Drone Inspection Specialist
Expert in drone-based infrastructure inspection with computer vision, thermal analysis, and 3D reconstruction for insurance, property assessment, and environmental monitoring.
Decision Tree: When to Use This Skill
User mentions drones/UAV?
├─ YES → Is it about inspection or assessment of something?
│ ├─ Fire detection, smoke, thermal hotspots → THIS SKILL
│ ├─ Roof damage, hail, shingles → THIS SKILL
│ ├─ Property/insurance assessment → THIS SKILL
│ ├─ 3D reconstruction for measurement → THIS SKILL
│ ├─ Wildfire risk, defensible space → THIS SKILL
│ └─ NO (flight control, navigation, general CV) → drone-cv-expert
└─ NO → Is it about fire/roof/property assessment without drones?
├─ YES → Still use THIS SKILL (methods apply)
└─ NO → Different skill needed
Core Competencies
Fire Detection & Wildfire Risk
- Multi-Modal Detection: RGB smoke + thermal hotspot fusion
- Precondition Assessment: NDVI, fuel load, vegetation density
- Defensible Space: CAL FIRE/NFPA 1144 compliance evaluation
- Progression Tracking: Spread rate, direction prediction
Roof & Structural Inspection
- Damage Detection: Cracks, missing shingles, wear, ponding
- Hail Analysis: Impact pattern recognition, size estimation
- Thermal Analysis: Moisture detection, insulation gaps, HVAC leaks
- Material Classification: Asphalt, metal, tile, slate identification
3D Reconstruction (Gaussian Splatting)
- Pipeline: Video → COLMAP SfM → 3DGS training → Web viewer
- Measurements: Roof area, damage dimensions, property bounds
- Change Detection: Before/after comparison for claims
Insurance & Reinsurance
- Claim Packaging: Documentation meeting industry standards
- Risk Modeling: Catastrophe models, loss distributions
- Precondition Data: Satellite + drone + ground integration
Anti-Patterns to Avoid
1. "Single-Sensor Dependence"
Wrong: Using only RGB for fire detection.
Right: Multi-modal fusion (RGB + thermal) for high-confidence alerts.
| Detection Source |
Confidence |
Action |
| Thermal fire only |
70% |
Alert + verify |
| RGB smoke only |
60% |
Alert + investigate |
| Thermal + RGB |
95% |
Confirmed fire |
2. "Ignoring Hail Pattern"
Wrong: Counting damage without analyzing spatial distribution.
Right: True hail damage has RANDOM distribution. Linear or clustered patterns indicate other causes (foot traffic, age).
3. "Thermal Temperature Trust"
Wrong: Using raw thermal values without calibration.
Right: Account for:
- Emissivity of materials (roof = 0.9-0.95)
- Atmospheric transmission (humidity, distance)
- Reflected temperature from surroundings
- Time of day (thermal lag)
4. "3DGS Frame Overload"
Wrong: Extracting every frame from drone video.
Right: Extract 2-3 fps with 80% overlap. More frames ≠ better reconstruction.
| Video FPS |
Extract Rate |
Result |
| 30 |
30 (all) |
Redundant, slow processing |
| 30 |
2-3 |
Optimal quality/speed |
| 30 |
0.5 |
Insufficient overlap |
5. "Insurance Claim Speculation"
Wrong: Estimating costs without material identification.
Right: Identify material → Apply correct cost matrix.
| Material |
Repair $/sqft |
Replace $/sqft |
| Asphalt shingle |
$5-10 |
$3-7 |
| Metal |
$10-15 |
$8-14 |
| Tile |
$12-20 |
$10-18 |
| Slate |
$20-40 |
$15-30 |
6. "Defensible Space Zone Confusion"
Wrong: Treating all vegetation equally regardless of distance.
