Interview with finalist for “Game Changer” AutoSens Award – Brian Deegan, Vision Research Engineer at Valeo
In a series of articles, we will be conducting short interviews with the shortlisted finalists for the inaugural AutoSens Awards 2017. We recently caught up with Brian Deegan, Vision Research Engineer at Valeo who has been nominated for the “Game Changer” Award.
Congratulations on being shortlisted for the “Game Changer” AutoSens Award. How does it feel to be a finalist?
This came as a very genuine surprise to me, I didn’t expect to get nominated at all. When I look at the other nominees in the category, they’re all very successful talented people. It’s an honour to be considered in the same category.
The nomination which has now helped you become a finalist cited your research on LED flicker – tell us more about that?
Watch AutoSens TV ~ (Click to subscribe)
► Join the AutoSens Awards celebrations in 2018
► Highlights from AutoSens in Detroit 2018
► Interview with Dave Tokic from Algolux at AutoSens Detroit 2018
The LED flicker work I am involved with came about as part of the IEEE P2020 working group on Automotive Image Quality standards. One of the challenges facing our industry is the lack of agreed standards for assessing camera image quality performance. When I found out that there would be a category specifically covering LED flicker, I asked if I could lead the working group, and I’m happy my offer was accepted.
To give a bit of background, LED flicker is an artifact observed in digital imaging where a light source appears to flicker (i.e. the light may appear to switch on and off or modulate in terms of brightness or colour), even though the light source appears constant to a human observer. This occurs when a light source is powered by a pulsed signal. The pulses occur at a high rate and appear constant to a human observer. LED lights have huge advantages over traditional lights, in terms of power saving and reliability. However, if a camera captures an image of this light with a short exposure time (e.g. during bright, sunny conditions), it is possible that the light will appear “on” in some images, and “off” in others. If you watch a video stream of an LED light, it will therefore appear to flicker on and off.
LED flicker poses a major challenge for automotive imaging. LED lights are now widely used as traffic lights, speed signs, traffic information signs etc. So for example, if you are using a camera for traffic light recognition, a red light may not be detected because of LED flicker. I have also seen several cases where LED running lights on a vehicle can appear to blink on and off at the same rate as a turn signal. This is a major concern for mirror replacement applications. A driver may mistake a flickering headlight for a turn signal. I have also heard of cases where drivers changed lanes to make way for a vehicle because they mistook a car with flickering headlamps for an emergency vehicle. So there are clearly definite safety concerns around LED flicker.
I was determined to get involved in LED flicker standards because I really felt that I understood the problem and its implications well, and I felt that I could help define a meaningful standard for the industry. So far, the work of the LED flicker working group has centered around defining the problem and defining standardized methods of assessing a camera system’s susceptibility to LED flicker. This will allow the industry to accurately characterize and benchmark a camera systems capabilities relating to LED flicker mitigation.
I am very happy with the level of engagement on this topic from across the industry and I believe we are making very good progress in this area.
You have a MSc in biomedical engineering – can you provide any insight as to what other medical technologies we will see more of in the automotive sector?
Biomedical engineering has already made it’s way into the automotive sector. A good example would be driver monitoring. By analyzing a drivers patterns, facial expressions, eye movements etc, we can already tell if a driver has become drowsy and provide an alert.
An extension of this will probably come in the next few years as cars become more autonomous. One key challenge already identified will be how to safely hand over between autonomous and manual driving. Monitoring the driver’s alertness state will be key to handling this transition safely.
In my biomedical engineering days I also did a lot of work on motion sickness. I was actually fortunate enough to work on a motion sickness study at NASA’s Johnson Space Center in Houston. I do think motion sickness will be a key challenge for fully autonomous vehicles. We’ve all seen the concept drawings of autonomous vehicles where people are reading etc while the car drives itself. Personally, I can’t read in a moving vehicle, I’ll get motion sick. Cyrel Diels gave an excellent talk on this topic during the 2016 AutoSens in Brussels. I do think minimizing the likelihood and severity of motion sickness will be a factor in the future designs of autonomous vehicles. It will probably influence cabin design, climate control, passenger monitoring etc.
Do tell us more about your personal and professional values
I think for me the most important thing is hard work. Talent, ability, whatever you call it, will only get you so far. To get the best out of yourself and those around you takes hard work. I was never the smartest person in the class, but I’ve usually been able to make up for it by working hard, by being passionate, by caring about the details, and by not being satisfied until you really get to the heart of whatever issue you’re working on.
Honesty and integrity are very important to me. If I make a mistake, I’ll own up to it. If I don’t understand something, I’ll ask the question. And I give credit to others where credit is due.
It’s a bit of a cliché, but teamwork is also vital. I’ve always tried to think of good will as being a currency – you build it up by doing a turn for others, and then you can spend it when you need a favour from someone else in turn. I find this the best way to think about it, to make sure there’s a good spirit of cooperation, that no one feels taken advantage of.
- Seeing in the dark – unique deep learning approach puts finalist Algolux in reach of the 2018 AutoSens “Software Innovation” Award ~ 15 August 18
- Robust and reliable localisation required for autonomous driving ~ 15 August 18
- The autonomous world is evolving and Velodyne will be a part of it ~ 14 August 18
Finally, in my own career I’ve come to a bit of a transition point. I’m now in a bit more of a senior role, and I got to this position through hard work on my part and the assistance of colleagues who took an interest and helped me out at crucial times. I now see it as part of my role to do the same for more junior colleagues. I’d like to be able to give them the same kind of support I got when I was in their position.
We meet lots of folk from Valeo, always happy – always invested in the company – why is that?
I think it’s because we see the value in what we’ve doing. Transportation as we know it is undergoing the biggest revolution since the combustion engine – and it’s a revolution that is badly needed. The work we do, not just in Valeo, will ultimately save lives and will hopefully go a long way to tackling climate change. This is an area worth working in.
Join us in wishing Brian Deegan all the best for being a finalist at the AutoSens Awards. View the full shortlist
Why not join the celebrations on 20 September at the Atomium Brussels – book your tickets to the Awards here