Cultural Heritage Images of Palmyra (CHIoP) Dataset

Published: 17 February 2021| Version 1 | DOI: 10.17632/snv3t78t9g.1


Cultural Heritage Images of Palmyra (CHIoP) dataset were gathered by scraping Flickr where images do not have any copyrights. Text search was used to collect CHIoP images, of cultural heritage sites in the Palmyra region in Syria, that have tags related to the search keywords. Later, the collected images were filtered manually to ensure that only images that positively related to each keyword search were included. The CHIoP dataset comprised a collection of 432 images distributed in 4 categories. Accordingly, each category represents a cultural heritage region of interest, then they were assigned names corresponding to each region as follows: Baalshamin (50 samples), Palmyra Theater (171 samples), Temple of Bel (113 samples), and Tetrapylon (98 samples). The CHIoP dataset could help researchers interested in the field of computer vision to test different approaches for image classification and object detection such as machine learning, deep learning, or transfer learning.



South Valley University


Artificial Intelligence, Computer Vision, Culture Heritage, Data Mining, Object Detection, Machine Learning, Content-Based Image Retrieval, Image Classification, Deep Learning