Spaces:
Sleeping
Sleeping
File size: 2,058 Bytes
8a7ddf0 9cf599c 8a7ddf0 9cf599c 8a7ddf0 9cf599c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | ---
title: Anomaly Detection API
emoji: 🔍
colorFrom: blue
colorTo: red
sdk: docker
pinned: false
license: mit
---
# 🔍 Anomaly Detection API
Real-time anomaly detection for electrical components using PatchCore + OpenCV classification.
## 🚀 Quick Start
### API Endpoint
**POST** `/infer`
**Request:**
```json
{
"image_url": "https://example.com/your-image.jpg"
}
```
**Response:**
```json
{
"label": "Normal",
"boxed_url": "https://cloudinary.com/boxed_image.jpg",
"mask_url": "https://cloudinary.com/anomaly_mask.png",
"filtered_url": "https://cloudinary.com/filtered_anomalies.png",
"boxes": []
}
```
### Example Usage
```bash
curl -X POST "https://YOUR_USERNAME-anomaly-detection-api.hf.space/infer" \
-H "Content-Type: application/json" \
-d '{"image_url": "https://example.com/test.jpg"}'
```
```python
import requests
response = requests.post(
"https://YOUR_USERNAME-anomaly-detection-api.hf.space/infer",
json={"image_url": "https://example.com/test.jpg"}
)
result = response.json()
print(f"Classification: {result['label']}")
print(f"Boxed Image: {result['boxed_url']}")
```
## 📋 Classification Labels
- **Normal** - No anomalies detected
- **Full Wire Overload** - Entire wire showing overload
- **Point Overload (Faulty)** - Localized overload points
## 🔧 Technical Details
- **Model:** PatchCore (anomaly detection)
- **Classification:** OpenCV-based heuristics
- **Response Time:** ~5 seconds
- **Max Image Size:** Unlimited (auto-resized)
## 🌐 Endpoints
| Endpoint | Method | Description |
|----------|--------|-------------|
| `/` | GET | API documentation |
| `/health` | GET | Health check |
| `/infer` | POST | Run inference |
## 📦 Output Files
All processed images are uploaded to Cloudinary:
- **boxed_url:** Original image with bounding boxes
- **mask_url:** Grayscale anomaly heatmap
- **filtered_url:** Filtered image showing only anomalous regions
## 🛠️ Built With
- PyTorch 2.4.1
- Anomalib (PatchCore)
- OpenCV
- Flask
- Cloudinary
## 📄 License
MIT License
|