研究目的
Investigating the potential use of remote sensing technologies for detecting small-scale looting of archaeological sites, specifically in the Ayios Mnason area of Cyprus, to establish a monitoring tool for cultural heritage protection.
研究成果
Remote sensing technologies, including aerial and satellite imagery, are effective for detecting small-scale looting in archaeological sites, as demonstrated in Cyprus. Techniques like image enhancement and object-oriented classification can identify soil disturbances, but verification through in situ inspection is essential. The methodology has potential as an early warning system for monitoring and protecting cultural heritage, though it is best applied in specific, known areas to minimize false positives.
研究不足
The study is limited by the small scale of looting attempts and the reliance on existing, non-scheduled imagery, which may not capture all events. The spatial resolution of datasets (e.g., 1.84 m for WorldView-2 multispectral bands) and spectral limitations (e.g., lack of near-infrared in Google Earth images) constrain detection accuracy. Automatic extraction methods produced false positives due to similar spectral characteristics of soil and other land uses, requiring a priori knowledge of the area.
1：Experimental Design and Method Selection:
The study utilized existing remote sensing datasets, including aerial and satellite images, to detect looting signs through various image processing techniques such as vegetation indices, principal component analysis (PCA), color transformations, and object-oriented classification. The methodology involved visual interpretation and automated analysis to identify soil disturbances indicative of looting.
2：Sample Selection and Data Sources:
The case study area was the archaeological landscape of Ayios Mnason in Politiko village, Cyprus. Data sources included aerial orthophotos from 1993, 2008, and 2014 (with resolutions of 1 m, 0.5 m, and 0.2 m respectively), a WorldView-2 multispectral satellite image from June 2011, and Google Earth images from 2008 to
3：5 m, and 2 m respectively), a WorldView-2 multispectral satellite image from June 2011, and Google Earth images from 2008 to List of Experimental Equipment and Materials:
2016.
3. List of Experimental Equipment and Materials: Equipment included high-resolution satellite and aerial sensors (e.g., WorldView-2), software for image processing (ENVI 5.3), and ground truthing tools like a GNSS system with real-time kinematic positioning for in situ mapping.
4：3), and ground truthing tools like a GNSS system with real-time kinematic positioning for in situ mapping.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Images were processed using histogram enhancements, pan-sharpening algorithms (Gram-Schmidt and NNDiffuse), vegetation suppression, PCA, color transformations (HSL and HSV), and object-oriented segmentation. In situ inspections were conducted to validate findings, with GNSS used for precise mapping.
5：Data Analysis Methods:
Data were analyzed using statistical techniques in ENVI software, including PCA for change detection, various vegetation indices for feature enhancement, and object-oriented classification for automatic extraction of looting marks. Results were cross-compared with ground truth data.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
