Use this URL to cite or link to this record in EThOS:
Title: Improving contrast for the detection of archaeological vegetation marks using optical remote sensing techniques
Author: Stott, David
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Airborne archaeological prospection in arable crops relies on detecting features using contrasts in the growth of the overlying crop as a proxy. This is possible because thecomposition of the soil in the features differs from the unmodified subsoil, and this exerts influence on the state of the crop. This influence is expressed as changes in crop canopydensity, structure, and in periods of resource constraint, variations in vegetation stressand vigour. These contrasts are dynamic, and vary temporally with local weather, andspatially with variations in drift geology and land use. This means that the archaeologicalfeatures have no unique spectral signature usable for classification. Rather, contrast isexpressed as relative, local variation in the crop. The extent to which the features are detectable using a particular technique is dependanton the strength of the contrast and the ability of the sensor to resolve it. Current practicerelies heavily on photography in the visible spectrum, but other sensors and processingtechniques have the potential to improve our ability to resolve subtle contrasts. This isimportant, as it affords the opportunity to extend the detection temporally and in soiltypes not normally considered conducive to detection. This work uses multi-temporal spectro-radiometry and ground-based survey to studycontrast at two sites in southern England. From these measurements leaf area index, vegetationindices, the red-edge position, chlorophyll fluorescence and continuum removalof foliar absorption features were derived and compared to evaluate contrast. The knowledgegained from the ground-based surveys was used to inform the analysis of the airbornesurveys. This included the application of vegetation indices to RGB cameras, theuse of multi-temporal and full-waveform LiDAR to detect biomass variations, and the useof various techniques with hyper-spectral imaging spectroscopy. These methods providea demonstrable improvement in contrast, particularly in methods sensitve to chlorophyllfluorescence, which afford the opportunity to record transient and short term contraststhat are not resolved by other sensors.
Supervisor: Cohn, Anthony ; Beck, Anthony ; Boyd, Doreen Sponsor: AHRC ; EPSRC
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available