Teklia/NorHand-v1-line

---
license: mit
language:
- nb
task_categories:
- image-to-text
pretty_name: NorHand-v1-line
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_examples: 19653
- name: validation
num_examples: 2286
- name: test
num_examples: 1793
dataset_size: 23732
tags:
- atr
- htr
- ocr
- historical
- handwritten
---

# NorHand v1 - line level

## Table of Contents
- [NorHand v1 - line level](#norhand-v1-line-level)
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)

## Dataset Description

- **Homepage:** [Hugin-Munin project](https://hugin-munin-project.github.io/)
- **Source:** [Zenodo](https://zenodo.org/records/6542056)
- **Paper:** [A Comprehensive Comparison of Open-Source Libraries for Handwritten Text Recognition in Norwegian](https://link.springer.com/chapter/10.1007/978-3-031-06555-2_27)
- **Point of Contact:** [TEKLIA](https://teklia.com)

## Dataset Summary

The NorHand v1 dataset comprises Norwegian letter and diary line images and text from 19th and early 20th century.

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

### Languages

All the documents in the dataset are written in Norwegian Bokmål.

## Dataset Structure

### Data Instances

```
{
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=4300x128 at 0x1A800E8E190,
'text': 'fredag 1923'
}
```

### 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

NorHand v1 - line level 数据集概述

数据集描述

NorHand v1 数据集包含19世纪和20世纪初的挪威信件和日记行图像及其文本。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挪威博克马尔语书写。

数据集结构

数据实例

json { "image": <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=4300x128 at 0x1A800E8E190>, "text": "fredag 1923" }

数据字段

• image: 一个包含图像的PIL.Image.Image对象。注意，当访问图像列（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: 图像类型，数据类型为image。
  • text: 文本类型，数据类型为string。
• 分割:
  • train: 训练集，包含19653个样本。
  • validation: 验证集，包含2286个样本。
  • test: 测试集，包含1793个样本。
• 数据集大小: 23732个样本。
• 标签:
  • atr
  • htr
  • ocr
  • historical
  • handwritten