Security systems integration on highways: are you up for the challenge?

Security systems integration on highways: are you up for the challenge?
As security solutions rapidly move beyond their conventional functions, highway traffic management is becoming a potential market for systems integrators (SI). Spread over several thousand kilometers in any given country, ensuring safe travel on highways is increasingly seen as a priority.

But the unique conditions on highways necessitate special attention and specific requirements. For SIs working in the vertical or looking to enter it, such requirements could be challenging. Here we talk to industry players to understand how SIs can overcome these difficulties and provide effective security solutions on highways.

How to optimize security solutions for low light?


Optimizing the solutions for low light is among the first priorities in this vertical. Fortunately, several modern cameras are able to provide usable images in such conditions, but installers will have to consider various parameters while installing them. 
security on highways
High speed and low light are two of the major
challenges to providing security on highways.


“For accurate analytics, camera mounting height and angle towards the road need to be chosen with the exact VCA application in mind,” said Constant Rutten, Marketing Application Design for Video Systems at Bosch Security Systems. “For example, for traffic monitoring the camera projection should be selected to minimize occlusion of cars, prevent direct sunlight and glare of reflecting surfaces. The glass window in the protective camera housing should be designed to minimize reflections in the optical path.”

Todd Brodrick, Director, Southwest-US at Pelco expanded the point further. 

“Many of the challenges can be solved with correct application and set end-user expectations.” Brodrick said. “Application challenges can be resolved with proper placement of the high-resolution camera, with the proper mounting mechanisms. For environmental challenges though, it is more difficult to bring a one-size-fits-all resolution as different locations have different issues.” 

Rutten also indicated that in extremely low light situations, cameras with highest light sensitivity should be chosen and may be combined with optional IR illumination for better results.

Adler Wu, Product Marketing Manager at Hikvision Digital Technology agreed to this, suggesting that IR products with high performance and precise algorithm can capture vehicle information, such as license plate, vehicle number, vehicle color and even vehicle type and much more. 

How to overcome challenges in license plate recognition?


An important part of any traffic surveillance solution is license plate recognition (LPR). According to Wu, the key aspects for efficient LPR include taking into account the differences in the license-plate types, traffic rules, and road infrastructure. 

“In terms of different vehicle types, with different symbol and text, we optimize the algorithm to improve accuracy,” Wu said adding that LPR could still be a challenge if vehicle license plate is damaged or covered. 
security in tunnels
Tunnels are especially challenging to provide
security solutions for. 


Jean-Pierre Picard, Product Marketing Manager at Genetec acknowledged such concerns, but added that countering them is not too difficult.  

“The most common challenges have to do with vehicles whose license plate is not visible - either because it is hidden by an object, such as a trailer hitch or a bike rack, or because there is no license plate, such as new vehicles,” Picard said. “But because the percentage of vehicles with blocked or missing license plates is fairly stable over time, it is still possible to produce an accurate vehicle count estimation, using LPR as the baseline data.”

Speaking along similar lines, Ashwin Amarapur, Director of the India-based VCA solutions provider, AllGoVision, said that customizing the VCA solution to identify license plate layout in different countries is critical. The solution has to be able to recognize and understand the differences in layout. 

“There is always a trade-off between having a high-contrast setting that allows capturing a wide image with all the surroundings, and selecting the region of interest, with a setting that is focused to understand what’s happening in that area,” Amarapur said.

How to optimize bandwidth consumption?

 
Modern streaming technology has come a long way since its initial days, but as image quality and application-requirements increase, there is a constant need to optimize bandwidth consumption. At present, H.264 is the most common standard on the market, and although there is much anticipation on the next version, H.265, the latter’s application- and device-support is still limited. 
bandwidth consumption video surveillance
Some experts are of the opinion that bandwith
consumption shouldn't be a cause of worry.

