The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: TypeError
Message: Couldn't cast array of type string to null
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 544, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 361, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2315, in cast_table_to_schema
cast_array_to_feature(
~~~~~~~~~~~~~~~~~~~~~^
table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
feature,
^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1861, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2071, in cast_array_to_feature
_c(children[name] if name in array_fields else null_array, subfeature)
~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1863, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2155, in cast_array_to_feature
return array_cast(
array,
...<2 lines>...
allow_decimal_to_str=allow_decimal_to_str,
)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1863, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2014, in array_cast
raise TypeError(f"Couldn't cast array of type {_short_str(array.type)} to {_short_str(pa_type)}")
TypeError: Couldn't cast array of type string to nullNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Multilingual OCR Annotation Dataset
Publisher: DataCluster Labs
1. Overview
Optical Character Recognition (OCR) is a computer vision technology that enables machines to detect, extract, and interpret text from images and scanned documents. It plays an important role in document digitization, information extraction, automation, and multilingual AI applications.
The Multilingual OCR Annotation Dataset by DataCluster Labs is designed to support the development, training, and evaluation of OCR and text-recognition models across multiple scripts and real-world environments.
The dataset covers four languages:
- Hebrew
- Thai
- Japanese
- Arabic
It includes both document images and scene-text images, providing examples of text found in structured documents as well as everyday surroundings.
2. Dataset Coverage
The dataset covers two broad categories of visual text data.
A. Document Images
Document images contain text presented in structured or semi-structured formats. Examples include:
- Forms and application forms
- Receipts and invoices
- Official and administrative documents
- Other scanned or photographed documents
These images can support OCR systems designed for document digitization, text extraction, and automated document processing.
B. Scene-Text and Street-View Images
Scene-text images contain text captured in natural, real-world environments. Unlike structured documents, this text may appear under varying lighting, viewing angles, backgrounds, and image-quality conditions.
Examples include:
- Shop signboards and storefront signage
- Billboards and advertising displays
- Shop hoardings
- Street signs and other public-facing signage
- Text appearing in everyday street environments
This category is relevant to scene-text detection and recognition systems that need to identify text in complex visual environments.
3. Language Coverage
The dataset includes text from four languages with distinct writing systems and recognition requirements.
| Language | Language Code | Relevant Characteristics |
|---|---|---|
| Hebrew | he | Right-to-left script |
| Thai | th | Complex character combinations and spacing |
| Japanese | ja | Hiragana, Katakana, and Kanji |
| Arabic | ar | Right-to-left script and connected characters |
This multilingual coverage can help researchers explore OCR performance across different scripts, character structures, and text orientations.
4. Annotation Structure and Formats
OCR datasets can contain annotations at different levels of detail, depending on the intended application and model requirements.
Common annotation levels include:
- Text-level: The transcription of text extracted from an image.
- Word-level: Individual words identified and annotated separately.
- Line-level: Text grouped and transcribed by individual lines.
- Paragraph-level: Larger blocks of text represented as paragraph-level units.
- Region-level: Text regions localized within an image using geometric coordinates.
The available annotation structure depends on the dataset release and project requirements.
Supported Annotation Formats
Annotations can be provided in commonly used machine-readable formats, including:
| Format | Description |
|---|---|
| JSON | Flexible format for structured OCR annotations |
| COCO JSON | Standardized format suitable for object and text-region detection annotations |
| XML | Structured representation of annotation metadata |
| CSV | Tabular representation of image and annotation records |
| TSV | Tab-separated representation, useful for text and metadata records |
The annotation format and level of detail can be selected according to the requirements of a particular OCR workflow.
5. Potential Applications
The dataset can support research and development across several OCR and computer vision applications:
- Multilingual OCR: Training and evaluating text-recognition systems across different scripts.
- Document Digitization: Extracting information from forms, receipts, applications, and other documents.
- Scene-Text Recognition: Detecting and recognizing text in shop signboards, billboards, hoardings, and street environments.
- Text Detection and Localization: Identifying the position of text regions within images.
- Document AI: Supporting intelligent document processing and automated information extraction.
- Multilingual AI Systems: Developing systems that work with diverse languages and writing systems.
6. Why Multilingual OCR Datasets Matter
Building reliable OCR systems requires more than recognizing printed English text. Real-world applications involve different scripts, writing directions, document layouts, fonts, image conditions, and environmental challenges.
Multilingual OCR datasets can help researchers:
- Improve recognition across different writing systems.
- Evaluate model performance on diverse scripts.
- Develop OCR systems for structured documents and natural scenes.
- Investigate challenges associated with text orientation, layout, and image quality.
- Build more inclusive document-processing and computer vision applications.
7. Challenges in OCR Dataset Development
Creating useful OCR datasets requires careful consideration of several factors:
- Script Diversity: Different languages have different character structures, writing directions, and recognition requirements.
- Image Quality: Blur, low resolution, shadows, and lighting variations can affect text recognition.
- Complex Backgrounds: Scene-text images often contain clutter, overlapping objects, and irregular text placement.
- Text Localization: Accurate identification of text boundaries is essential for detection and recognition tasks.
- Annotation Consistency: Standardized annotation guidelines and quality checks are important for reliable training data.
8. Who Can Use This Dataset?
This dataset may be useful for:
- Researchers working on OCR and computer vision.
- Machine learning engineers developing text-detection and recognition models.
- Document AI and intelligent document processing teams.
- Organizations building multilingual information extraction systems.
- Students and academic institutions exploring multilingual OCR.
9. Custom OCR Dataset and Annotation Services by DataCluster Labs
In addition to publishing datasets, DataCluster Labs provides custom data annotation and dataset development services for computer vision and AI projects.
We help organizations convert their existing image and document collections into structured, machine-learning-ready datasets. We can also develop datasets from scratch based on specific project requirements.
Our OCR-related services include:
- Multilingual OCR annotation.
- Text detection and transcription.
- Word-, line-, and paragraph-level annotation.
- Text-region localization using bounding boxes or polygons.
- Document and form annotation.
- Scene-text annotation for signboards, billboards, hoardings, and street-view images.
- Custom dataset creation and annotation in formats such as JSON, COCO JSON, XML, CSV, and TSV.
- Annotation quality checks and dataset structuring.
Our approach focuses on delivering structured, consistent, and project-specific datasets that support real-world AI development.
Website: https://datacluster.ai/
Contact: sales@datacluster.ai/
10. License and Usage
Please refer to the license specified for this dataset release before using, modifying, or redistributing the data.
Users should also verify that their intended use complies with applicable licensing and data-use requirements.
11. Citation
If you use this dataset in your research or development work, please cite it as:
DataCluster Labs. Multilingual OCR Annotation Dataset.
Kaggle: [Insert Dataset URL]
12. About DataCluster Labs
DataCluster Labs works on AI and machine learning solutions, with a focus on high-quality datasets, data annotation, and AI development.
We support organizations and research teams in building reliable AI systems through custom data collection, annotation, and dataset development.
- Downloads last month
- 166