More than 270 000 pedestrians lose their lives on the world’s roads each year accounting for 22% of the total 1.24 million road traffic deaths. (WHO, 2013) Jaywalking is one of the problems. However, jaywalkers are not always the ones to blame. There are cities that are heavily centered around road traffic and which do little to provide pedestrians with means for safe and frequent road crossing. Some cities are aware of this problem and want to come up with solutions. FLOW is here to help. Using smart video analysis, we can help you detect where the jaywalkers cross and tell their numbers. Keep reading and we will show you how exactly this can be done and how this can help you make better decisions that improve pedestrian safety.

If you are new to FLOW - read this article or watch our webinar to get familiar with the basics. Links to more FLOW specific guides can be found below.

Motion filters in FLOW

Property filters in FLOW

Spatial filters in FLOW

Set operations in FLOW

Working with time in FLOW

Using widgets to view traffic data in FLOW Insights

If you do not have the FLOW Insights demo kit yet - download it here for free!

In order to only detect the pedestrians in our data, we need to drag and drop a category filter to the canvas. We double click it to open its menu and we tick the pedestrians' box and untick the car box which is ticked by default.

If we only want to detect jaywalkers' total number, without knowing which direction they came from, we can do this with only one gate. If you want to do both directions skip to the advanced counting solution part. Then we create a gate in the footage with the create zones/gate tool in the top left corner by drawing a line.

We draw this line in the middle of the road. We name this gate “both way jaywalking detection” Then we click on the gate and hold to drag it onto the canvas and connect it below the category filter. This method will let us get jaywalker counts for both directions. When using a single gate we could also be counting some people getting out of cars. If we want to count only crossing jaywalkers we will need to use 2 gates which will be the case in the following part. Delete this gate if you would like to try the advanced counting solution next. If not, skip to the widgets part. To do this either push the delete button on your keyboard or double click the gate and click “Delete gate”.

If we want to detect jaywalkers crossing from each direction separately we need 2 gates and 2 movement directions. We draw gates on each side of the road and name them “Jaywalker detection left” and “Jaywalker detection right”. Next we draw 2 movement direction lines. One from left gate to right gate and one from right gate to left gate. We name them “Crossing from left” and “Crossing from right” respectively. Be careful to follow the direction in which the movement is drawn.

Then we drag these movements onto the canvas to see their values. Double-check the direction, which is under the Movement filter in the canvas, so it correctly represents the direction. Do not forget to connect them under category to count only pedestrians.

If we want to count jaywalkers in the different sections of the crossing we can create separate gates for them. We can set up detection for specific jaywalking paths. This can be done in the same way as described above with the only difference of drawing smaller gates that cover just portions of the crossing.

By looking at the trajectories we can identify where the pedestrians are crossing. This allows us to decide on the best placement for the zebra crossing.

To better see the individual values, you can drag the value widget on each of the parameters in the canvas that you want to monitor. After you do so, go to the dashboard to see all of the widgets you have added.

You have learned how to detect jaywalkers from both directions at once and from each direction separately. These insights can help you to decide things like whether zebra crossing is needed and where. It can also help you find out whether jaywalkers are present at all and if so if there are enough of them to consider changes in the design of the roads and surrounding crossings. Altering the traffic light cycles on crossings may also help reduce jaywalking or perhaps you might decide that the overall number of pedestrian traffic is so big that construction of an overpass or underpass would be the best solution to improve the traffic flow and safety. We hope that you have enjoyed this guide. To browse more articles about FLOW click here.

Check out our highway traffic camera guide here

Links

• Learn about the use-cases for pedestrians and bicycles for FLOW Insights here.

• Do you have any questions? Contact us!

• Follow us on Linkedin.

• Learn more about FLOW from guides or on our web.

• Visit our homepage to view products and news.

Thank you for your interest! Have a great day!

Let the traffic flow - with FLOW!

Reference: (WHO, 2013) - https://www.who.int/mediacentre/news/notes/2013/make_walking_safe_20130502/en