Advanced Driving Hub

"New app could soon turn every car into a speed camera – and report traffic offences at the touch of a button."

https://www.rac.co.uk/drive/news/driving-tech/new-app-could-soon-turn-every-car-into-a-speed-camera/

About the only upside is the police being swamped by so many reports they can't possibly process them all. And hopefully owners of such devices will be prosecuted when they inevitably speed themselves (find me a driver who /never/ speeds). Afterall, if catching bad driving is so important, why shouldn't the phone/app report the owner for such outrageous behaviour?

"Show me a driver who says they have never exceeded a speed limit and I will show you either a liar or a menace."

I can't remember who it was said that but I think it was someone in officialdom related to driver training?

Anyway, I wonder how the app will get type approval for recording all of these offences. If the local constabulary's bad driver reporting system is anything to go by I'd say this one would probably last... hmmm... maybe a day or so before it is overwhelmed. It will also take an army of people to review every individual piece of footage etc.

It's madness, but I would not bet against TPTB trying it out.

jont- wrote:About the only upside is the police being swamped by so many reports they can't possibly process them all. And hopefully owners of such devices will be prosecuted when they inevitably speed themselves (find me a driver who /never/ speeds). Afterall, if catching bad driving is so important, why shouldn't the phone/app report the owner for such outrageous behaviour?

Interesting to note, I know for sure that both Avon & Somerset, and our force (Gloucestershire) will prosecute drivers who send in dashcam footage to report other road users if it also reveals they've committed an offence themselves.

One of the most common ones is people who become 'Dash Cam Warriors' and try to chase people to catch them or prove an offence, and in the process they are clearly speeding or dwdc themselves..

One of the Bristol examiners is a current serving traffic officer for A&S and she told me that what they do is have police staff do an initial sort of the obvious nonsense ones, and then anything which *may* have an offence on will be reviewed by a traffic officer. They batch them up and then part of their rota will be that traffic officers will occasionally have a few hours where they are allocated to review them.

I'm a special with Gloucestershire and we have an online portal to submit dashcam footage in a similar way. I've sent a message to the inspector who runs the RPU ops to ask what happens to it, will update when I get an answer, but I suspect it'll be much the same as to confirm an offence it has to be reviewed by a warranted officer.

Being a software guy by background, with a strong interest in the technical side of photography too, I’m trying to imagine how such a device might work.

I think it is possible, if you measure the distance between two points and then film a vehicle driving between them, to count the number of video frames and determine speed quite accurately. But that’s not what’s being claimed, is it, there’s no measured distance?

The only thought I have is it might use data from the camera’s autofocus software to guess the distance at some moment, and then to repeat it a few hundred ms later, and calculate speed based on the difference in focus over time. That would be interesting as a toy, but laughably inaccurate, and nobody is going to take it as serious evidence.

My suspicion here is that RAC (among many other publications) have been played for suckers, in advertising an App that’s based on snake-oil, hoping to make a fast buck for the developers. Happy to be proven wrong.

I doubt it is any different to any current in car camera
It simply records standard video on a loop and then the rest of the app is the ability to hit a button to say keep the last 30 seconds - at which points it saves that and restarts the loop - the saved section later being available for upload to a portal.
speed if calculated would be time for car to go between known points on the road as now...

akirk wrote:speed if calculated would be time for car to go between known points on the road as now...

If done manually, video frames (25 or 30 fps) and count the line markings.

Speed of the camera car could be by GPS, but typically not reliable.

I was told that by an accident investigator who had done research on dashcams as evidence

From my own messing about, there can be considerable lag.

My understanding is that whilst actual location measured by GPS can be a few metres adrift, speed derived from it is typically very accurate indeed - albeit with a computational lag of a few seconds. I believe it simply measures the doppler shift in the signals from the satellites, hence location accuracy is irrelevant when measuring speed.

But that would only yield speed of the “observer”. This app claims to deduce speed of an independent target vehicle, as well as the observer. Yes, that could be measured by counting the video frames between two known points. And if the offence was serious enough to merit the effort, an investigator could manually measure the distance based upon features such as, say, a pillar box an a tree. But GPS coordinates are nowhere near accurate enough to be used as a substitute for manual measurements.

I remain sceptical of this app. That said, I’ve published a few iphone Apps myself, just for fun. I wish I’d though of this idea myself as I reckon it might be a good earner for the developers.

Can't remember details, but apparently there was a crashed car that, according to its dashcam, 'moved' across a road.

I was setting up for trials a few years back, logging the location of equipment. Using 3 GPS units, together in an A4 plastic wallet. One was 11 metres away from the other two.

Leaving aside the improbable suggestion that this App might measure speed of other traffic, I see the Daily Mail is covering it too.

https://www.dailymail.co.uk/news/articl ... EDING.html

The Mail article helpfully explains that

Those using the app while driving will not face a risk of using a mobile phone behind the wheel, as they will simply be required to touch th home screen to record a breach.

Maybe not a specific offence of using a mobile phone, but surely still distracted driving?

Personally if I’ve just been cut up by some hooligan then I want to spend the following few seconds getting my wits together and assessing effects upon other traffic, rather than venting my anger by stabbing at my phone. The article is illustrated by a picture of a phone in a suction mount slap-bang in the middle of the windscreen which, I suspect would indeed be typical placement for many people anxious to use such a feature.

I don’t have a dashcam, though I’ve been tempted. A non-negotiable requirement for me would be that it could be positioned in such a way as to not impede my own vision in the slightest and in fairness, most purpose-built dashcams seem to be designed with that in mind. Phones, however, are a fundamentally different shape from dashcams…