Changelog

2.20.0

[MEX]. Improved support for detecting the CURP and ElectorKey.
[ARG]. Extraction of the field JWT from the QR in idcard 2024 model.

2.19.0

Added support for CUE ESP foreign cards.
Use the parameters type: foreign_card and model: ESP
Improved support for the NIC Identity document.
Use the parameters type: id_card and model: NIC

2.18.2

Added support for the CRI Identity document.
The parameter type is optional.

2.18.1

New features:

Optimized extraction of category codes and dates in driver's licenses.

Fixes:

Retrieval of images from the back side of Identity documents.
Retrieval of images of documents processed with multilingual OCR.
Normalization of dates extracted from driver's licenses.

2.18.0

Added support for processing driver's licenses with table-based designs.
Added support for the COL Identity document.
Added support for the DOM Identity document.
Added support for the NIC Identity document.
Improved QR code detection.

2.17.2

New features:

The parameter model is optional.

2.17.1

New features:

Added support for an empty model parameter.

Fixes:

Return the content of the images when specified retrieveImages.

2.17.0

New features:

Added support for processing rotated images.
Added support for HND foreign card.
Added support for QR codes in the ARG Identity document.

2.16.2

New features:

Improved processing of SLV Identity document.

2.16.1

New features:

Improved processing of ARG Identity document.

2.16.0

New features:

Added support for loading configuration from environment variables.
Updated the OCR extractor.
Updated document templates (ARG).

Fixes:

Fixed a possible null pointer issue.
Improved validation for empty License scenarios.

2.15.0

New features:

Improved matching logic: the system no longer returns an error when no side match is found.
Improved matching behavior when multiple matches are detected.
Updated document templates for PER.
Integrated facephi-rest-client.

Fixes:

Errors are now reported in the correct order of occurrence to improve traceability.
Improved processing of Identity documents by automatically using the most suitable document mode when available.
Added source type information to the universal output.
Removed the facephi-flow dependency to simplify the architecture and reduce possible conflicts.

2.14.5

Updated document templates:
- ARG: fixed the scanning of Address in some documents.
- ARG: fixed the field name Numero de Tramite.

2.14.4

Updated document templates:
- ARG: support for the 2024 model

2.14.3

Updated document templates (ARG).

2.14.2

Updated the document reader SDK to Version 4.0.2 to disable IQA checks.

2.14.1

Updated the document reader SDK to Version 4.0.1 to correctly extract the document number from CHL Identity documents.

2.14.0

New POST Endpoint /api/v1/process_multi to process multilingual documents.
Refactored log recording with a consistent format.
Updated the OCR extractor.
Integration test updates for the new endpoints and configuration parameters.
Bug fixes and performance improvements in document processing.

2.13.0

New GET Endpoint /api/v1/config to query the service's current configuration.
New POST Endpoint /api/v1/config to update the service configuration without needing to restart it.
Fix: return the document images in the universal output.
Improved error messages.
Updated ESP Identity document templates.

2.12.1

Updated document templates (Canada).

2.12.0

Updated the OCR extractor.
Added the universal response output format.
Updated facephi-recogntion v4.0.1

2.11.0

Updated templates for RENTA v18.

2.10.1

Fixed a License issue to ensure compliance and the proper operation of the OCR service.

2.10.0

Unified the configuration parameters into a single file.
Updated the OCR extractor.
A new IDCard interpreter with some improvements is available. The previous interpreter is used by default for compatibility reasons.

2.9.1

Set the attribute isCropped of the Endpoint /process to true by default and marked as obsolete (deprecated).

2.9.0

Updated the OCR extractor.
Improvements in log recording.

2.8.1

Updated the OCR extractor.
Fixed the templates for the Mexico IDCard.

2.8.0

Updated templates for RENTA v15.
Fixed the PDF height and width as double type.
Updated the OCR extractor.
Added the Argentina Id Card Version 0004. And removed Version 0003.
Updated the text-interpreter and the templates.

2.7.0

PDF: reduced the default threshold to 0.5 to improve text detection.
PDF: added the optional parameter pageNums. It is a list of page numbers to process. The count starts at 1. If it is not present, all pages will be processed.
PDF: ignored "page not parsed" errors on the 2nd page and onwards.
PDF: added the output warnings. This field contains information about the pages that were not processed. In the future, it could be extended to include other information.
Updated the text-interpreter:
- PDF: the model fields are included even if they are not detected or are empty.
Updated document templates:
- PDF: the IVA.json and RENTA.json models correctly process multi-word fields.
Updated the License backend v2.0.0.
- Added high availability.
Updated the OCR extractor.

2.6.0

Parallelized PDF processing in the OCR engine.
Improvements in log recording.
Bug fixes.

2.5.1

Added support for income tax in El Salvador, form F-11 V17. Model name: RENTA.
Added support for VAT in El Salvador, form F07 V14. Model name: IVA.
Added support for multi-page PDF in the OCR engine.
Improvements in log recording. The log level is propagated from the service configuration to the OCR engine.
Updated the OCR extractor.
Updated the REST engine.
The ForceDetection option of ProcessBody has been deprecated.

2.5.0

Improvements and optimizations in the OCR engine for documents and invoices.
Improvements in the OCR engine for the raw endpoint.
Improvements in log recording.

2.4.6

Improvements in the service health check.
Improvements in license management.

2.4.4

Fixed the Telmex model in special cases such as PUBLICO EN GENERAL as a line in the middle of the address.
Fixed the CFE model when dates do not have the correct format.
Fixed minor interpreter issues.

2.4.3

Added support for Mexico IDCard/ForeignCard.

2.4.2

Fixed dependencies, PDF engine with gcc7.

2.4.1

Fixed an avx512 error in openvino builds for invoices

2.4.0

Updated the Version of Facephi Rest Manager to 2.5.0. New configuration parameters added to the json "logger_path", "logger_level", "logger_rotation", "logger_max_files".
Updated the invoice OCR engine with significant performance improvements.

2.3.4

Fixed the barcode reader for some images from Argentina.

2.3.3

Updated facephi-rest-manager to v2.4.2.
New configuration parameter client_max_body_size to limit the maximum request body size.

2.3.2

BREAKING CHANGE: the endpoint process now receives the parameter files instead of image to allow reading double-sided documents.
Updated the templates for Argentina documents: v1, v2, v3.
Added support for IDCard documents of type OCR, MRZ and barcode.
Added support for IdCard documents with two sides.
Added support for templates based on XML files.

2.3.1

BREAKING CHANGE: the raw endpoint has a new parameter called type which allows you to choose the type of document to process.
Fixed the RENTA template.
Fixed the PDF reader.
Added the ability to configure the OCR service.

2.3.0

The raw endpoint has a new parameter type.
CFE and TELMEX templates fixed.
Added improvements in image processing.
Bug fixes.

2.2.0

Support for new models.
Added text extraction from PDF.
Added PDF to image conversion.
Bug fixes.

2.1.3

Documentation fixed.
Fixed the CFE and TELMEX models.
Modified the result of /api/v1/process to add more information.

2.1.2

Added OCR Latin 2.0
Added the Process, Health, Version and Raw endpoints.

PreviousDeployment on Kubernetes NextRelease notes

Last updated 6 days ago