Identification of yellow/blue ribbons: a simple computer vision problem

Speaker: Chung-hong Chan (Hong Kong)

Community: Hong Kong R Usergroup

Category: Open Source

Language: English (with English Slides)

Tag: Computer Vision Data Science

Advanced
Photo of Chung-hong Chan

About Speaker

Computational Social Science Researcher

About the Topic

人類眼晴可以簡易辨別黃、藍絲帶圖像,但電腦只會看成是一群像素 (Pixels) 。電腦視覺 (Computer Vision) 就是要讓電腦能夠從圖像理解圖像的意義,辨別物件是最常見的電腦視覺問題。

本研究展示如何從雨傘革命期間收集的三十多萬張 Facebook 頭像圖,用開源軟件如 OpenCV, Scikit-learn 等等辨識出黃、藍絲帶等等相關圖像。

Human eyes can identify images of yellow/blue ribbons easily but what the computers see are just a bunch of pixels. Computer vision (CV) is a subfield of artificial intelligence with the aim to teach the computer to see. Object identification is a common CV problem.

The purpose of this study is to demonstrate how to use open source software such as OpenCV and Scikit-learn to identify yellow / blue ribbons and associated objects from over 300k Facebook profile pictures systematically collected during the Umbrella Revolution.