Right: CAL FIRE zones have different requirements:
| Zone |
Distance |
Requirement |
| 0 |
0-5 ft |
Ember-resistant (no combustibles) |
| 1 |
5-30 ft |
Lean, clean, green (spaced trees) |
| 2 |
30-100 ft |
Reduced fuel (selective thinning) |
Data Collection Strategy
Satellite Data (Regional Context)
- Sentinel-2: 10m resolution, NDVI, fuel moisture (SWIR bands)
- Landsat-8: 30m resolution, historical baseline, thermal band
- Planet: 3m resolution daily, change detection
- Application: Regional risk mapping, before/after events
Drone Data (Property Detail)
- RGB Mapping: 2-5cm GSD, orthomosaic, 3D model
- Thermal Survey: Moisture detection, heat signatures
- Close Inspection: Damage documentation, detail photos
- Application: Individual property assessment
Ground Truth
- Slope Measurement: GPS transects for topographic risk
- Soil Sampling: Moisture content for fire risk
- Material Verification: Confirm roof type
- Application: Calibration and validation
Quick Reference Tables
Fire Detection Confidence Levels
| Signal Combination |
Confidence |
Alert Priority |
| Thermal >150°C + Smoke |
95% |
CRITICAL |
| Thermal fire model |
80% |
HIGH |
| Hotspot >80°C |
70% |
MEDIUM |
| Smoke only |
60% |
MEDIUM |
| Hotspot 60-80°C |
50% |
LOW |
Roof Damage Severity
| Type |
Low |
Medium |
High |
Critical |
| Missing shingle |
- |
- |
Always |
- |
| Crack |
<1" |
1-3" |
>3" |
Multiple |
| Granule loss |
<10% |
10-30% |
>30% |
- |
| Ponding |
- |
Small |
Large |
Active leak |
Wildfire Risk Factors (Weighted)
| Factor |
Weight |
High Risk Indicators |
| Defensible space |
20% |
Non-compliant zones |
| Vegetation density |
20% |
NDVI >0.6, high fuel load |
| Slope |
15% |
>30% grade |
| Roof material |
10% |
Wood shake, Class C |
| Structure spacing |
10% |
<30ft between buildings |
| Access/egress |
10% |
Single road, narrow |
3DGS Quality Settings
| Quality Level |
Iterations |
Time |
Use Case |
| Preview |
7K |
5 min |
Quick check |
| Standard |
30K |
30 min |
General use |
| High |
50K |
60 min |
Documentation |
| Inspection |
100K |
3 hrs |
Damage measurement |
Reference Files
Detailed implementations in references/:
fire-detection.md - Multi-modal fire detection, thermal cameras, progression tracking
roof-inspection.md - Damage detection, thermal analysis, material classification
insurance-risk-assessment.md - Hail damage, wildfire risk, catastrophe modeling, reinsurance
gaussian-splatting-3d.md - COLMAP pipeline, 3DGS training, inspection measurements
Integration Points
- drone-cv-expert: Flight control, navigation, general CV algorithms
- metal-shader-expert: GPU-accelerated 3DGS rendering
- collage-layout-expert: Visual report composition
- clip-aware-embeddings: Material/damage classification assistance
Insurance Workflow
1. Pre-Event Assessment (Underwriting)
├─ Satellite: Regional risk context
├─ Drone: Property-level risk factors
└─ Output: Risk score, premium factors
2. Post-Event Inspection (Claims)
├─ Drone survey: Damage documentation
├─ 3DGS: Measurements, change detection
└─ Output: Claim package, cost estimate
3. Portfolio Risk (Reinsurance)
├─ Aggregate: TIV, loss curves
├─ Model: AAL, PML, concentration
└─ Output: Treaty pricing, structure
Key Principle: Inspection accuracy depends on multi-source data fusion. Single-sensor assessments miss critical context. Always correlate drone findings with satellite baseline and weather data for defensible conclusions.
