---
title: Context-CrackNet Pavement Crack Analyzer
emoji: 🔬
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: "5.49.1"
app_file: app.py
pinned: true
license: mit
short_description: Segment & measure pavement cracks in pixel space
---

# 🔬 Context-CrackNet — Pavement Crack Analyzer

Upload a pavement photograph → crack segmentation + pixel-space measurements.

📄 **Paper:** [Construction and Building Materials (2025)](https://doi.org/10.1016/j.conbuildmat.2025.141583)
🏛 North Dakota State University · ✍ Blessing Agyei Kyem et al.

## How to use

1. The app defaults to **Context-CrackNet** hosted from `Blessing988/context-cracknet-weights`
2. Select the **Training Dataset** checkpoint you want to use
3. Upload a pavement image
4. Click **🚀 Analyze Cracks**

All measurements are in **pixel units** — no camera parameters required.

| Metric | Description |
|--------|-------------|
| Crack Area (px²) | Total crack pixel count |
| Crack Length (px) | Medial-axis skeleton length |
| Mean Width (px) | 2 × mean Distance Transform at skeleton |
| Max Width (px) | 2 × max Distance Transform at skeleton |
| Crack Count | 8-connected components |
| Coverage (%) | Area ÷ (H×W) × 100 |