According to Rutten, bandwidth consumption can be controlled in a number of ways. Smart cameras dynamically adjust camera settings to ensure most efficient bandwidth consumption and also allow compression parameters to be set for different parts of the visuals, which are known as encoder regions. This gives SIs the ability to adapt the encoder compression ratio for various regions of an image. An unimportant region can be set to use more compression and thus reduce bitrates, while important regions can be assigned a lower compression ratio to show more details. 

Although this is indeed a potential solution, Vermeulen suggests the use of analytics at the edge as an alternative to do away with bandwidth-consumption concerns.  

“For proper video analytics performance, one needs a high quality image and sufficient processing power,” Vermeulen said. “By performing the video analytics at the edge, near or in the camera, video transfer can be limited to human eye viewing or even be discarded at all until a significant incident occurs.”

On the other hand, as the available bandwidth rates continue to increase, some solution providers downplay concerns on bandwidth consumption. 

“There is no reason why you can’t open up the bandwidth,” said Dr. Alan Hayes, Founder and MD of AMG Systems. “10GB is readily available and 4GB is within reach. Shaving off a few GB at the edge becomes irrelevant as network bandwidth is generally increasing. The more compression you put in, the more information you throw away, which ultimately hampers the VCA in the control room to a certain extend.”

What’s the right frame rate for highways? 


Solution providers are generally of the opinion that a framerate of 30 fps is sufficient in most cases, but Rutten opined that 60 fps combined with high light sensitivity and dynamic range is needed for free-flow applications at high speed to provide consistent results in widely varying lighting conditions. 

“The image quality of each frame will be more or less the same, as shutter speed etc., are determined by other factors like the speed of the objects and lighting conditions,” Rutten said. “When applying VCA or LPR on the captured images, ask your supplier if the analytics software can make use of the additional frames captured at 60 fps.”

When to use wireless technology?


According to Jeremy Koh, VP of Worldwide Sales at Firetide, a division of UNICOM Global, there are two scenarios where wireless technology comes in handy during traffic monitoring installations. The first is where there is no fiber connection, while the second is when you have to fix a fiber-cutoff issue. 

Speaking specifically on scenarios where there are no fiber connections at all, and where the distance that need to be covered could be long, Koh said that a multi-hop wireless network may be the preferred option. But here, the challenge would be to create an ideal network for real time monitoring, in which the data-rate doesn’t drop from hop to hop. 
wireless on highways
Wireless is used when there is no
fiber connection or when you have
to fix a fiber-cutoff issue. 


“You would need wireless technology that has very low latency,” said Koh. “If possible with a latency number that is linear, for example, for ten hops, there should be a latency number of 10 milliseconds, while 20 hops should have 20 milliseconds.”

Speaking on the different options available, Hayes said three kinds of wireless technologies could be used for traffic monitoring - line of sight (LoS), cellular and non-line of sight.

While selecting a wireless solution for any situation, the major factors to consider are bandwidth and distance capability, resilience of the technology, and the environment. In Hayes’ opinion, LoS is better than the rest due to their higher bandwidth availability. Non-LoS solutions are also more expensive, while cellular networks can be intermittent and irregular.

The downside of LoS solutions is that they are limited by the “line of sight”, which could cause an issue when it comes to connections over long distances. But Hayes feels this shouldn’t be a concern in this vertical, where the role of wireless technology is often limited to short stretches. Usually wireless systems are used to cover small gaps in cabled network, especially on undivided highways (carriageway) where data needs to be transmitted from one side to another. 

Is there a "best highway solution"?


Highways are a potential segment for SIs, as traffic management is increasingly becoming an important aspect of security. But the technological challenges presented here could create considerable difficulties for those seeking to work in the vertical.

To make things complicated, requirements of a project could significantly vary depending on the location, in effect doing away with the concept of a single best solution. 

The key then, is to understand the individual conditions and concerns in each project and tailor a solution that could make it unique. There is also a need to remain up-to-date with the technological developments that could potentially improve the quality or solve a current issue. In short, a proactive approach to solving problems is ideal for providing solutions in this vertical. 


Share to:
Comments ( 0 )