Project Overview
A curated overview of verified Kurdish optical character recognition and handwritten text recognition research. The collection covers printed Sorani OCR, historical print, isolated handwritten characters, handwritten lines and Kurdish scene text. Publications, datasets and reported results are attributed to their original creators and evaluation settings.
Technologies & Methods
Applications
- Document Digitization
- Historical Print Processing
- Handwriting Recognition
- Scene Text Recognition
Related Publications
Exploiting Script Similarities to Compensate for the Large Amount of Data in Training Tesseract LSTM: Towards Kurdish OCR
Saman Idrees , Hossein Hassani (2021)
This study adapts an Arabic Tesseract LSTM model to printed Sorani Kurdish using limited Kurdish ground truth and synthetic font-based training material. The authors report 95.45% average character …
An extensive dataset of handwritten central Kurdish isolated characters
Rebin M. Ahmed , Tarik A. Rashid , Dr. Polla Fattah , Abeer Alsadoon , Seyedali Mirjalili (2021)
This Data in Brief article describes a collection of isolated handwritten Central Kurdish characters. The paper reports 40,940 images from 390 writers and tables containing 35 labels. A later …
A vast dataset for Kurdish handwritten digits and isolated characters recognition
Peshraw Ahmed Abdalla , Abdalbasit Mohammed Qadir , Mohammed Y. Shakor , Ari M. Saeed , Abdalla Taha Jabar , Ali Abdalla Salam , Hedi Hamid Hama Amin (2023)
This dataset paper introduces K-PIT and K-ZHMARA for isolated Central Kurdish handwriting recognition. K-PIT contains 245,000 character images across 35 classes and K-ZHMARA contains 70,000 digit …
Kurdish standard EMNIST-like character dataset
Hamsa D. Majeed , Goran Saman Nariman , Renas Sardar Azeez , Bawar Bilal Abdulqadir (2024)
This Data in Brief article presents an EMNIST-like handwritten character dataset for Sorani Kurdish. The resource contains 58 character-shape categories and the repository reports 406,000 images. The …
Making Old Kurdish Publications Processable by Augmenting Available Optical Character Recognition Engines
Blnd Yaseen , Hossein Hassani (2024)
This 2024 arXiv preprint adapts Tesseract 5 for historical printed Sorani Kurdish publications using transcribed lines from pre-1950 material. The abstract reports 1,233 line images with …
KSTRV1: A scene text recognition dataset for central Kurdish in (Arabic-Based) script
Sardar Omar Salih , Karwan Jacksi (2025)
This Data in Brief article introduces a scene-text dataset covering Kurdish Sorani and Badini in Arabic-based script alongside Arabic and English. It reports 1,420 natural scene images, 19,872 real …
A Modified Vision Transformer for Kurdish Cursive RTL Handwritten Text Recognition
Faraedwn M. Salih , Abdulbasit K. Al-talabani (2026)
This 2026 Research Square preprint presents a modified Vision Transformer approach for handwritten Sorani Kurdish line recognition and introduces DASNUS. The dataset contains 11,475 lines from 867 …
Related Datasets
KurdishOCR Ground Truth
Published: January 1, 2021 | Size: 522 line-image/transcription pairs reported by the associated paper
Ground-truth material released for printed Sorani Kurdish OCR research by Saman Idrees and Hossein Hassani. The associated study reports 522 line-image/transcription pairs. The repository distributes …
Central Kurdish Isolated Handwritten Characters
Published: January 1, 2021 | Size: Approximately 41,000 isolated handwritten character images; exact current v3 total not independently verified
A public collection of isolated handwritten Central Kurdish characters associated with the 2021 Data in Brief publication. The original paper reports 40,940 images from 390 writers, while a later …
K-PIT and K-ZHMARA
Published: December 22, 2022 | Size: 315,000 images combined
A deposited Central Kurdish handwriting bundle containing K-PIT for isolated characters and K-ZHMARA for digits. K-PIT contains 245,000 images across 35 classes and K-ZHMARA contains 70,000 images …
Kurdish Standard Characters EMNIST-like Dataset
Published: July 13, 2023 | Size: 406,000 images reported by the repository
An EMNIST-like handwritten-character resource for Sorani Kurdish. The repository reports 406,000 images across 58 character-shape categories. The associated paper verifies Sorani coverage; Latin …
KurdishOldTextsOCR
Published: January 1, 2024 | Size: 1,203 TIFF images and 1,203 text files in the inspected release
Ground truth for OCR of historical printed Sorani Kurdish publications. Direct archive inspection found 1,203 TIFF images and 1,203 text files. The associated paper reports 1,233 line images, so the …
KSTRV2
Published: September 7, 2025 | Size: 1,420 scene images, 19,872 real word samples and 20,000 synthetic instances reported for the multilingual collection
Corrected public release of the Kurdish scene-text recognition resource associated with KSTRV1. The multilingual collection covers Sorani and Badini in Arabic-based Kurdish script alongside Arabic and …
DASNUS
Published: March 10, 2026 | Size: 11,475 handwritten lines from 867 writers
A Sorani Kurdish handwritten-line dataset containing 11,475 lines from 867 writers. The creators define writer-separated training, validation and test splits containing 8,042, 1,700 and 1,733 lines …
KHLD - Kurdish Handwritten Lines Dataset
Published: April 22, 2026 | Size: 47,944 handwritten line images reported by the creators
A public Sorani Kurdish handwritten-line resource created by Mohammed S. Ibrahim, Hevi Ahmed, Yassin Hussein, Tarik A. Rashid and Polla Fattah. The creators report 47,944 line images and 4,802 …