Faces extracted from Trove newspaper photographs, 1880-1912
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
This is a collection of faces extracted from digitised newspapers available through http://trove.nla.gov.au.
Photographs were harvested from the newspapers using the Trove API. OpenCV2 was used to extract faces from the photographs.
The filenames of the faces include important contextual information. for example:
This filename contains four fields separated by hyphens.
18950209 – date the original newspaper article was published. In this case 1895–02–09 in ISO format, or 9 February 1895.
460 – the Trove identifier for the newspaper in which the article was published. By appending this to a Trove url you can retrieve more information via the web interface or API:
web – http://nla.gov.au/nla.news-title460
API – http://api.trove.nla.gov.au/newspaper/title/460?key=[your API key]
139706071 – the Trove identifier for the article. By appending this to a Trove url you can retrieve more information via the web interface or API:
web – http://nla.gov.au/nla.news-article139706071
API – http://api.trove.nla.gov.au/newspaper/139706071?key=[your API key]
1 – the index of the face extracted from the article.
You can also explore the same dataset using my Face API.
Tim Sherratt (@wragge) email@example.com 24 June 2015