v1.0.0 Live API

API Developer Documentation

High-performance automated solvers for Rotational Captchas and 2 Similar Objects Matching Captchas with 1–5ms execution latency.

Try in Interactive Playground ⚡ Read Auth Setup

Authentication Requirement

Pass your API key in either an HTTP header or query parameter for all endpoints:

Header: X-API-Key: rk_xxxxxxxxxxxx

Query: ?api_key=rk_xxxxxxxxxxxx

POST /api/predict/rotational

Latency: ~2.5 ms

Solves Rotational Captchas (Outer Ring & Inner Disc) using OpenCV seam cross-correlation. Returns exact rotation target angle, slider drag angle, correction angle, and peak confidence score.

Request Payload (JSON)

{
  "outer_image": "<base64_string_or_data_uri>",
  "inner_image": "<base64_string_or_data_uri>"
}

Response Format (JSON)

{
  "predicted_rotation": 67.95,
  "label_angle": 67.95,
  "processing_time_ms": 2.54
}

POST /api/predict/similar-objects

Latency: ~8.0 ms

Evaluates matching similarity between 2 candidate objects/images for "Select 2 Similar Objects" captchas using OpenCV ORB feature keypoints and HSV color histogram correlation.

Request Payload (JSON)

{
  "image_a": "<base64_string_or_data_uri>",
  "image_b": "<base64_string_or_data_uri>"
}

Response Format (JSON)

{
  "is_match": true,
  "similarity_score": 0.895,
  "confidence": "high",
  "matching_keypoints": 24,
  "color_similarity": 0.942,
  "feature_similarity": 0.848,
  "processing_time_ms": 7.82,
  "solver": "opencv_feature_matcher"
}

Code Implementation Examples

import requests
import base64

API_URL = "http://localhost:8000"
API_KEY = "rk_your_api_key_here"

def get_base64(path):
    with open(path, "rb") as f:
        return base64.b64encode(f.read()).decode("utf-8")

# 1. Rotational Captcha Solve
response = requests.post(
    f"{API_URL}/api/predict/rotational",
    json={
        "outer_image": get_base64("outer.png"),
        "inner_image": get_base64("inner.png")
    },
    headers={"X-API-Key": API_KEY}
)
print("Rotational Result:", response.json())

# 2. Similar Objects Match Solve
match_res = requests.post(
    f"{API_URL}/api/predict/similar-objects",
    json={
        "image_a": get_base64("obj1.png"),
        "image_b": get_base64("obj2.png")
    },
    headers={"X-API-Key": API_KEY}
)
print("Similar Objects Result:", match_res.json())

const fs = require('fs');

const API_URL = 'http://localhost:8000';
const API_KEY = 'rk_your_api_key_here';

function getBase64(path) {
  return fs.readFileSync(path, { encoding: 'base64' });
}

async function runSolver() {
  // Rotational Captcha
  const res = await fetch(`${API_URL}/api/predict/rotational`, {
    method: 'POST',
    headers: {
      'X-API-Key': API_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      outer_image: getBase64('outer.png'),
      inner_image: getBase64('inner.png')
    })
  });
  console.log('Rotational Angle:', await res.json());
}

runSolver();

curl -X POST "http://localhost:8000/api/predict/rotational" \
  -H "X-API-Key: rk_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "outer_image": "iVBORw0KGgoAAAANSUhEUgAA...",
    "inner_image": "iVBORw0KGgoAAAANSUhEUgAA..."
  }'