Innovative Ways to Count Pedestrians and Bicyclists

By James Bonanno

Traffic volume is one of the most important pieces of information that transportation professionals consider when building or updating road infrastructure. Volumes are systematically collected using automated technology at locations statewide to provide reliable average daily counts of vehicles along corridors and in intersections.

Traffic volumes are also important in the planning and design of pedestrian and bicyclist infrastructure; however, such data is not systematically collected the way that it is for vehicles. Pedestrian and bicycle traffic is usually counted on a case by case basis that gives a snapshot of what traffic could be like in an area. If the weather is bad on the days that the counting is planned to take place, counts may be significantly lower than on an average day with nice weather. Vehicle traffic data on the other hand, is collected frequently throughout the year and compounded into an average daily traffic volume. Recording this data over a longer period of time results in a better representation of the actual traffic volume the corridor may experience in an average day.

Vehicle counts are collected frequently thanks to the automated technologies used to collect the data. One method is laying a simple rubber tube across the road, attached to a counter. Every time a car drives over it, a count is registered. These counts give planners and engineers a good representation of how the corridor is being used and helps them in their decisions to improve the corridor. Pedestrian and bicycle traffic however, has traditionally been recorded manually. This process is time consuming, monotonous, expensive, and can yield unreliable results. In busy urban areas with high pedestrian and bicycle traffic, it can be very difficult for an employee to keep up with the amount of people and bicycles passing by. Video cameras can be used, but an employee must still go through the videos and manually count each pedestrian and bicyclist, as the automated video-counting technology is not always reliable.

The New Jersey Department of Transportation has acknowledged this shortcoming in pedestrian and bicycle traffic data and has begun researching new technologies to implement that will provide accurate pedestrian and bicycle traffic data. Infrared Beam Counters, Passive Infrared Counters, Piezoelectric Pads, Laser Scanners, and Computer Vision are all emerging technologies in pedestrian counting that NJDOT has considered. Each is a viable option for counting pedestrians and bicyclists but each also has its own drawbacks.

    1. Infrared Beams:Consists of a two small boxes that are connected with a beam of infrared light. When the beam of light is broken by a pedestrian or bicyclist a count is added. These two boxes must be deployed horizontally across an area such as a sidewalk. It can provide an accurate count of individuals passing by but has trouble deciphering the number of people in groups (i.e., walking side by side), and differentiating between pedestrians, bicyclists, and animals.
Melbourne’s pedestrian counting sensors.
    1. Infrared Counters: These devices detect heat signatures that pass, and count each individual signature that passes a certain threshold. This threshold of heat allows the sensor to differentiate between a human and an animal, but cannot tell a pedestrian from a bicyclist. These devices are deployed horizontally, resulting in the same problem when it comes to counting groups of pedestrians. Groups of pedestrians may create heat signatures so close together that the device count the group as a single pedestrian.
    2. Piezoelectric Pads: These are pads or mats that lay in the ground and count pressure signatures that pass over them. If a number of these mats are placed side by side, they can each count a different pedestrian and add them together to provide an accurate count. Piezoelectric Pads must be placed two to four inches into the ground. This is not a problem when pedestrians are walking on soft dirt or grass, but the pads can be costly to install in concrete. In addition, they cannot differentiate between pedestrians and bicyclists.
ECO Multi-Counter on Route 52 Bridge
  1. Laser Scanners: Deployed either overhead or horizontally, laser scanners emit infrared light and catch the reflected pulses to create an image that is then counted. They can differentiate between pedestrians and bicyclists, and when deployed overhead can accurately count pedestrians in groups. However, these scanners are heavily affected by weather conditions. Rain, fog, and snow can reduce the accuracy of the counts.
  2. Computer Vision: Video provided by cameras is analyzed by computers to count pedestrians and bicyclists in an area, but the count accuracy is drastically affected by the available light and weather, and many times require human checks.

In addition, there are tools available to count bicyclists, such as the same rubber tube used to count cars, or induction loops installed in the pavement. These devices do a good job at counting bicyclists but cannot count pedestrians at all.

These are some of the current technologies being considered by the New Jersey Department of Transportation to improve the accuracy of pedestrian and bicycle counts. Which technology is installed will ultimately depend on multiple factors having to do with each devices count accuracy, cost, availability, and deployment method.

A good example of pedestrian counting can be taken from Melbourne, Australia. The city uses overhead sensors to count pedestrians and bicyclists and decipher which direction they are moving. Twenty-eight sensors are at work counting twenty-four hours a day, seven days a week and the information is all publicly available on the internet. They can do this because the sensors track movements instead of images resulting in anonymous travel data.

In New Jersey, the new Route 52 Bridge to Ocean City was built with technology to count pedestrians and bicyclists. The multi-use path on the side of the bridge uses a combination of infrared beam and induction loop to count and differentiate between pedestrians and bicyclists. You can read more about it here.

Having more systems such as this deployed in New Jersey could provide excellent pedestrian and bicycle travel information and help planners and engineers make future pedestrian and bicycle infrastructure decisions.


Sources:

http://www.pedestrian.melbourne.vic.gov.au/
http://cait.rutgers.edu/research-reports/automatic-pedestrian-counter

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s