Semiwiki 400x100 1 final
WP_Term Object
(
    [term_id] => 95
    [name] => Automotive
    [slug] => automotive
    [term_group] => 0
    [term_taxonomy_id] => 95
    [taxonomy] => category
    [description] => 
    [parent] => 0
    [count] => 750
    [filter] => raw
    [cat_ID] => 95
    [category_count] => 750
    [category_description] => 
    [cat_name] => Automotive
    [category_nicename] => automotive
    [category_parent] => 0
)

New Vision for Traffic Cameras in 2016

New Vision for Traffic Cameras in 2016
by Roger C. Lanctot on 01-29-2016 at 7:00 am

 A picture may be worth a thousand words, but a traffic camera video may be worth millions. TrafficLand is poised to transform forever the use of traffic camera video information with demonstrations at the CES show in Las Vegas. In the process, TrafficLand will be overcoming years of industry ambivalence toward the use of traffic camera info across a wide spectrum of transportation applications.

We have a strange relationship with traffic cameras in the U.S. and elsewhere in the world. We like to see the traffic monitoring cameras, but we hate to see speed and red-light cameras.

Traffic cameras gather video, usually low quality, that can be viewed online or via television or mobile devices or on the massive video walls in local Department of Transportation traffic management centers. We use video cameras to monitor traffic conditions and still cameras to enforce speed limits and intersection controls.

We are also using still cameras to gather license plate information at toll booths or via police car-mounted devices. Lastly, we have cameras mounted in police cars and on police officers to gather incident information.

Video and still cameras are increasingly seen as the gold standard, the reliable source of ground truth. But we are simultaneously ambivalent regarding these valuable sources of information.

The past year has seen multiple police shootings caught on camera. Most of the cases that have made it to the evening news have involved police officers shooting and often killing sometimes unarmed black people.

In what has become a well-established pattern, there often seem to be unnecessary administrative delays in releasing the video connected to these incidents. When the video is released it often appears to provide incontrovertible proof of police misconduct. Yet time after time authorities find no wrong-doing by the officers involved.

When it comes to speed and red light cameras it is equally complicated, though lives are not directly at risk. Researchers such as the Insurance Institute for Highway Safety (IIHS) have concluded that red-light cameras and speed cameras are effective at achieving their stated purposes of enforcing speed limits or reducing red light running, but opposition to these devices continues to grow even as they proliferate.

Typically, speed and red-light infractions enforced by cameras do not come with driver’s license demerits and, in some cases, are not considered enforceable civil violations. Opposition to these systems has arisen from groups with three main objections:

[LIST=1]

  • Violation of privacy
  • The use of cameras as a means to raise revenue
  • The unenforce-ability of these privately generated tickets (as in the recent case of Round Rock, Tex. – http://tinyurl.com/jazhpok – Round Rock Ending Red Light Camera Program)
  • The belief that red light cameras reduce red light running but cause an increase in rear-end collisions for drivers attempting to stop short

    License-plate reading cameras appear to be seeing broad acceptance as a tool for collecting tolls or for catching electronic/cashless toll violators. There appear to be some objections in some regions of the U.S. and the world to unlimited license plate reading.

    One of the most hotly contested issues worldwide is the usage and storage of all of these types of camera information for law enforcement purposes. Policies vary on the amount of time the information gathered from cameras can be stored and the ways that it can be used. These battles continue, especially in the context of law enforcement priorities in the U.S. and the more privacy-obsessed environment in Europe.

    Generally speaking, it looks like police dash and body cams are here to stay and we will be seeing more of these videos in 2016 and beyond. License plate readers, too, will see wider adoption, particularly for toll collection and law enforcement. And red light and speed cameras are popping up all over the landscape even as local resistance intensifies even in the face of IIHS endorsements.

    In this wider context, traffic video cameras play a unique role. Traffic video cameras represent, in some ways, the ultimate ground truth when it comes to traffic conditions.

    Traffic information providers struggle to determine current and expected traffic conditions and, by extension, the travel times for different routes. This information is vital to traffic management executives, fleet operators, commuters and average citizens. But there is no universally accepted means for measuring traffic quality or accuracy, which gives traffic camera video a unique source of power and influence.

    Perhaps the most frustrating aspect of traffic information are the various sources of “lag” in the systems for delivering traffic data – real-time and predictive. There is lag in identifying events, in wirelessly communicating the information, and in interpreting the impact.

    A wide range of apps from the ubiquitous Waze to TomTom, HERE, Telenav, iGo, and Navmii integrate local traffic and routing information but have no means to overcome this inherent lag. The traffic information is comprised of what is called “flow” data (the calculated speed per road segment and time) and “journalistic” data (the details including time, location and status of traffic-impacting events – road works, sporting events, jams and/or accidents).

    In some respects, traffic camera video is like getting the homework answers from the back of the teacher’s guide. Why use algorithms if you can just look at the cameras? Seeing is believing, after all.

    Few of the listed apps enable access to traffic camera video, which is available in the U.S. from TrafficLand. Navigation application suppliers prefer to provide an overlay of traffic information analysis with route guidance rather than providing the raw video.

    The challenge in providing raw video feeds is that the user must look directly at the video to understand precisely what is happening. TrafficLand established itself as the lead go-to source for this information six years ago with in-dash proofs of concept with multiple industry suppliers. Only a handful of these implementations made it to vehicle dashboards.

    The low level of adoption seemed like a statement by the automotive industry that their customers “couldn’t handle the truth” – ie. the ground truth of live traffic camera video. It was too distracting. At the same time, there is a risk that the information contained in traffic camera video will contradict the traffic information available in the on-board app created by the car maker. In essence, real-time traffic camera video has the potential to undermine the credibility of existing app-based traffic information.

    TrafficLand has taken all of these concerns to heart and is in the process of redefining the collection and interpretation of traffic camera video. Thanks to the work of partners including Harris Corporation, Metrotech, TrafficVision and others, TrafficLand is converting its traffic camera feeds into data – in real-time – in a way that can be integrated with existing algorithm-based traffic solutions.

    TrafficLand’s new real-time data feeds will include speed by lane, headway (distance between cars), stopped vehicle in roadway, vehicle count, wrong way vehicle, intersection pressure/performance and incident detection. These enhancements come along with weather-related information including roadway visibility, surface wetness, snow and, later in 2016, precipitation and road ice.

    TrafficLand is benefiting in part from DOT’s that are stepping up the quality of their camera installations to higher resolution equipment. But TrafficLand’s approach to converting the video feeds to actionable data is revolutionary and proceeding regardless of the quality of the existing video sources.

    TrafficLand is the largest integrator of DOT traffic cameras with more than 32,000 DOT and other roadway cameras available worldwide. The company is best known for its DOT installations in the U.S. but is rapidly expanding its global operations.

    TrafficLand’s efforts are creating new types of information to assist drivers – most notably its visibility calculation – and new ways of identifying incidents directly from the camera feed – as illustrated in the pictures at the top of this post. (The sequence of three images, seconds apart, demonstrate the ability of TrafficLand to identify a traffic incident – in this case a collision – in real-time.) In fact, TrafficLand’s ability to identify incidents within seconds of their occurrence will put pressure on app developers to integrate and deliver that information in a timely manner.

    So, 2016 begins with major advances in traffic video collection, interpretation and distribution. TrafficLand has enabled a major advance in establishing the real-time ground truth of traffic information. It remains to be seen how quickly car makers and app developers can respond and take advantage of the new resource.

    More articles from Roger…

    Share this post via:

  • Comments

    0 Replies to “New Vision for Traffic Cameras in 2016”

    You must register or log in to view/post comments.