Innovations at Traffic Signals for Cycling: Would a Thermal Imaging Camera Improve Detection

The first person detected by the thermal imaging camera in Portland. 

I presented on the topic "Innovations at Traffic Signals for Cycling" at the most recent meeting of the Oregon Section of the Institute of Transportation Engineers. There were a few new things we had done in the past year that I felt compelled to share information about. The most recent was a Thermal Imaging Sensor. Before I go into detail on that, let's review some of the recent innovations we have implemented in the past year.
In June, we repurposed our school zone enforcement light as a bike detected confirmation light. We have installed these at two locations and the response from their use has been very positive from the user community.  One of the reminders from the meeting was something that Rick Nys, the Clackamas County Traffic Engineer mentioned. He highlighted that the County used feedback from BikePortland.org to guide their improvements of the first bicycle signal in their system. He also mentioned that the response from the bicycling community was so positive that "it makes them want to do more of these types of projects".

The other innovations that I described included the Dutch Bicycle Countdown to Green signal I have described before, but haven't found a way to bring to the U.S.



The newest innovation that we're studying is the use of a thermal imaging camera made by Flir/Traficon. Traficon was recently purchased by Flir and it is interesting to consider a company like this getting into the traffic business.

Flir cameras come from military applications

 That hasn't always worked so well for other companies, but I am hopeful that they will be in it for the long haul. The good news with this situation is that there is an existing expertise with Traficon which is a Belgian company. The company representatives that we met with during the install of this unit suggests that we should be able to do what we want to detect bikes and distinguish their heat signature from that of a car. We'll study the equipment similar to what we did with the microwave and video detection at the Broadway Bridge NW Lovejoy intersection.

Here's a summary of the presentation in pictures.

The location of the test is SE Cesar E Chavez and Lincoln, a popular
bike route that is busy during the commute periods. 

I mentioned it was busy, this was after the morning peak and
people on bikes outnumbered vehicles even on a rainy day.

The push button here is mounted so the person on a bike
can reach it, but it isn't easy. The other direction is impossible to
activate from the street.

A peek in the cabinet of the equipment, the display was temporary,
but we do have access to the images back in the Traffic Operations Center.

I am looking forward to having this studied and learning a bit more about future applications.

Traffic Signal System of Tomorrow

Today at our regional meeting of the transportation operations staff, we discussed the operation of a future traffic signal system that advances beyond mid 1990s software programming. The system would have a graphical user interface that matches the functionalities of Google maps and the system would send text messages based on predefined criteria that alert us to problems such as the signal failing due to a loss of power or a conflicting movement. These are very basic things that we haven't set up to my knowledge. We use the public to tell us when signals are in red flash (and they are pretty good at it too).

We are not so good at knowing if the detection has failed or if the bus is calling for priority like we had set it up to do.

As we discussed that, I came back to reality recognizing that the budgets are tight and getting tighter. Yet there is an increasing need to make decisions based on data and to be data driven. The cross roads for this is a continuous investment in technology and communication that will have our systems report about how well they are operating and if there is some maintenance function that is necessary. The logical next step would be an expert system that prioritizes functions and action for an agency.