Loading repository dataβ¦
Loading repository dataβ¦
suchitrachendake97 / repository
π AegisBridge Creator | AI & Civil Engineering Innovator π Built an AI-powered Bridge Health Monitoring System using HTML, CSS, JavaScript, Python & Gemini Vision AI. Detects structural damage, generates engineering reports, health scores, risk assessments, and remediation plans aligned with Indian standards. Passionate about AI, infrastructure,
A powerful AI-powered bridge structural health monitoring system built using HTML, CSS, JavaScript, Python, and the OpenRouter API. This project provides an intelligent inspection assistant that analyzes bridge photographs, detects structural damage, and delivers detailed engineering reports in real time through a simple web interface. The system runs locally and communicates with the Google Gemini 2.0 Flash Vision AI (via OpenRouter) to generate accurate damage assessments aligned with Indian engineering standards.
AegisBridge
β
βββ AegisBridge.html
βββ server.py
βββ run.bat
βββ README.md
βββ LICENSE
Clone the repository:
git clone https://github.com/suchitrachendake97/AegisBridge.git
Move into the project folder:
cd AegisBridge
No additional dependencies to install β Python 3 is all you need.
Start the local server:
python server.py
You will see:
AegisBridge Proxy Server (Python)
URL β http://localhost:5000
Open your browser and go to:
http://localhost:5000
AegisBridge uses the OpenRouter API to access Google Gemini 2.0 Flash Vision AI. OpenRouter is a free gateway that gives you access to 300+ AI models through a single API key.
Open your browser and go to:
https://openrouter.ai
Click Sign Up and create a free account (Google login supported)
After logging in, go to:
https://openrouter.ai/keys
Click Create Key
Give it a name (e.g. AegisBridge) and click Create
Copy the key
β οΈ Save this key somewhere safe. You won't be able to see it again after closing the page.
Start the server and open the app in your browser:
http://localhost:5000
Click the βοΈ API Setup button in the top right corner of the screen
In the popup modal that appears:
OpenRouter API Key fieldgoogle/gemini-2.0-flash-001 (Free Β· Recommended)You will see a confirmation message in the chat:
β
Ready! Running google/gemini-2.0-flash-001.
| Model | Cost | Best For |
|---|---|---|
google/gemini-2.0-flash-001 | β FREE | Fast analysis β Recommended |
google/gemini-flash-1.5 | β FREE | Stable alternative |
meta-llama/llama-3.2-11b-vision-instruct:free | β FREE | Backup option |
anthropic/claude-3.5-sonnet | π° Paid | Most detailed analysis |
π‘ All FREE models support image vision and work perfectly for bridge inspection.
The system is powered by Google Gemini 2.0 Flash Vision and is designed to behave like an expert bridge structural inspector:
User sends a bridge image and asks:
Analyze this bridge and tell me about the damage.
AEGIS responds:
I can see significant diagonal cracking on the pier face,
approximately 0.5β1mm wide, indicating moderate shear stress.
There is also surface corrosion staining near the waterline.
Health Score: 58 / 100
Severity: MODERATE
Risk Level: ELEVATED
Recommended Actions:
- Immediate: Document crack locations with measurements
- Short-term (30 days): Epoxy injection and surface sealing
- Long-term: Full NDT assessment per IRC:SP:35
User:
What is epoxy injection?
AEGIS:
Epoxy injection is a crack repair technique where low-viscosity
epoxy resin is injected under pressure into cracks to restore
structural integrity. It is specified under IS:456 for concrete
repairs and is suitable for cracks between 0.05mm and 6mm wide.
You can also start AegisBridge using:
run.bat
This will:
http://localhost:5000| Metric | Achievement |
|---|---|
| Analysis Time | < 10 seconds per image |
| Deployment Cost | Zero β free API tier |
| Damage Categories | Cracks, Corrosion, Spalling, Rebar, Deformation |
| Health Score | 0β100 with confidence percentage |
| Severity Levels | NONE / MINOR / MODERATE / SEVERE / CRITICAL |
| Standards Covered | IRC:6, IRC:21, IRC:SP:35, IS:456, IS:1893, RDSO |
| Voice Support | Input + Output (Chrome / Edge) |
| Accessibility | Any modern browser β laptop or smartphone |
This project is released under the MIT License.
Suchitra Chendake GitHub: https://github.com/suchitrachendake97