Digital Humanities
@ Pratt

Inquiries into culture, meaning, and human value meet emerging technologies and cutting-edge skills at Pratt Institute's School of Information

Content Based Image Search

The amount of photos taken has increased exponentially since the popularization of digital photography. According to Facebook’s annual report “On average more than 350 million photos per day were uploaded to Facebook in the fourth quarter of 2012. Over 240 billion photos have been shared on Facebook.” That breaks down to 4000 photos uploaded per second. The sheer amount of images and the frequency with which photos are reposted can make it difficult to locate information about an image.

How does this relate to digital humanities? In DH projects there may be multiple people involved in creating metadata, often through students or crowdsourcing. Inconsistencies can occur and make it nearly impossible to locate information about an item through keyword searches. One approach to tracing instances of a photo is through a content based image searche. I will demonstrate three search engines that can be used for this process.

Leave a Reply

Your email address will not be published. Required fields are marked *