Abstract
This study evaluates the recently proposed Document Attention Network (DAN) for extracting key-value information from Uruguayan birth certificates, handwritten in Spanish. We investigate two annotation strategies for automatically transcribing handwritten documents, fine-tuning DAN with minimal training data and annotation effort. Experiments were conducted on two datasets containing the same images (201 scans of birth certificates written by more than 15 different writers) but with different annotation methods. Our findings indicate that normalized annotation is more effective for fields that can be standardized, such as dates and places of birth, whereas diplomatic annotation performs much better for fields containing names and surnames, which can not be standardized.
| Original language | English |
|---|---|
| Title of host publication | Document Analysis and Recognition: ICDAR 2024 Workshops, Proceedings |
| Editors | Harold MOUCHÈRE, Anna ZHU |
| Publisher | Springer, Cham |
| Pages | 40-54 |
| Number of pages | 15 |
| ISBN (Electronic) | 9783031706455 |
| ISBN (Print) | 9783031706448 |
| DOIs | |
| Publication status | Published - 2024 |
| Externally published | Yes |
Publication series
| Name | Lecture Notes in Computer Science |
|---|---|
| Volume | 14935 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Bibliographical note
Publisher Copyright:© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
Funding
The research that originated the results presented in this publication was partly supported by the Agencia Nacional de Investigación e Innovación (ANII) and the France 2030 CollabNext project.
Keywords
- Automatic information extraction
- Birth certificates transcription
- Handwritten text recognition
- Normalized and diplomatic annotation