Slashdot Mirror
Speed Tickets Challenged Based On Timestamped Photos
SEWilco writes "A businessman has challenged automated tickets of his vehicles by calculating the vehicle speed based upon the tickets, which include timestamps of two photos." Maybe more word problems should be on the police academy curriculum.
Mr. Foreman’s tickets were all issued in Forest Heights, a town of about 2,600 where officials expected $2.9 million in ticket revenue this fiscal year, about half the town’s $5.8 million budget.
Couldn't get people to pay taxes for that new community pool there? Sheesh.
which include timestamps of two photos.
The obvious response? They will start sending ONE timestamped photo.
I got a ticket from one of those things 2 weeks ago; when it flashed, I looked down. I was doing 48. I've checked my speedometer using a GPS, and it's accurage. They aren't supposed to take a picture until 10 miles over the limit (the limit there is 40, so it shouldn't have taken a picture until 50). The ticket that came in the mail said I was doing 52.
I talked to a lawyer, and was told to just pay the bill, less trouble and less expensive in the long run.. so, that was $218.
The real kicker on the ticket was that each offense must be reviewed by a real cop with a badge number. The cop's name? Officer Dollar.
Bastards.
Never trust an atom. They make up everything.
And at this scale, it's got absolutely nothing to do with Heisenberg.
Those who fail to understand communication protocols, are doomed to repeat them over port 80.
There have been studies that show a huge increase in collision, especially rear-end collisions at intersection cameras.
There have been many scandals with towns setting their yellow lights to have durations significantly below the correct, and often legally required minimum times.
There is a huge trend for these to be cash cows for local governments by means of fraud. And they wonder why people hate them.
Looks like this guy has identified a town where the cameras are 'miscalibrated' and are raking in tons of dough from everyone that isn't as smart as this guy.
Did you read the article?
If the time difference is 0.363 seconds, there's not much time for acceleration. Assuming braking at 1 g (approximately the maximum a car can do) that's a difference of about 8 mph -- which is substantial, but the pictures show the brake lights as generally being off, suggesting a much lower rate of acceleration or deceleration.
Also, it's clever to invoke Heisenberg any time we're talking about velocity and position, but I think these objects are large enough to assume that the uncertainty is relatively small :)
The real travesty here is that the judge let other tickets issued by the same devices stand after it was demonstrated to him that they are not reliable. If there is reason to believe that the device was wrong in one case, there is reason to believe that it was wrong in every case.
The truth is that all men having power ought to be mistrusted. James Madison
"In Prince George’s County, cameras are operated entirely by municipalities, which can set them up within half-mile school zones. The devices are installed by vendors that typically receive about 40 percent of the payout on each ticket, with the rest going to local, county and state government."
How could anyone have thought that this was a good idea? If the only thing the private corps are doing is the installation, why are they getting 40% of all future proceeds? If the private corps are doing the on-going process of operating and maintaining the cameras, then you just incentivized them to do whatever causes more tickets to be mailed out.
My guess is that it's the later, and the local municipalities are more than happy to incentivize the private corps to break the law, since they're getting 60% cuts. Then, when scandals like this one break out, they wash their hands of the matter and say we didn't know what was happening, it was that corrupt private contractor.
From the article:
Optotraffic representatives said the photos are not intended to capture the actual act of speeding, and are taken nearly 50 feet down the road from sensors as a way to prove the vehicle was on the road.
How does proving that a car was on the road prove that it was speeding?
A lawyer with some spare cash can rent an instrumented "bait car" with certified-instruments that will be admissible in court and prove once and for all that the cameras lie, then sue the city on behalf of all who were convicted or who plead guilty under what amounts to duress.
The city can then sue the vendor for the 40% cut it paid back.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
I'm at work, so I can't look it up, but do a google/youtube search on "atlanta speed limit 55" or something like that.
TL;DR: Some college kids decided to go the speed limit on Atlanta's 295 loop, which is posted at 55mph, but traffic travels around 70+ mph. They got five cars and blocked all lanes, and went 55 mph. The video editing is atrocious, but the point is very good.
The government intentionally posts low speed limits so everyone is guilty. Once everyone is guilty, they are free to pull over anyone, at any time, for any reason, and cite "speeding" as the reason.
No they saying he was able to DECELERATE 15mph in .363 seconds. RTFA.
Sig Battery depleted. Reverting to safe mode.
Or, decellerating from 50 to 20 in .363 seconds, technically less as the article implies the brake lights are not lit in hte photo. Perhaps significantly less as standard incandescent lights as use in the brake light fixtures of typical trucks take a (relatively) significant amount of time to fully illuminate, and a (relatively) significant amount of time for the fillament to cool and go completely dim.
The photos are taken 50 feet from the sensor according to the company.
But he is using time stamps which are placed in the image as they are being written to disk, (probably microsd card) NOT as they are being taken.
Pictures taken are held in memory until they are processed (converted from raw to jpeg). At the time they are processed the timestamp in inserted into the image.
It took .363 seconds to process, timestamp, and write out the first image. That's is ALL that time stamp measures.
So upon a technical review, this guy should have lost this case.
Sig Battery depleted. Reverting to safe mode.
Further, assuming a 3/4 second reaction time, it takes 55 feet at 50mph for a driver to even get a foot on the brake pedal, which is 5 feet more than the manufacturer's expert claimed the average distance from sensor to photograph was. so, assuming the .363 second difference occurred after the driver realized the camera had taken his picture, he would actually not have the full .363 seconds to decelerate to meet the mean time theorem criteria for the photograph.
also if you had read the article you would have also noticed that the pictures are taken roughly 50ft after the car passes the speed sensors
according to google... 50 mph = 73.3333333 feet per second
so that gives the vehicle about an additional .68 seconds to decelerate before the first picture is even taken.
so assuming your calculations above are right that is roughly 15 mph of braking the car could do before the first photo is even taken....50mph-15mph=35 mph...which could put him at the speed limit before the first photo is taken.
this was just some quick estimation, but i think the calculations work out.
