Teklia/Esposalles-line

---
license: mit
language:
- ca
task_categories:
- image-to-text
pretty_name: Esposalles-line
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_examples: 2328
- name: validation
num_examples: 742
- name: test
num_examples: 757
dataset_size: 3827
tags:
- atr
- htr
- ocr
- historical
- handwritten
---

# Esposalles - line level

## Table of Contents
- [Esposalles - line level](#esposalles-line-level)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)

## Dataset Description

- **Homepage:** [The Esposalles Database](http://dag.cvc.uab.es/the-esposalles-database/)
- **Source:** [IEHHR](https://rrc.cvc.uab.es/?ch=10&com=evaluation&task=1)
- **Paper:** [The ESPOSALLES database: An ancient marriage license corpus for off-line handwriting recognition](https://doi.org/10.1016/j.patcog.2012.11.024)
- **Point of Contact:** [TEKLIA](https://teklia.com)

## Dataset Summary

The Marriage Licenses ground-truth is compiled from the Marriage Licenses Books conserved at the Archives of the Cathedral of Barcelona.

Note that all images are resized to a fixed height of 128 pixels.

### Languages

All the documents in the dataset are written in Catalan.

## Dataset Structure

### Data Instances

```
{
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1244x128 at 0x1A800E8E190,
'text': 'donsella filla de Onofre Esquer morraler de Bara y'
}
```

### Data Fields

- `image`: a PIL.Image.Image object containing the image. Note that when accessing the image column (using dataset[0]["image"]), the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the "image" column, i.e. dataset[0]["image"] should always be preferred over dataset["image"][0].
- `text`: the label transcription of the image.

原始信息汇总

Esposalles - line level 数据集概述

数据集描述

数据集摘要

该数据集的 Marriage Licenses 真实数据来源于巴塞罗那大教堂档案馆保存的 Marriage Licenses 书籍。所有图像都被调整为固定高度 128 像素。

语言

数据集中的所有文档均以加泰罗尼亚语书写。

数据集结构

数据实例

一个典型的数据实例包含以下内容： json { image: <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1244x128 at 0x1A800E8E190>, text: donsella filla de Onofre Esquer morraler de Bara y }

数据字段

• image: 一个 PIL.Image.Image 对象，包含图像。注意，在访问图像列（使用 dataset[0]["image"]）时，图像文件会自动解码。解码大量图像文件可能会花费大量时间，因此建议先查询样本索引再访问 "image" 列，即 dataset[0]["image"] 应始终优先于 dataset["image"][0]。
• text: 图像的标签转录文本。

数据集信息

特征

• image: 图像数据类型
• text: 字符串数据类型

数据分割

• train: 2328 个样本
• validation: 742 个样本
• test: 757 个样本

数据集大小

总共 3827 个样本。

标签

• atr
• htr
• ocr
• historical
• handwritten