Skip to content

Nervoload/IRIS

Repository files navigation

IRIS

IRIS: Image Recognition Integrated Signalling - Colour Detection and Dictation for the Visually Impaired

We utilize state-of-the-art Artificial Intelligence to extract & identify subjects & features from live video footage, for active verbal communication of colour information directly in front of the user.

Detecting objects with a Convolutional Neural Network, we then utilize Support Vector Machines to classify different text-colour pairs under particular HSV value thresholds.

To run mannually, run the object_detection file to generate an image, then subsequently utilize the classifier colorProcessing to run the the text to audio generation.

About

IRIS - Colour recognition software from live video footage.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors