Kurdish Sign Language(KuSL2023)Dataset

The KuSL2023 dataset, a compilation of 71,400 sign language images, amalgamates the ASL alphabet dataset and the ArSL2018 dataset, encompassing 34 sign language combinations. The ASL alphabet dataset contributes 78,000 RGB photos, while the ArSL2018 dataset, assembled in Al Khobar, Saudi Arabia, by forty contributors, comprises 54,049 images representing the Arabic alphabet. Clean-up procedures, including centering pictures and resolving imperfections, were employed to standardize the images. Notably, the 23 most commonly used letters from the Kurdish Alphabet–Arabic script were derived from the ArSL2018 dataset, with the remaining Kurdish alphabet letters sourced from the ASL alphabet collection. Researchers and developers can access the KuSL2023 dataset for training and testing machine learning models for sign language recognition and related tasks, with the dataset freely available for non-commercial use.