도로명변환기 OCR 시험실 — third-party notices

All OCR inference runs on the user's device. This application has no image-upload or cloud OCR endpoint.

PaddleOCR models (Apache License 2.0)
  https://huggingface.co/PaddlePaddle/korean_PP-OCRv5_mobile_rec_onnx
  revision: 5c6f574b8e2230adf4287b33e736d71b9fabd28e
  https://huggingface.co/PaddlePaddle/PP-OCRv5_mobile_det_onnx
  revision: e6f4fa85f00e168c862bc462aebca69eef9b3d3d
  The character dictionary is extracted from the recognition model's inference.yml.
  Full license: models/PaddleOCR-LICENSE.txt

ONNX Runtime Web 1.22.0 (MIT)
  Copyright Microsoft Corporation. All rights reserved.
  https://github.com/microsoft/onnxruntime
  Full license: vendor/onnxruntime-LICENSE.txt

Tesseract.js 6.0.1 / Tesseract.js-core 6.0.0 (Apache License 2.0)
  https://github.com/naptha/tesseract.js
  https://github.com/naptha/tesseract.js-core
  Full licenses and bundled dependency notices:
  vendor/tesseract-LICENSE.txt
  vendor/tesseract-core-LICENSE.txt
  vendor/tesseract.min.js.LICENSE.txt
  vendor/worker.min.js.LICENSE.txt

Tesseract Korean and English traineddata (Apache License 2.0)
  https://tessdata.projectnaptha.com/4.0.0/kor.traineddata.gz
  https://tessdata.projectnaptha.com/4.0.0/eng.traineddata.gz
  https://github.com/tesseract-ocr/tessdata

Example image is a synthetic test fixture, not a photographed mail item.
It is rendered using Noto Sans CJK KR (SIL Open Font License 1.1).
The font file itself is not distributed with the application.

Asset download URLs and SHA-256 digests: assets-manifest.json