The smartphone app that allows the public to present evidence of speeding: will it actually work?
Our story earlier this week about a new smartphone app that uses AI to analyze video and lets users report suspected speeding drivers to the police for a potential app drew a lot of comments, including by Rod King of the 20’s Plenty For Us road safety campaign, whose volunteers are testing the app.
> New smartphone app allowing the public to submit evidence of speeding
Among the concerns raised about the Speedcam Anywhere app was whether police would have the desire, not to mention the resources, to deal with such submissions, as well as whether there were evidentiary issues related to the captured images.
However, responding to road.cc readers in the comments to the original article, King said he believed the technology “will be transformational” and that he was “completely convinced it will make a huge difference”. for road safety.
He said: “I think it’s fair to say the forces don’t expect this. So don’t expect systems to automatically adapt to new technology.
“I’m confident it will be accepted, but we have a very uneven set of constabularies on the app, particularly ‘where the people are’. There will be pioneer forces that see the benefits and others that will just migrate from Gatso [speed cameras].
“The point about uploading the video and the report is that the report saves the analysis of the video. But if it’s disputed, the video can also be analyzed manually. It’s much easier than analyzing most dashcam or headcam submissions.
“It will be transformational. But some forces may take longer to transform than others. I’m convinced it will make a huge difference. »
We will be reaching out to the app developers to get a number of concerns raised by road.cc readers answered, as well as to get thoughts on the matter from road safety professionals.
In the meantime, Plenty For Us 20s director King gives more details on the app and how it works, which will hopefully answer a number of questions raised in the comments. He also pointed out that more information is available on the the organization’s website.
Most speed sensing “devices” use a function within the device to measure the speed of a vehicle. If it’s radar or laser, it’s measuring speed by bouncing a wave off the oncoming vehicle and measuring the doppler effect produced because the vehicle is moving .
A problem with this is that you can rarely use the device in the path of the vehicle and therefore you are not measuring the speed of the vehicle as it is reduced by the cosine from the line of sight to the direction of travel. Because the device independently assesses the speed it needs calibration.
Speedcam Anywhere is different. It’s not the smartphone that measures the speed. The app buffers a video frame and when you press the shutter as a car passes and is centered on the screen, it selects a video clip from the previous second and the next second. This is then uploaded to the cloud along with the GPS position. The server then:
- Uses ANPR to look up vehicle make, model, and year;
- Finds the wheelbase (WB) of the vehicle;
- Uses AI to analyze video and locate wheel centers;
- Find the still image in the video clip where the front wheel passes a point on the road. Takes its timestamp (T1);
- Find the still image in the video clip where the rear wheel passes the same point on the road. Takes its timestamp (T2);
- Uses the simple physical calculation v=s/t i.e. v= WB/(T2-T1);
- Search the map to find the speed limit at this point;
- Creates a two-second video clip overlaid with timestamps;
- Creates an A4 report showing location, vehicle image, vehicle details, location, speed and speed limit along with timestamped still images used;
- Returns a summary to the application.
The app user can then download the report and the video clip to upload to the police dash cam. Here the police can (if they wish) review the video to verify the speed of the vehicle.
Because the wheels always follow the direction of the vehicle, the angle of approach does not matter. You can use the app with oncoming or departing cars as long as you can see the license plate. A line of sight with one edge of the image perpendicular to the road has proven to be the most effective.
The requirement for a home office-approved device only exists for speeding convictions. This is an anomaly that assumes the speed is measured on the device. When it comes to video, precedents have already been set where drivers have been prosecuted for reckless or dangerous driving based on video evidence. This includes checking for speeding tickets.
However, this requires expert analysis and often measurement of road markings or landscape to provide a fixed distance to measure travel time. Speedcam Anywhere eliminates the need for that expert analysis by using AI and the car’s wheelbase as a fixed distance.
The antisocial driving offense in section 59 only requires reasonable grounds to believe that a motor vehicle is being used on any occasion in a manner that contravenes section 3 or section 34 of the Highway Traffic Act 1988 (reckless and reckless driving and prohibition of off-road driving). conduct) and causes or is likely to cause concern, distress or annoyance to members of the public. Therefore, under “reasonable grounds”, a breach of section 59 can be recorded.
Speedcam Anywhere could get type approval for “process” rather than “device”. In this case, the police and members of the public can use it directly for the enforcement of a speeding violation.
This really uses a great combination of modern databases, smart analytics, and AI to provide accurate speed measurement using a smartphone. This is not only a great invention, but also a huge step forward for the application. When driving in the future, any pedestrian you see could be a Speedcam Anywhere pedestrian. Especially in urban and village settings, drivers should beware of blasting in public places between buildings we call streets. After all, 20 Plenty is where the people are.