New York State Bridge Security Case Study

NYSBA Improves Bridge Security and Operation with Smart Thermal Cameras

The Hudson River is an essential waterway for the flow of commerce on the East Coast. The watercourse spans 315 miles, starting in upstate New York, flowing past the capital of Albany, and continuing down to New York City, where it forms a border with the state of New Jersey. The New York State Bridge Authority (NYSBA) is a government-owned entity that maintains and operates five of the bridges spanning the Hudson River: the Bear Mountain Bridge, the Newburgh- Beacon Bridge, the Mid-Hudson Bridge, the Kingston-Rhinecliff Bridge, and the Rip Van Winkle Bridge.

“The Hudson River is a vital pathway to and from the Port of Albany, so the security of the bridges and the water under the bridges is paramount,” says Greg Herd, director of information technology at NYSBA. He and his IT department are in charge of the technologies used at the bridges to monitor traffic and toll transactions, and provide an overall secure environment. He says bridge security challenges on those structures include trespassers, inclement weather, and traffic incidents ranging from minor fender benders to serious accidents. “Without the technologies that we have at our disposal now, [security] would be an extremely difficult task,” says Herd.

9/11 Spurs Major Change in Bridge Security

This article originally appeared in Security Management, October 2013.
This article originally appeared in the October 2013 Security Management.

The 9-11 terrorist attacks caused a major shift in the way the NYSBA approached bridge security. “Since {then] we’ve had several major security initiatives at the bridge authority both in monitoring and in access control,” says Herd. “It had initially started out mainly as perimeter fencing and perimeter alarms in the traditional sense, with motion detectors and fence vibration devices in the plan.”

Initially, research into video motion detection resulted in the finding that the software wasn’t mature. It yielded “a tremendous number of false alarms, and no products that I found were able to detect stopped motion,” he says. “That was something that we had wanted to implement so that we could detect vehicles that may have stopped on the roadway. Response time is paramount when you’re on a three lane highway with very little shoulder room and you have a vehicle stop over.”

In July 2011, a consultant suggested Herd take a look at the SightLogix SightSensor, a motion-based video surveillance tool, for possible use on and around the NYSBA bridges. “As we looked into SightLogix, we quickly realized that the analytical capability of their product brought to us everything that we’d been looking for,” says Herd. They implemented the system in March 2012.

Eliminating Nuisance Alerts is a Must

As Herd points out, one of the biggest challenges the sensors help them manage is vehicle traffic. And the NYSBA bridges get a lot of it. According to a recent report, more than 58 million vehicles crossed the five bridges combined in 2012; the busiest of those bridges, the Newburgh-Beacon Bridge, saw nearly 25 million vehicles cross last year.

The analytic capabilities of the SightSensor also allow operators to know when a car slows down to a certain speed, and operators can set parameters to target the alerts more precisely. ”You can say, ‘I only want to know when an object is doing less than three miles an hour, and it’s this particular size,” which will help eliminate small animals from triggering an alarm, says Herd.

SightLogix worked with Herd and systems integrator Transdyn, which designed and installed the sensors for the NYSBA. A company spokesperson explains that the SightSensor product offers an array of capabilities that make it ideal for a bridge environment, including being able to handle all types of weather. He referred to the version that uses thermal imaging, which is what the NYSBA installation has. “Thermal cameras can see in any lighting condition, so there is no need for adding visible lighting. [The image) looks the same during the day as at night, for the most part. It does much better in inclement weather, so if it’s raining, if it’s snowing, if it’s foggy, these devices can see what a normal camera, which looks at the visible spectrum of light, would miss,” he says.

Thermal Imaging Mitigates Impact of Weather and Lighting

Inclement weather and light-reflecting river water are notorious for causing false alarms. SightLogix points out that the thermal imaging also comes in handy when dealing with car headlights, reflections off water, shadows from clouds going by, and “all sorts of things that are outdoors that make the average visible camera system a poor choice for doing automated detection.”

