doniramdani820/geetest-slider-api

Geetest Slider Detection API

🚀 ONNX-based slider position detection for Geetest captcha using YOLOv8 model.

🔧 Setup Requirements

Files to Upload:

• —app.py - Main FastAPI application
• —requirements.txt - Python dependencies
• —Dockerfile - Container configuration
• —best_model.onnx - Your trained YOLO model
• —data.yaml - Model configuration file

Environment Variables:

Set in Space Settings → Repository secrets:

API_SECRET_KEY = DASDAS2

🎯 API Endpoints

POST /predict

Predict slider position from captcha image.

Headers:

Content-Type: application/json
Authorization: Bearer DASDAS2

Request Body:

{
  "image": "base64_encoded_image_string",
  "confidence_threshold": 0.5
}

Response:

{
  "success": true,
  "bbox": {
    "x1": 100,
    "y1": 50, 
    "x2": 120,
    "y2": 70,
    "confidence": 0.85
  },
  "slider_position": 150.5,
  "confidence": 0.85,
  "message": "Slider position detected successfully"
}

GET /health

Check API health status.

🚀 Usage

const response = await fetch('https://doniramdani820-geetest-slider-api.hf.space/predict', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer DASDAS2'
  },
  body: JSON.stringify({
    image: base64_image_string,
    confidence_threshold: 0.5
  })
});

const result = await response.json();
console.log('Predicted position:', result.slider_position);

🔍 Model Requirements

• —Format: ONNX (YOLOv8)
• —Input: 640x640 RGB images
• —Output: Bounding boxes for slider position
• —Classes: Based on your data.yaml configuration

🛡️ Security

• —API key authentication required
• —Rate limiting applied
• —Input validation and sanitization

📊 Performance

• —ONNX Runtime 1.18.0 for optimal inference
• —CPU optimized execution providers
• —Memory efficient session configuration