Our aim is to detect eye movements in a way to make drawings on the screen. This is not only a fun project but also makes applications more accessible to people with different abilities.
We worked with Arduino UNO and an Arduino camera module (OV7670) to detect the eye movement by using C++. Then we moved on to Raspberry Pi 3 and used Python to gather the data to display. We wanted to upload and store OpenCV which is an API used for image detection. Since the size of the library was too large, we needed to upload it on the server Microsoft Azure. We used Pygame to convert the movement of the eye into drawing.
What we learned?
Since we initially struggled to get an image from the Arduino Uno this made us realise how important it is to conduct research about which hard-/ and software is most suitable and the easiest to use for tackling the set challenge prior to getting started. Moreover, we could see the difference in effort and work required by two different approaches and hence understand how essential it is to brainstorm in all directions before committing to one path. Besides, we got an insight into two different hard-/ and software systems by working with the Arduino UNO (C++) and the Raspberry Pie 3 (Python). Furthermore, we learned how to incorporate the diverse skill sets each team member brought along.
Outlook on further applications
The system we have built can be integrated into different devices. Of course, this can be used as a tool to create Art by eye movement only. However, this application is relevant for other disciplines too.
Applications can be found in several fields, for instance scientific research, as neuroscientists investigate how we perceive images (on which parts of the image we focus) when seeing it for the first time in comparison to seeing a known image.
It would be of significance for people who cannot access electronic devices due to physical circumstance since this application can be used to enable the interaction with an electronic device, such as a laptop, by eye motions only.
It bears relevance to marketing research too, since several eye motions can be layered over one another to analyse what in particular attracts the attention of customers when presented with an advertisement.