop marks screening threshold of the second step to complete the "two-step classification method" to identify suspected sites.The accuracy of the "two-step classification method" proposed by the paper was verified by using geophysical exploration methods,which provided effective information for the excavation of the Han-Wei Capital City ruins and revealed the applicability of multi-temporal X-band SAR data in archaeological exploration.(3)At present time,some problems exist in radar remote sensing archaeology,such as the data sources used in heritage sites and the remains themselves both have a singular spatial scale,which indicates spatial resolution or size,resulting a high omission ratio.To solve the problem a multi-scale archaeological remains identification method,which is based on micro-topography marks,was studied and proposed using the intensity and phase information of multi-resolution and multi-sensor SAR data.This method was applied at Hangu Pass of Han Dynasty in three different ways.First,a time-domain average of intensity information of multi-temporal,moderate-resolution,L-band PALSAR images was performed to suppress the speck noise and enhance useful information,and helped to recognize the trend of defensive line of the Hangu Pass on a large scale and identify some fine remains like the tower of the pass on a small scale.Second,a interferometric process of high-resolution,X-band TerraSAR/TanDEM images was carried out to extract the high-precision digital elevation model(DEM)of the site,which supported to judge the overall trend of the defensive line on a large scale.Third,procedures like filtering and geocode were taken on a multi-temporal,ultrahigh-resolution,X-band TerraSAR image,which is acquired at Staring Spotlight imaging mode,helping to identify fine remains like beacon tower and outer wall of rammed earth.The result shows that in radar remote sensing archaeology,multi-resolution and multi-sensor SAR data accompanied with micro-topography marks can be helpful to identify remains at different scales accurately.(4)Facing the problems that the attribute information obtained by GPR is rarely used in archaeological exploration and thus cases and experiences are lacked to use the attribute information jointly at present time,a novel instantaneous attributes integration(IAI)method is proposed to distinguish the interfaces of cultural layers.On the basis of analyzing the time series spectrum characteristics of instantaneous amplitude,phase and frequency of' GPR signal,this method multiplies the instantaneous amplitude and normalized instantaneous frequency to achieve the integration of instantaneous attribute information and the late signal enhancement,and finally improves the positioning accuracy of the interface between two adjacent layers.This method was applied in the case of identifying the ancient roads at the Hangu Pass of Han Dynasty,and achieved a more reliable result than traditional GPR data processing method,which helped to implement a successful radar remote sensing archaeological exploration over a typical site at He-Luo Region based on GPR data.
Keywords/Search Tags:He-Luo Region, Radar Remote Sensing Archaeology, Temporal-Crop Marks, Micro-topography Marks, Attribute Marks, SAR, GPR
PDF Full Text Request
Related items