| 1. |
- Blomqvist, Christopher, et al.
(author)
-
Joint Handwritten Text Recognition and Word Classification for Tabular Information Extraction
- 2022
-
In: 2022 26th International Conference on Pattern Recognition (ICPR). - 9781665490627 - 9781665490634 ; , s. 1564-1570
-
Conference paper (peer-reviewed)abstract
- In this paper, we present a system for extracting tabular information from loosely structured handwritten documents. The system consists of three parts, (i) a u-net like CNN-based method for text detection and segmentation, (ii) a new attention-based method for simultaneous text recognition and classification of word-parts, and (iii) a method for matching the word parts into a tabular structure for each entry. A key contribution is the observation that the new attention-based recognition and classification module makes it possible for improved spatial analysis of the tabular information. The method is evaluated on a unique historical document: The Swedish Wealth Tax of 1571, consisting of 11,453 pages of hand-written tax records. The evaluation shows that the system provides a significant improvement to the state-of-the-art to the problem of tabular extraction from loosely structured historical documents.
|
|
| 2. |
- Blomqvist, Christopher, et al.
(author)
-
Reading the ransom: Methodological advancements in extracting the Swedish Wealth Tax of 1571
- 2023
-
In: Explorations in Economic History. - : Elsevier BV. - 0014-4983. ; 87
-
Journal article (peer-reviewed)abstract
- We describe a deep learning method to read hand-written records from the 16th century. The method consists of a combination of a segmentation module and a Handwritten Text Recognition (HTR) module. The transformer-based HTR module exploits both language and image features in reading, classifying and extracting the position of each word on the page. The method is demonstrated on a unique historical document: The Swedish Wealth Tax of 1571. Results suggest that the segmentation module performs significantly better than the lay-out analysis implemented in state-of-the art programs, enabling us to trace many more text blocks correctly on each page. The HTR module has a low character error rate (CER), in addition to being able to classify words and help organize them into tabular formats. By demonstrating an automated process to transform loosely structured handwritten information from the 16th century into organized tables, our method should interest economic historians seeking to digitize and organize quantitative material from pre-industrial periods.
|
|
