Document Extraction

Extract structured data from documents using AI-powered OCR and layout understanding. Supports multi-page documents and returns page-level extracted fields, classification, and confidence scores.

POST

/api/gw/services/document_extraction

---

Authorizations

x-api-key · string · required

API key used to authenticate requests.

---

Body

imageBase64 · string · required

Base64 encoded document image or file.

filename · string

Original filename of the document.

---

Response

200

success · boolean

Indicates whether the request was processed successfully.

requestId · string

Unique identifier for the API request.

correlationId · string

Identifier used for tracing the request across systems.

data · object

Contains OCR processing results and metadata.

data.pages · number

Total number of processed pages.

data.ocrTotalMs · number

Total OCR processing time in milliseconds.

data.artifactsId · string

Identifier for stored artifacts/results.

data.results · array

List of extracted page-level results.

data.results[].pageIndex · number

Index of the processed page.

data.results[].status · string

Processing status of the page (e.g., SUCCESS, FAILED).

data.results[].docType · string

Detected document type (e.g., INVOICE).

data.results[].docSide · string

Document side processed (e.g., FRONT, BACK).

data.results[].fields · object

Key-value pairs of extracted fields (e.g., invoice_number, total_amount).

data.results[].confidence · number

Confidence score for the extracted data (0 to 1).

data.results[].warnings · array

List of warnings encountered during processing.

data.results[].pageImages · string

Base64 or reference string for the processed page image.

data.failureReason · string

Reason for failure if processing was not successful.

error · object | null

Contains error details if the request fails; otherwise null.

durationMs · number

Time taken to process the request in milliseconds.

processedAt · string

Timestamp when the request was processed (ISO 8601 format).

---

Curl

curl -X POST "https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction" \
  -H "Content-Type: application/json" \
  -d '{
    "imageBase64": "<base64-document>",
    "filename": "invoice.pdf"
}'

import requests

url = "https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction"

payload = {
    "imageBase64": "<base64-document>",
    "filename": "invoice.pdf"
}

print(requests.post(url, json=payload).json())

fetch("https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    imageBase64: "<base64-document>",
    filename: "invoice.pdf"
  })
})
.then(res => res.json())
.then(console.log);

<?php
$url = "https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction";

$data = [
  "imageBase64" => "<base64-document>",
  "filename" => "invoice.pdf"
];

$ch = curl_init($url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($data));
curl_setopt($ch, CURLOPT_HTTPHEADER, ['Content-Type: application/json']);

$response = curl_exec($ch);
curl_close($ch);

echo $response;
?>

package main

import (
  "bytes"
  "net/http"
)

func main() {
  json := []byte(`{
    "imageBase64":"<base64-document>",
    "filename":"invoice.pdf"
  }`)

  http.Post(
    "https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction",
    "application/json",
    bytes.NewBuffer(json),
  )
}

import okhttp3.*;

public class Main {
  public static void main(String[] args) throws Exception {

    OkHttpClient client = new OkHttpClient();

    String json = "{\"imageBase64\":\"<base64-document>\", \"filename\":\"invoice.pdf\" }";

    RequestBody body = RequestBody.create(json, MediaType.parse("application/json"));

    Request request = new Request.Builder()
      .url("https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction")
      .post(body)
      .build();

    Response response = client.newCall(request).execute();
    System.out.println(response.body().string());
  }
}

require 'net/http'
require 'uri'
require 'json'

uri = URI("https://dev-idf-gw.cognetlabs.org/api/gw/services/document_extraction")

payload = {
  imageBase64: "<base64-document>",
  filename: "invoice.pdf"
}

http = Net::HTTP.new(uri.host, uri.port)
request = Net::HTTP::Post.new(uri, {'Content-Type': 'application/json'})
request.body = payload.to_json

response = http.request(request)
puts response.body

200

{
"success":true,
"requestId":"909ca3d2fe34",
"correlationId":"3f1b300063d94ed9a4ea4460a5656c2c",
"data":{
"pages":1,
"ocrTotalMs":320,
"artifactsId":"art_12345",
"results":[
{
"pageIndex":0,
"status":"SUCCESS",
"docType":"INVOICE",
"docSide":"FRONT",
"fields":{
"invoice_number":"12345",
"total_amount":"450.00"
},
"confidence":0.95,
"warnings":[
"string",
"string"
],
"pageImages":"string"
}
],
"failureReason":"string"
},
"error":null,
"durationMs":274,
"processedAt":"2026-04-29T13:58:18.7342517Z"
}

{
    "error": "ImageBase64 is required"
}