1---2name: drone-inspection-specialist-23description: Advanced CV for infrastructure inspection including forest fire detection, wildfire precondition assessment, roof inspection, hail damage analysis, thermal imaging, and 3D Gaussian Splatting4 reconstruction. Expert in multi-modal detection, insurance risk modeling, and reinsurance data pipelines. Activate on "fire detection", "wildfire risk", "roof inspection", "hail damage", "thermal analysis",5 "Gaussian Splatting", "3DGS", "insurance inspection", "defensible space", "property assessment", "catastrophe modeling", "NDVI", "fuel load". NOT for general drone flight control, SLAM, path planning,6 or sensor fusion (use drone-cv-expert), GPU shader development (use metal-shader-expert), or generic object detection without inspection context (use clip-aware-embeddings).7allowed-tools: Read,Write,Edit,Bash(python:*,pip:*),Grep,Glob,mcp__firecrawl__firecrawl_search,WebFetch,mcp__stability-ai__stability-ai-generate-image8metadata:9 source: https://skillmd.com/skills/curiositech/drone-inspection-specialist-210 category: AI & Machine Learning11 pairs-with:12 - skill: drone-cv-expert13 reason: Core drone navigation and CV14 - skill: clip-aware-embeddings15 reason: Semantic understanding of inspected areas16 tags:17 - inspection18 - fire-detection19 - thermal20 - gaussian-splatting21 - insurance22---2324# Drone Inspection Specialist2526Expert in drone-based infrastructure inspection with computer vision, thermal analysis, and 3D reconstruction for insurance, property assessment, and environmental monitoring.2728## Decision Tree: When to Use This Skill2930```31User mentions drones/UAV?32├─ YES → Is it about inspection or assessment of something?33│ ├─ Fire detection, smoke, thermal hotspots → THIS SKILL34│ ├─ Roof damage, hail, shingles → THIS SKILL35│ ├─ Property/insurance assessment → THIS SKILL36│ ├─ 3D reconstruction for measurement → THIS SKILL37│ ├─ Wildfire risk, defensible space → THIS SKILL38│ └─ NO (flight control, navigation, general CV) → drone-cv-expert39└─ NO → Is it about fire/roof/property assessment without drones?40 ├─ YES → Still use THIS SKILL (methods apply)41 └─ NO → Different skill needed42```4344## Core Competencies4546### Fire Detection & Wildfire Risk47- **Multi-Modal Detection**: RGB smoke + thermal hotspot fusion48- **Precondition Assessment**: NDVI, fuel load, vegetation density49- **Defensible Space**: CAL FIRE/NFPA 1144 compliance evaluation50- **Progression Tracking**: Spread rate, direction prediction5152### Roof & Structural Inspection53- **Damage Detection**: Cracks, missing shingles, wear, ponding54- **Hail Analysis**: Impact pattern recognition, size estimation55- **Thermal Analysis**: Moisture detection, insulation gaps, HVAC leaks56- **Material Classification**: Asphalt, metal, tile, slate identification5758### 3D Reconstruction (Gaussian Splatting)59- **Pipeline**: Video → COLMAP SfM → 3DGS training → Web viewer60- **Measurements**: Roof area, damage dimensions, property bounds61- **Change Detection**: Before/after comparison for claims6263### Insurance & Reinsurance64- **Claim Packaging**: Documentation meeting industry standards65- **Risk Modeling**: Catastrophe models, loss distributions66- **Precondition Data**: Satellite + drone + ground integration6768## Anti-Patterns to Avoid6970### 1. "Single-Sensor Dependence"71**Wrong**: Using only RGB for fire detection.72**Right**: Multi-modal fusion (RGB + thermal) for high-confidence alerts.73| Detection Source | Confidence | Action |74|------------------|------------|--------|75| Thermal fire only | 70% | Alert + verify |76| RGB smoke only | 60% | Alert + investigate |77| Thermal + RGB | 95% | Confirmed fire |7879### 2. "Ignoring Hail Pattern"80**Wrong**: Counting damage without analyzing spatial distribution.81**Right**: True hail damage has RANDOM distribution. Linear or clustered patterns indicate other causes (foot traffic, age).8283### 3. "Thermal Temperature Trust"84**Wrong**: Using raw thermal values without calibration.85**Right**: Account for:86- Emissivity of materials (roof = 0.9-0.95)87- Atmospheric transmission (humidity, distance)88- Reflected temperature from surroundings89- Time of day (thermal lag)9091### 4. "3DGS Frame Overload"92**Wrong**: Extracting every frame from drone video.93**Right**: Extract 2-3 fps with 80% overlap. More frames ≠ better reconstruction.94| Video FPS | Extract Rate | Result |95|-----------|--------------|--------|96| 30 | 30 (all) | Redundant, slow processing |97| 30 | 2-3 | Optimal quality/speed |98| 30 | 0.5 | Insufficient overlap |99100### 5. "Insurance Claim Speculation"101**Wrong**: Estimating costs without material identification.102**Right**: Identify material → Apply correct cost matrix.103| Material | Repair $/sqft | Replace $/sqft |104|----------|--------------|----------------|105| Asphalt shingle | $5-10 | $3-7 |106| Metal | $10-15 | $8-14 |107| Tile | $12-20 | $10-18 |108| Slate | $20-40 | $15-30 |109110### 6. "Defensible Space Zone Confusion"111**Wrong**: Treating all vegetation equally regardless of distance.112**Right**: CAL FIRE zones have different requirements:113| Zone | Distance | Requirement |114|------|----------|-------------|115| 0 | 0-5 ft | Ember-resistant (no combustibles) |116| 1 | 5-30 ft | Lean, clean, green (spaced trees) |117| 2 | 30-100 ft | Reduced fuel (selective thinning) |118119## Data Collection Strategy120121### Satellite Data (Regional Context)122- **Sentinel-2**: 10m resolution, NDVI, fuel moisture (SWIR bands)123- **Landsat-8**: 30m resolution, historical baseline, thermal band124- **Planet**: 3m resolution daily, change detection125- **Application**: Regional risk mapping, before/after events126127### Drone Data (Property Detail)128- **RGB Mapping**: 2-5cm GSD, orthomosaic, 3D model129- **Thermal Survey**: Moisture detection, heat signatures130- **Close Inspection**: Damage documentation, detail photos131- **Application**: Individual property assessment132133### Ground Truth134- **Slope Measurement**: GPS transects for topographic risk135- **Soil Sampling**: Moisture content for fire risk136- **Material Verification**: Confirm roof type137- **Application**: Calibration and validation138139## Quick Reference Tables140141### Fire Detection Confidence Levels142| Signal Combination | Confidence | Alert Priority |143|-------------------|------------|----------------|144| Thermal >150°C + Smoke | 95% | CRITICAL |145| Thermal fire model | 80% | HIGH |146| Hotspot >80°C | 70% | MEDIUM |147| Smoke only | 60% | MEDIUM |148| Hotspot 60-80°C | 50% | LOW |149150### Roof Damage Severity151| Type | Low | Medium | High | Critical |152|------|-----|--------|------|----------|153| Missing shingle | - | - | Always | - |154| Crack | <1" | 1-3" | >3" | Multiple |155| Granule loss | <10% | 10-30% | >30% | - |156| Ponding | - | Small | Large | Active leak |157158### Wildfire Risk Factors (Weighted)159| Factor | Weight | High Risk Indicators |160|--------|--------|---------------------|161| Defensible space | 20% | Non-compliant zones |162| Vegetation density | 20% | NDVI >0.6, high fuel load |163| Slope | 15% | >30% grade |164| Roof material | 10% | Wood shake, Class C |165| Structure spacing | 10% | <30ft between buildings |166| Access/egress | 10% | Single road, narrow |167168### 3DGS Quality Settings169| Quality Level | Iterations | Time | Use Case |170|---------------|------------|------|----------|171| Preview | 7K | 5 min | Quick check |172| Standard | 30K | 30 min | General use |173| High | 50K | 60 min | Documentation |174| Inspection | 100K | 3 hrs | Damage measurement |175176## Reference Files177178Detailed implementations in `references/`:179- `fire-detection.md` - Multi-modal fire detection, thermal cameras, progression tracking180- `roof-inspection.md` - Damage detection, thermal analysis, material classification181- `insurance-risk-assessment.md` - Hail damage, wildfire risk, catastrophe modeling, reinsurance182- `gaussian-splatting-3d.md` - COLMAP pipeline, 3DGS training, inspection measurements183184## Integration Points185186- **drone-cv-expert**: Flight control, navigation, general CV algorithms187- **metal-shader-expert**: GPU-accelerated 3DGS rendering188- **collage-layout-expert**: Visual report composition189- **clip-aware-embeddings**: Material/damage classification assistance190191## Insurance Workflow192193```1941. Pre-Event Assessment (Underwriting)195 ├─ Satellite: Regional risk context196 ├─ Drone: Property-level risk factors197 └─ Output: Risk score, premium factors1981992. Post-Event Inspection (Claims)200 ├─ Drone survey: Damage documentation201 ├─ 3DGS: Measurements, change detection202 └─ Output: Claim package, cost estimate2032043. Portfolio Risk (Reinsurance)205 ├─ Aggregate: TIV, loss curves206 ├─ Model: AAL, PML, concentration207 └─ Output: Treaty pricing, structure208```209210---211212**Key Principle**: Inspection accuracy depends on multi-source data fusion. Single-sensor assessments miss critical context. Always correlate drone findings with satellite baseline and weather data for defensible conclusions.