Ontario Innovators: Studying the human eye to develop robots — and surveillance systems — that may keep you safer

By Daniel Kitts - Published on October 11, 2016
York University’s Doug Crawford (left) is fascinated by the way the human brain processes eyesight. (York University)



An aging population means an increasing number of people have vision problems that can significantly affect their day-to-day lives. At the same time, technology has advanced to the point that a growing number of machines are able to “see” — in some ways more clearly and precisely than any humans can.

It’s in this world of fading eyesight and watchful machines that York University’s Vision: Science to Applications (VISTA) operates. The interdisciplinary research program recently received a $33.3 million grant from the federal government to continue its work. TVO.org spoke to Doug Crawford, VISTA’s scientific director, about how much is still unknown about human eyesight and how studying it can potentially lead to medical and technological breakthroughs.

Most people take eyesight for granted. What do you find so fascinating about it?

What fascinates me about vision is that it doesn’t really end with the eye. It really involves so much of the brain. And I’m a neuroscientist who works on how we use vision to guide movement and to keep track of what things are and where they are as we walk through space.

You once said many of the things you study are mundane activities like washing the dishes. If they’re mundane, why study them?

The mundane things are the things we take for granted when we’re healthy. Washing the dishes, driving your car, going to work — so many neurological conditions can affect those processes and degrade not only your quality of life but your productivity at work. And because vision involves so many areas of the brain, it ends up getting affected by a lot of different neurological disorders, everything from Alzheimer’s to Parkinson’s to stroke.

We can’t think of vision in isolation, [thinking] it’s the eyes or even one part of the brain. It’s an entire system that involves the entire brain and the entire body.

York University vice president Rob Haché has said that some of VISTA’s research will look at vision science “at the interface between humans and technology.” Explain.

Our computer science team can really make important contributions to people’s wellbeing and safety. An example of that is an investigator name Michael Jenkin who specializes in visually-guided robotics. Another investigator named John Tsotsos has been involved in the development of assistive robotic devices. There’s a very big interest right now in elder-care facilities and hospitals having robots that can do simple tasks to help out the elderly: assist them in getting them from here to there in the hospital, or making sure they’re safe within their home environment. Those things were science fiction 10, 20 years ago, but now they’re becoming reality.

Some of the research VISTA is doing is in attentive visual machines. What are those?

We take for granted that we have this wonderful faculty of attention as human beings. We’re surrounded by so much information —way more than the brain could process. And the way that we’ve learned to deal with that is that we focus our attention on the task at hand. And that’s something we just do automatically. When you’re developing a robot that has to perform certain tasks you need to choose what part of the visual environment the machine pays attention to, because if you try to process everything that’s out there, it becomes a huge computational problem.

VISTA will also look into how vision can be used in surveillance technology, which already makes many people nervous. What new types of surveillance technology will VISTA explore?

More and more cameras are being used to monitor what’s going on in our environment. Right now you picture a person that’s sitting in a room with a whole bunch of video monitors. And if you watch police dramas, like I do sometimes, the person often misses what’s going on. So what you need are ways of detecting events that are unusual and possibly dangerous. And it isn’t just in neighbourhoods. It could be flying over a region where you’re concerned about fires starting, and how does the video system detect that there’s an event that the pilot wouldn’t see themselves.

You mentioned the public concern of “Is Big Brother watching me?” That obviously is an ethical issue and it’s one that we intend to address head-on. It’s certainly something that we don’t take for granted as we move forward with these technologies.

This interview has been edited for length and clarity. 

Thinking of your experience with tvo.org, how likely are you to recommend tvo.org to a friend or colleague?
Not at all Likely
Extremely Likely

Most recent in Politics