Recognized / Not Recognized, (Looking for the popular news image algorithm)


The project Recognized / Not Recognized revolves around the question: ‘What features determine ‘successfull’ news photos?’

From a selection of worldwide press agency databases (which include amateur witness photos) I scraped all the news photos in relation to the ten events that received the most coverage in the past five years. This resulted in a database of approx. 850.000 images. By using image recognition software from Google (and some proxies), I determined how many times an individual image exists on the google indexed internet. In other words, the reproduction rate of each image is calculated.

Press photos from a big news event have a wide range of variables but are in a way very uniform - since they are all witness photographs –, therefore my database consists of an ideal set for a cross-sectional study. 

uv-print on transparant sheet, plexiglass 2cm, stickers, tl-light. Steel frame 3m x 70cm -> 2 pieces

Coding: Muhammad Atif Ayaz, Emile den Tex

Made possible by: Stimuleringsfonds voor Creatieve Industrie, Amsterdams Fonds voor de Kunsten & Museum of the Image, Digital Methods Initiative - University of Amsterdam.

Extra Magazine, spread