No, they are saying he was able to decelerate 15 MPH in the ~50 foot distance between where his vehicle was when it was supposedly clocked, and where it was when its photo was snapped.You RTFA.
Optotraffic representatives said the photos are not intended to capture the actual act of speeding, and are taken nearly 50 feet down the road from sensors as a way to prove the vehicle was on the road. ... “Their speed is not measured by the photos. The speed is measured before the photos are taken.”
Of course, that does bring up the question of why they need 2 photos if they aren't using them to determine the vehicle's speed.
The photos are clearly intended to prove that the vehicle was at that place at that time. If the vehicle was not at that exact place at that exact time, they are inaccurate and should be inadmissible in court.
I don't think that's right. A time stamp on disk might be placed in the image as it gets written out, but that's only accurate with 1 second granularity anyway, making those time stamps useless. This is talking about a time stamp that contains much more precise time stamping information, likely burned into the (possibly non-digital) image by physical hardware in the camera, which almost certainly means that it is generated at the same time the picture is generated.
If it is being burned into the image after the fact, then the camera vendor is being dumb, particularly since the whole purpose of those photos is to prove that an infraction really occurred, and burning in the time stamps after the photo is taken is basically tampering with evidence.
Check out my sci-fi/humor trilogy at PatriotsBooks.
Going from 50 mph to 20 mph in 0.363 seconds means he would have had to been decelerating at about 3.7Gs.
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
So the rich are free to speed as much as they want, only because they can afford to do so?
Any sect, cult, or religion will legislate its creed into law if it acquires the political power to do so.
As was noted in the article, however (and in rebuttal to the vendor who argued your point), the defendant noted that none of the photos showed his vehicles (company vehicles) with their brake lights on.
While that doesn't mean they weren't speeding prior to the intersection, the calculations and absence of break lights raise reasonable doubt.
If the camera is not taking photos of the vehicles while the violation is being committed, what proof is there that the vehicle was actually breaking the law? They could snap two photos of any vehicle crossing the intersection and claim it was speeding at the sensor location. If they want to use the photos as proof, take a photo at the sensor location and then again at the start of the intersection. That known distance would allow for a more accurate representation of the alleged speed of the drivers.
I use irony whenever I can, but my shirts are still wrinkled...
Too late. We're officially confiscating your Math license.
I'm guessing it snaps a few pictures to make sure that it gets a clear shot of the plates and possibly the driver.
"In America, first you get the sugar, then you get the power, then you get the women..." -H. Simpson
In New Orleans, they did JUST THAT!!
They put up what I thought were stoplight cameras...no problem I stop.
I didn't know they were also speed cameras....and there is a lag time between getting popped, and you receiving your ticket in the mail...a fucking MONTH!!
This was on my route to work..I got 7 of them...lucky it was ONLY 7.
But on the back of each ticket...it says clearly that "this is not a moving violation and will not go on your record".
This is NOTHING but revenue generation.
The nice folks in Jefferson Parish, next to us..voted these damned things out...with they people in Orleans Parish would do the same.
Light travels faster than sound. This is why some people appear bright until you hear them speak.........
So they're tampering with the evidence by putting false timestamps on the photos.
Oh crap, I need a license to perform Math?
You mean I've been estimating path loss and link margins illegally all this time?
Since it's 2011 and I started in 1988, that means it's been... D'oh!
I can see the fnords!
OCCAM suggests that the photo time stamps measure the interval betweens processing and storing of the photos and nothing at all to do with when the photos are actually snapped.
Speed is measured by radar, not photo evidence.
Sig Battery depleted. Reverting to safe mode.
Not necessarily .. sunlight of a passing car might cause glare. They might have one of those illegal covers over their plate to "hide" their license plate. It might be raining and you get a waft of rain on the first picture due to rooster tail or puddle splash. If you are saying there are two pictures, at the same time? Yes I agree. However most cameras that I have seen have two separate pictures at a minor time distance apart for just those reasons.
good use of the iphones new location tracker.
It is nowhere near accurate enough for that. The best it can do is say that you were probably on that road at approximately that time, or maybe that you were nowhere near the camera site at the time.
Just to add some points of comparison:
Normal hard braking is about 0.4 Gs.
Skilled hard braking is around 0.7 Gs.
Around 1 G seems to be the limit for skilled braking with performance tires and a great road surface.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
which almost certainly means that it is generated at the same time the picture is generated
No. That's exactly my point. The time stamps are generated AFTER the picture is taken.
In order for the time stamps to measure EXACT time the picture is taken you need a realtime clock running in the focal plane. NASA does this.
Off the shelf CCDs do not have this.
The time stamp is inserted at processing time. Its not in the raw image.
Lets imagine a system where an interrupt is triggered when the image capture is complete. If, at that time, the timestamp is inserted into the raw image, then it will be accurate to within one interrupt response time, which is going to be accurate enough to support the charge of speeding.
If the timestamp is inserted into the image after some arbitrarily long interval during which jpeg encoding is done, files are saved to flash memory, etc. etc. then you are doing it wrong and your cases deserve to be thrown out.
Then the company could have easily said that. But they didn't, even though they sent a guy down to the court to defend their device.