Loading...
Loading...
Contributors:
Dataset URI:
wj3r25d8jvThis dataset presents a structured collection of digital images of traditional textile motifs documented in East Nusa Tenggara (Nusa Tenggara Timur/NTT), Indonesia, particularly from the Sumba and Flores region. It was developed as part of a cultural heritage documentation initiative to preserve the visual characteristics and contextual information associated with traditional Indonesian textiles while providing reusable data for image processing and computer vision research. The dataset contains original textile images, augmented images, and structured metadata. Each original image is associated with five augmentation variants: rotation, horizontal flip, vertical flip, brightness adjustment, and grayscale transformation. The relationship between the original and augmented images is explicitly maintained in the metadata, enabling researchers to distinguish field-documented source images from computationally generated variants. Each record contains metadata describing the textile type, motif name, motif category, province, district/city, philosophical meaning, production technique, and documentation date. The collection represents diverse visual categories, including geometric, fauna, flora, human/anthropomorphic, and combined motifs. It also documents several traditional weaving techniques, including tenun ikat, tenun ikat lungsi, and tenun ikat pahikung. The dataset is designed for interdisciplinary research in computer vision, image processing, machine learning, deep learning, digital cultural heritage, and traditional textile studies. Potential applications include motif classification, visual feature extraction, similarity analysis, clustering, content-based image retrieval, transfer learning, and development of automated traditional textile recognition systems. Beyond visual recognition, the inclusion of cultural and geographical metadata enables researchers to connect computational analysis with the cultural context of each textile. The dataset can therefore serve as a foundation for developing AI-based approaches to the documentation, identification, and preservation of Indonesian traditional textile motifs. When developing machine-learning models, augmented images originating from the same original image should remain within the same data partition to prevent data leakage between training, validation, and testing sets.
Keyword(s):
Computer Vision; Image Processing; Machine Learning.
MeSH Term(s):
Grants and Funding:
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)