Detect and quantify urban expansion from multi-temporal built-up rasters. Use when the user wants to analyze changes, compare multi-temporal rasters, compute indices, or generate assessment reports.
1 次安装
记忆
Geoskill: Building Change Audit
试用Object-level building change detection between two epochs. Identifies new, demolished, expanded, reduced, split, and merged buildings from footprint vectors. Use when comparing two building datasets, auditing construction changes, or generating building change reports.
它能做什么
Object-level building change detection between two epochs. Identifies new, demolished, expanded, reduced, split, and merged buildings from footprint vectors. Use when comparing two building datasets, auditing construction changes, or generating building change reports.
技能文档
Building Change Audit
Compares two epochs of building footprints to detect object-level changes.
Trigger
Use when the user wants to:
- Compare two building datasets for changes
- Identify new/demolished buildings
- Audit construction changes between epochs
- Generate building change statistics
CLI Usage
# Basic comparison
python scripts/building_change_audit.py \
--before-buildings before.geojson \
--after-buildings after.geojson
# With custom IoU threshold
python scripts/building_change_audit.py \
--before-buildings before.geojson \
--after-buildings after.geojson \
--match-iou 0.3 --area-tolerance 0.1
# Specify output directory
python scripts/building_change_audit.py \
--before-buildings before.geojson \
--after-buildings after.geojson \
--output-dir ./audit-output
Parameters
| Parameter | Default | Description |
|---|---|---|
--before-buildings | required* | Before epoch buildings (GeoJSON/Shapefile) |
--after-buildings | required* | After epoch buildings (GeoJSON/Shapefile) |
--match-iou | 0.1 | IoU threshold for matching |
--area-tolerance | 0.05 | Area ratio tolerance for "unchanged" |
--bbox | — | Bounding box W,S,E,N for auto-downloading ms-buildings |
--date-range | — | Date range START,END for auto-download |
--aoi-file | — | AOI polygon file (GeoJSON) |
--cache-dir | — | Override the default cache directory |
--output-dir | ./building-change-output | Output directory |
*Required unless --synthetic or --bbox+--date-range is supplied.
Output
| File | Description |
|---|---|
building_changes.geojson | Change features with type and metrics |
report.html | HTML summary report |
output-manifest.json | Machine-readable manifest |
Change Types
| Type | Meaning |
|---|---|
new | Building present only in after epoch |
demolished | Building present only in before epoch |
expanded | Matched, area increased > tolerance |
reduced | Matched, area decreased > tolerance |
unchanged | Matched, area within tolerance |
split | One before → multiple after |
merge | Multiple before → one after |
Exit Codes
| Code | Meaning |
|---|---|
| 0 | Success |
| 2 | Argument error |
| 3 | Dependency missing |
| 6 | Data validation failure |
| 7 | Processing failure |
数据下载
本 skill 可自动从 Microsoft Planetary Computer 下载 Microsoft Building
Footprints (ms-buildings) data, generating two snapshots — one for
--before-buildings and one for --after-buildings — so you can run a
change audit without supplying your own GeoJSON.
python building_change_audit.py --bbox 116,39,117,40 --date-range 2024-06-01,2024-06-30 --output-dir
--bbox W,S,E,N: WGS-84 边界框 (西, 南, 东, 北)--date-range START,END: 日期范围 (YYYY-MM-DD,YYYY-MM-DD)--aoi-file: 替代 --bbox 的 GeoJSON 多边形--cache-dir: 缓存目录 (默认 ~/.geoskill_cache)
当用户只给 --bbox + --date-range (没有 --before-buildings/--after-buildings) 时，
skill 自动下载两份 ms-buildings 数据 (一前一后)。
注意：ms-buildings 是单一静态数据集, 两次下载返回的通常是同一份 parquet
blob — 真正的 temporal change detection 需要在脚本外做区域/日期过滤。
由于 ms-buildings 用的是 abfs:// URL (urllib 不支持)，下载会失败并自动
fall back 到 --synthetic 模式。
相关技能
Topology-aware cadastral parcel change detection between two epochs. Identifies new, deleted, expanded, reduced, split, and merged parcels from cadastral vector data (宗地/地块/调查图斑). Use when comparing two epochs of land parcels, auditing boundary changes, or generating cadastral change ledgers.
1 次安装
Extract building footprints and estimate height, floor count proxy, and volume from DSM/DTM/LiDAR data. Produces 2.5D urban models for 3D city modeling, population downscaling, and risk exposure analysis.
1 次安装
Estimate building footprint density and floor area ratio (FAR) from building footprints and heights using kernel density estimation.
Monitor construction progress from multi-temporal satellite imagery. Classifies project stages (clearing, earthwork, foundation, structure, finishing, completed) using spectral indices and detects stagnation. Use when tracking infrastructure projects, auditing construction timelines, or generating progress reports from remote sensing data.
1 次安装
Multi-temporal shoreline change rate analysis. Generates transects, computes Endpoint Rate (EPR) and Linear Regression Rate (LRR), and identifies erosion hotspots. Use when monitoring coastline retreat/accretion or identifying erosion-prone segments.
1 次安装