Smart Document ScannerVision + OCR Workbench
Checking API
Computer Vision + OCR Document Pipeline

Turn an angled document photo into inspectable data.

Detect the document, correct perspective, improve readability, extract text, and inspect every processing artifact in one workbench.

Open scanner workbenchRead API docs

INPUT_FRAME

DetectCorrectEnhanceOCR
Pipeline

How the five-stage pipeline works

Each stage writes an inspectable artifact and reports progress to the active job.

  1. 01

    Detect

    Find the dominant four-corner document contour.

  2. 02

    Correct

    Apply a homography transform for a stable top-down view.

  3. 03

    Enhance

    Balance contrast, suppress noise, and sharpen text.

  4. 04

    OCR

    Read text lines with per-line confidence from the OCR engine.

  5. 05

    Parse

    Map recognized text into document-specific fields.

Structured output

Targeted document types and schemas

Automatic classification is available, or choose a target type before processing.

Business card

business_card
  • name
  • company
  • email
  • phone

ID card

id_card
  • nik
  • name
  • birth_date
  • blood_type

Receipt

receipt
  • merchant
  • reference_number
  • date
  • total

Paper document

paper_document
  • title
  • document_number
  • date
REST API

Use the same pipeline from your application.

The browser workbench and API call the identical processing service.

POST /api/process
curl -X POST http://localhost:8000/api/process \
  -F "file=@document.jpg" \
  -F "doc_type=auto"