Skip to content

This is an IoT-enabled wearable gesture recognition device which captures hand movements using a accelerometer and recognizes correct alphanumeric characters using an SVM Algorithm.

Notifications You must be signed in to change notification settings

amlannandy/Gesture-Keyboard

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Gesture Keyboard

This is an IoT enabled device that uses Machine Learning to type out text/audio messages with hand gestures.

The gesture keyboard module consists of an Arduino Nano micro-processor with an MPU-6050 attached to it, which is turned on and off by a switch. Keeping the button pressed, one has to draw the alpha-numeric character they want to type. It captures the movement of the module through its 6 axis(accelerometer + gyro) and compares it against the 1500+ samples that we have recorded beforehand and predicts the correct character.

We use pyGARL (Python Gesture Analysis and Recognition Library), a Python library which uses SVM (Support Vector Machine) algorithms to analyze the samples that we have recorded and normalizes the data. When the module records a movement, it matches it against the normalized data and predicts the correct one.

Achievements

We made this during Cisco ThingQbator Mid-Cohort Hackathon 2019 and won 1st place.

About

This is an IoT-enabled wearable gesture recognition device which captures hand movements using a accelerometer and recognizes correct alphanumeric characters using an SVM Algorithm.

Topics

Resources

Stars

Watchers

Forks

Languages