---
title: Acne Detection API
emoji: 🔬
colorFrom: indigo
colorTo: gray
sdk: docker
pinned: false
license: mit
---

# Acne Detection API

FastAPI backend untuk deteksi jerawat menggunakan YOLOv8/YOLOv26s.

## API Endpoints

| Method | Endpoint | Deskripsi |
|--------|----------|-----------|
| GET | `/` | HTML Interactive Test UI |
| POST | `/api/predict` | Deteksi dari upload gambar |
| POST | `/api/predict/json` | Deteksi dari base64 JSON |
| POST | `/api/predict/image` | Return gambar annotated langsung |
| GET | `/api/classes` | Daftar kelas yang dideteksi |

### Example Usage

```bash
curl -X POST "https://username-acne-api.hf.space/api/predict" \
  -F "image=@wajah.jpg" \
  -F "conf_threshold=0.05" \
  -F "skin_type=berminyak"
```

### Skin Type Parameter

| Nilai | Deskripsi |
|-------|-----------|
| `berminyak` | Kulit berminyak (default) |
| `kering` | Kulit kering |
| `sensitif` | Kulit sensitif |
| `kombinasi` | Kulit kombinasi (berminyak di T-zone, kering di pipi) |

### Response (Expert Rules)

```json
{
  "expert_rule": {
    "rule_number": 1,
    "degree": "ringan",
    "dominant_lesion": "comedone",
    "skin_type": "berminyak",
    "input_summary": {
      "severity_level": "mild",
      "detected_classes": ["comedone"],
      "lesion_counts": {"comedone": 3}
    }
  },
  "recommendation": {
    "active_ingredient": "Salicylic Acid (SA) 2%",
    "product_type": "Serum Eksfoliasi",
    "dosage": "2-3 tetes",
    "frequency": "1x sehari (malam)",
    "time_of_day": "malam",
    "instructions": "Oleskan tipis pada area jerawat...",
    "contraindications": ["Hindari penggunaan bersamaan dengan retinoid..."]
  }
}
```