Overview
This release introduces an enhanced OCR capability powered by a custom-trained model designed to improve the extraction of identity document information. The previous OCR implementation relied on a predefined model that supported only a limited number of fields and struggled with documents containing varied layouts, formats, and non-standard structures.
The new OCR model removes these limitations by enabling extraction of all required fields from supported GCC identity documents, improving accuracy and ensuring broader compatibility with different document templates. Additionally, Arabic language support has been incorporated to better process identity documents issued across GCC countries.
Key Improvements
1. Complete Field Extraction
- The new OCR model can extract all configured and required fields from supported identity documents.
- Eliminates dependency on predefined field limitations of the earlier OCR implementation.
- Supports dynamic identification and extraction of document fields.
2. Arabic Language Support
- Enables OCR processing of identity documents containing Arabic text.
- Improves recognition and parsing accuracy for bilingual or Arabic-only documents.
3. Improved Accuracy & Flexibility
- Handles structured and unstructured document formats.
- Supports varied layouts and templates commonly found in GCC identity documents.
- Enhanced field mapping and parsing logic ensures better data consistency.
Supported Touchpoints
The enhanced OCR functionality is available across the following workflows:
- Uploading identity documents from My Profile (Employee)
- Uploading identity documents from Employee Profile (Admin)
- Bulk upload of identity documents by Admin
- Uploading identity documents through Tasks by Employee
- Uploading identity documents through Tasks by Admin on behalf of an Employee
Impact
These improvements ensure that OCR functionality delivers complete and accurate data extraction, allowing customers to fully leverage automated document processing for identity documents across GCC regions.
READ MORE
Comments
0 comments
Please sign in to leave a comment.