Datasets
Dataset
odia-handwritten-ocr
by tell2jyoti
Odia Handwritten OCR Dataset Dataset Description This dataset contains 182,152 handwritten Odia character images prepared for training OCR models. The dataset covers all 47 OHCS (Odia Handwritten Character Set) characters with balanced class distribution.
MITimage-classification 77 0100K < n < 1M rows
published 5 Jan 2026
Preview
First 5 rows of default / train, from the Hugging Face dataset viewer.
| image_path | filename | class_id | character | type | augmentation |
|---|---|---|---|---|---|
| Odia_handwritten/Odia_Compound_Character/Odia_Compound_Character/binary/13314_resized_binary.jpg | 13314_resized_binary.jpg | 13 | ଖ | compound | binary |
| data/balanced/augmented_images/class_27/aug_27_brightness_1701.jpg | aug_27_brightness_1701.jpg | 27 | ତ | compound | balanced_brightness |
| Odia_handwritten/Odia_Compound_Character/Odia_Compound_Character/grayscale/15917_resized.jpg | 15917_resized.jpg | 15 | ଘ | compound | grayscale |
| Odia_handwritten/Odia_Handwritten_Simple_Character/Odia_simple_character/Inverted_Odia_SimpleCharacter/0000012407_resized_inverted_binary.jpg | 0000012407_resized_inverted_binary.jpg | 1 | ଆ | simple | Inverted_Odia_SimpleCharacter |
| data/balanced/augmented_images/class_33/aug_33_blur_748.jpg | aug_33_blur_748.jpg | 33 | ଫ | compound | balanced_blur |
Cite this
MITtextCite
Tags
task_categories:image-classificationtask_categories:object-detectionlanguage:orlicense:mitsize_categories:100K<n<1Mformat:jsonmodality:textlibrary:datasetslibrary:pandaslibrary:polarslibrary:mlcroissantregion:usocrodiahandwriting-recognitionindian-languagescharacter-recognition