Homemade Speed Trap Made By Former UVA CS Professor (cvilletomorrow.org)

An anonymous reader writes: Irritated by speeders in his neighborhood and frustrated with the City of Charlottesville's inability or unwillingness to enforce the speed limit, a former professor in the Computer Science department of the University of Virginia created a program in openCV to track vehicle speed on his residential neighborhood street: "You'll find that almost 85 percent of the cars going by are violators [of the neighborhood's 25mph limit]". This includes a city bus doing 34mph.

Comments

legalism is a crap philosophy.

[Anonymous Coward]

If everybody is speeding, maybe the speed limit is too low.

Re:legalism is a crap philosophy.

[codeButcher]

If everybody is speeding, maybe the speed limit is too low.

Probably not, as The Fine Article states near the top: "... installed a camera on his roof and began writing speed-monitoring software after a 12-year-old pedestrian was injured by a car last October."

Re:legalism is a crap philosophy.

[MrL0G1C]

Perhaps the kid was jumping in front of the car and would have been injured anyway.

Two things about this, one, slower vehicles are much easier to avoid for careless kids and two, speed kills, every extra ten miles an hour exponentially increases the likelihood of the pedestrian being killed when hit.

Re:legalism is a crap philosophy.

[Aereus]

That's all fine and good until you throw snow and ice into the mix, then all those objects become wrapped around cars and cause accidents from the excessive braking/swerving required to navigate them during inclement weather. I've lost count of how many signs and poles I've seen bent over clear to the ground after storms, or cars losing control in S-curves from the "scenic/safer" road design.

Re:legalism is a crap philosophy.

[N1AK]

Two things about this, one, slower vehicles are much easier to avoid for careless kids and two, speed kills, every extra ten miles an hour exponentially increases the likelihood of the pedestrian being killed when hit.

So set speed limits at 10mph, or 5mph, or ban cars entirely if decreasing fatalities is always a justification for decreasing a speed limit, because if it isn't then you need a more credible case.

In the UK the normal speed limit in a residential area is 30mph. 20mph limits in the vicinity of schools are becoming more common. In general UK speed limits are quite relaxed, and especially on non-urban roads policing of moderate speeding is very limited; It is not at all unusual to find traffic averaging 80+mph on UK motorways (interstates) which have a 70mph limit, and you could comfortably do 90mph if traffic is flowing with no real risk of a ticket.

All of this should make the UK a very dangerous place for pedestrians if speed limits alone were a primary driver of road fatalities, but they aren't. The UK averages 3.6 fatalities per billion kilometres driven. The US average (where limits are on average lower) is 7.1, which is effectively double. It seems much more likely that issues like car quality, driver certification, road design, car design etc are far more influential.

Re:legalism is a crap philosophy.

[nierd]

Speed limits should be set based on the first only. The entire objective is reduce the probability of and accident to an acceptable level, not reduce it's severity.

Yeah - my seatbelts, crumple zone, side impact airbags, front impact airbags, passenger airbags, re-enforced pillars, side impact beams, automatic roll shut off, bumpers, safety glass, etc. all are very helpful at reducing the probability of an accident. That's a seriously wrong statement.

How to improve Slashdot

[enriquevagu]

Add a link to the summary.

Re:25 mph?

[Viol8]

But the Think Of The Children campaigners will tell you that injuries to kids are X % less at 25 mph than 35 (or whatever). Which of course is a Reductio Ad Absurdum argument since you can then argue than 15mph causes less deaths than 25 etc and eventually get to the point where you end up with a 5mph limit and a man walking in front of the vehicle with a red flag and a whistle to warn people ahead. These people refuse to countenance the fact that there must be a trade off between road safety and society - which like it or not depends on motor vehicles - being able to function.

Re:therefore the speed limit is invalid

[silas_moeckel]

If that 85% are not crashing into things seems rather safe to say they are correct. Would be hard to find any road where nearly 1 in 6 cars has an accident each time they drive through.

Re:Dear black and whiter

[stdarg]

You're not considering that the 4 lane avenue is wider and probably busier, and thus scarier to kids. A sleepy residential street on the other hand is the kind that a kid might cross without really thinking about because 9 times out of 10 (or more) a car isn't coming.

Therefore you can't assume the streets have to be treated the same just because of proximity.

Re:This speed limit is reckless

[jratcliffe]

It is very well established in the civil engineering literature how to set the speed limit for vehicle usersafety.

That 85% rule of thumb might work well for an Interstate highway, but it's a terrible metric for a local street, where the priority should be the safety of pedestrians.

Re:Dear black and whiter

[ultranova]

The residential street is 100m from a 35mph four lane avenue. No fence, no limits. What stops the same careless kid to step onto the avenue to the left, instead of the street in front?

The fact that it's a four lane avenue rather than a residential street. Kids aren't generally stupid, just inexperienced about entitled assholes who think they're above the law.

Following the "think of the careless children!" reasoning, we'd either have to:

No, we don't "have to". We also have the option of simply enforcing existing speed limits with a special emphasis on residential streets and other low-limit areas. Just make the fine proportional to ((speed - limit) / limit) and unwillingness to enforce shouldn't be an issue anymore.

Re:OT Re:legalism is a crap philosophy.

[omnichad]

the term "exponentially" is being abused.

No. Unless you're talking about x^1, exponentially is not being abused. What else are you going to call it? Linear? What's your threshold for exponential? x^3?