Another offering of the SightSensor is what SightLogix calls geo-referencing. The company notes that each pixel in the images generated by the sensors is associated to a GPS position, giving the sensor even more intelligence about the size of objects in the field of view. “In effect, we’re making [the sensor) very, very smart by saying, ‘we’re going to tell you where everything is, not only the latitude and the longitude, but how big an object is, where it is, what direction it’s moving in, and how fast.”

Herd notes that the geo-referencing capability has allowed them to set up zones around the bridges to better deal with boat traffic. “[The] SightLogix product enabled us to configure zones to say, ‘okay, if a boat is moving or a vehicle is moving, not a problem, but if they stop for more than [a certain] amount of time, go into alarm.’ Then we needed to have that alarm coming to the local facility, but then also feed the alarm into a command center that monitors all of our facilities.”

The camera system also reduces false alarms with electronic image stabilization, which removes movement caused by wind and passing cars, a necessity for reducing false positives in a bridge application.

The NYSBA uses a video management system from Videolnsight, which easily integrates with the SightLogix sensors, according to Herd. When an alarm is triggered, the SightSensor passes the alarm into the video management system. Herd explains that the system can be programmed to take a series of actions, such as sending an e-mail or putting text into a variable message sign.

The NYSBA monitors the incoming images and alarms from a central station, says Herd, but it also has backup locations in the event of an emergency. So whenever there is an incident or an accident that impedes traffic, the system sounds an alert. Operators can then dispatch help to the scene and use the system to monitor and evaluate the event.

A “Huge Return on Investment”

The system has been operational since 2012. Herd says that the sensors have provided a huge return on investment and that the NYSBA relies on the technology on a 24/7 basis. “If we didn’t have the SightLogix sensors, we would be looking at traditional methods of blocking access and monitoring individuals. And that would mean a considerable amount of chain link fence, traditional motion detectors, and all the equipment that goes behind traditional detection,” he says.

“As an example, one of our sites has a large area by the river where there’s pretty much free access alongside woods. If we had to secure that area without the SightSensors, it would be thousands of feet of chain link fence augmented with outdoor motion sensing devices. And then we would need low-light or thermal surveillance cameras to investigate every time a motion sensor tripped,” he explains.

Herd notes that moving forward, NYSBA will continue to keep an eye out for products that can help secure the bridges and waterway even better. He says that he is always looking at new technology and assessing whether it might reduce costs and increase security.

This article first appeared in Security Management, October 2013. 



4 Comments

  • Donald Graham

    A very good article. However, the article spins only the positive about IR and motion-based analytics, but doesn’t admit to the proven negatives, such as: 1) IR does not perform well in fog, high moisture content, rain, snow, and ice. In my experience, especially in New York applications, IR peformance can frequently degrade in a water based environment (even midwave IR cameras). 2) Motion-based video processing engines’ performance is directly impacted by the ability of the camera to “see” the environment. In the studies that Raytheon has performed, IR cameras certainly mitigate the problems mentioned above, and their performance is certainly better than day/night visible cameras. However, they still suffer from weather related events, and in New York, this is a very common problem.

    Having said all of that, I have looked at SightLogix for potential Raytheon applications, and I feel their products are some of the best in the business.

    • Donald, thanks for your comment. As you say, fog and other weather can impact a thermal camera’s performance and needs to be considered when designing a system. And I agree that thermal cameras will do significantly better than visible cameras in these situations. SightSensors help address this concern with thermal cameras as they use a much higher amount of on-board image processing to detect very small temperature differences and can maintain robust detection accuracy under a broader range of conditions, including fog.

      Here’s a real world example of a SightSensor detecting an intruder at 200 meters during very dense fog – you can see the thermal camera (at left) in this case was doing much better than the visible camera (at right) which is essentially blinded. This clip was taken along the Delaware River and is similar to the New York region you mention.

      httpv://youtu.be/Deeiphnb5rQ

      That said, when working with customers who will use our cameras in areas where fog is common, we’ll suggest de-rating the camera’s published detection distance as needed relative to the specific environment where they’ll be used.

      Thanks again for the comment.

      Eric

  • We are using Sightlogix in the Norwegian Artic climate, and are very happy with the results also in these harsh environments.
    If anyone want video from different weather conditions – send me a notice and I will link you one.

Leave a Reply