考古遗址遥感探测 | Archaeological Site Detection
Fuses LiDAR micro-topography, multispectral vegetation anomalies and SAR backscatter to automatically screen suspected archaeological sites and assign anomaly grades, providing remote sensing leads for large-scale archaeological surveys.
The method works in three layers: a large-window detrending of the DEM extracts local relief (highlighting micro-landforms such as mounds and depressions); NDVI is computed and detrended to identify crop marks caused by buried remains; and SAR backscatter is z-scored to detect moisture/structural anomalies. The three anomaly layers are normalized, fused with weights (or by maximum), graded by thresholds (none/low/high), and local peaks are used to locate suspected site points.
Dependencies / 依赖
pip install 'numpy' 'rasterio' 'scipy'
Usage / 使用方法
Basic Usage
python geoskill-archaeology-site-detection.py --bbox 116.0 39.0 117.0 40.0 [other parameters]
Example 1 (Synthetic Data, Offline)
python geoskill-archaeology-site-detection.py --bbox 116 39 117 40 --synthetic --output-dir ./out
示例 2（真实多波段影像（DEM/Red/NIR/SAR））
python geoskill-archaeology-site-detection.py --input scene.tif --output-dir ./out
Example 3 (Switch to Maximum-Value Fusion)
python geoskill-archaeology-site-detection.py --input scene.tif --fusion max --output-dir ./out
Example 4 (Adjusting Weights and Classification Thresholds)
python geoskill-archaeology-site-detection.py --input scene.tif --w-relief 0.5 --w-spectral 0.3 --w-sar 0.2 --high-threshold 0.8 --output-dir ./out
Example 5 (Larger Detrending Window (Regional Landforms))
python geoskill-archaeology-site-detection.py --input scene.tif --window 25 --footprint 9 --output-dir ./out
Output / 输出
| File | Format | Description |
|---|---|---|
anomaly_score.tif |
GeoTIFF | Fused anomaly score [0,1] |
anomaly_level.tif |
GeoTIFF | Anomaly grade (0 none / 1 low / 2 high) |
suspected_sites.geojson |
GeoJSON | Suspected site points (with score and grade) |
detection_report.json |
JSON | Summary statistics and top sites |
output-manifest.json |
JSON | Run manifest |
Data Source / 数据源 / Source
Multi-band GeoTIFF with band order DEM / Red / NIR / SAR. Or use --synthetic to generate physically consistent simulated data (fully offline).
Privacy / 隐私声明 / Privacy
- Runs offline by default;
--syntheticmode requires no network at all. - All processing is done locally; user data is never uploaded.
License / License
MIT
name: geoskill-archaeology-site-detection description: 'LiDAR micro-topography, multispectral anomaly and SAR fusion for suspected archaeological site detection with anomaly grading'
考古遗址遥感探测 | Archaeological Site Detection
融合 LiDAR 微地形、多光谱植被异常与 SAR 后向散射，自动筛查疑似考古遗址并给出异常等级，为大范围考古调查提供遥感线索。
方法分三层：对 DEM 做大窗口去趋势提取局部起伏（突出土丘/凹陷等微地貌）；计算 NDVI 并去趋势识别地下遗存导致的作物标志 (crop mark)；对 SAR 后向散射做 z-score 识别湿度/结构异常。三层异常归一化后按权重融合（或取最大），阈值分级（无/低/高）并用局部峰值定位疑似遗址点。
依赖
pip install 'numpy' 'rasterio' 'scipy'
使用方法
基本用法
python geoskill-archaeology-site-detection.py --bbox 116.0 39.0 117.0 40.0 [其他参数]
示例 1（合成数据，离线）
python geoskill-archaeology-site-detection.py --bbox 116 39 117 40 --synthetic --output-dir ./out
示例 2（真实多波段影像（DEM/Red/NIR/SAR））
python geoskill-archaeology-site-detection.py --input scene.tif --output-dir ./out
示例 3（改用最大值融合）
python geoskill-archaeology-site-detection.py --input scene.tif --fusion max --output-dir ./out
示例 4（调整权重与分级阈值）
python geoskill-archaeology-site-detection.py --input scene.tif --w-relief 0.5 --w-spectral 0.3 --w-sar 0.2 --high-threshold 0.8 --output-dir ./out
示例 5（更大去趋势窗口（区域地貌））
python geoskill-archaeology-site-detection.py --input scene.tif --window 25 --footprint 9 --output-dir ./out
输出
| 文件 | 格式 | 说明 |
|---|---|---|
anomaly_score.tif |
GeoTIFF | 融合异常评分 [0,1] |
anomaly_level.tif |
GeoTIFF | 异常分级（0 无 / 1 低 / 2 高） |
suspected_sites.geojson |
GeoJSON | 疑似遗址点（含评分与等级） |
detection_report.json |
JSON | 统计摘要与 Top 站点 |
output-manifest.json |
JSON | 运行清单 |
数据源 / Source
多波段 GeoTIFF，波段顺序 DEM / Red / NIR / SAR。 或使用 --synthetic 生成物理一致的模拟数据（完全离线）。
隐私声明 / Privacy
- 默认离线运行，
--synthetic模式完全无网络。 - 所有处理在本地完成，不上传用户数据。
License
MIT