African foods for deep learning-based food recognition systems dataset.

Autor:	Ataguba G; Department of Computer Science, Dalhousie University, Halifax, Nova Scotia B3H 4R2, Canada., Ezekiel R; Department of Computer Science, Federal University, P.M.B 1154 Lokoja, Nigeria., Daniel J; Department of Computer Science, Federal University, P.M.B 1154 Lokoja, Nigeria., Ogbuju E; Department of Computer Science, Federal University, P.M.B 1154 Lokoja, Nigeria., Orji R; Department of Computer Science, Dalhousie University, Halifax, Nova Scotia B3H 4R2, Canada.
Jazyk:	angličtina
Zdroj:	Data in brief [Data Brief] 2024 Jan 24; Vol. 53, pp. 110092. Date of Electronic Publication: 2024 Jan 24 (Print Publication: 2024).
DOI:	10.1016/j.dib.2024.110092
Abstrakt:	African foods have socio-cultural significance that extends through migration, tourism, and marriage. Africans travel and integrate within the continent through intermarriages. There are over a thousand cultural aspects that differ such as language, food, dressing, beliefs and customs. Food is one of the cultural aspects that Africans embrace quickly as they migrate and integrate socio-culturally. We considered the limited representation of African food research in the HCI community and propose to contribute the rich significant food datasets from two African countries: Cameroon and Ghana. List of Cameroonian foods collected are: Ekwang, Eru and Ndole. In addition, the list of Ghanaian foods we collected are: Jollof Rice, Palm-nut Soup and Waakye. Given the cultural diversity and our study's goal for cultural inclusion, we interacted with at least two locals from the selected countries, and they confirmed that these foods were universally recognized within their respective countries. The datasets were collected from YouTube, Facebook, the field (restaurants), Creative Common Attribution Google Images, and other Creative Commons Attribution sources. A total of 204 images of Ekwang, 206 images of Eru and 205 images of Ndole were collected. In addition, we collected a total of 347 images of Jollof Rice, 392 images of Palm-nut Soup and 400 images of Waakye. We present a meta-data description of the data, quality assessments of our dataset and opportunities for the HCI community to explore in the future. Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. (© 2024 The Authors. Published by Elsevier Inc.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu