Creative Commons source reading · Data quality and algorithm evaluation
OrnAsia: A dataset of asian ornaments for image classification and cultural identification
Nayeem MD; Sraboni SZ; Biswas SS; Islam MM
Originally published in Data Brief · 2025-10-18
From the article
Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.
Abstract
AbstractThis article presents a curated dataset comprising 1088 high-resolution original images of six traditional Asian ornaments: Bangles, Rings, Earrings, Tikka, Necklaces, and Payel. The dataset was developed to support the classification and identification of culturally significant jewellery using computer vision and artificial intelligence techniques. All images were captured in real-world settings, including local markets and remote areas of Mirpur, Dhaka, Bangladesh, using smartphone cameras under natural lighting conditions. The collection process ensured variation in lighting angles, perspectives, and ornament positioning to enhance the dataset’s robustness and realism. The data was organized into six distinct categories, with both training and testing subsets clearly defined to facilitate machine learning model development.…
Editorial reading guide
- What data created and tested the model, and was evaluation independent?
- What is the cost of a false positive, false negative or out-of-distribution decision?
