2006 June 10 Saturday
Texas Governor Proposes Border Cameras On Internet
Governor Rick Perry of Texas expects to have cameras up on the Texas border with Mexico within 30 days and these camera feeds on the internet.
June 6, 2006 — The Texas government is installing hundreds of cameras along the Mexican border and enlisting Americans who use the Internet to monitor illegal activity.
Within 30 days, the Texas Department of Homeland Security will enable citizens and law enforcement officials to watch alleged crimes through the Internet as they occur, using surveillance cameras along the 1,200-mile border with Mexico. Texas officials expect the cameras to capture images of drug trafficking, trespassing, theft, rape and kidnapping, all common to border areas.
If anyone has a good source for exactly how many cameras they are installing with the distance between cameras please post in the comments. I'd like to get an idea of how much it would cost to install cameras along the entire border.
I can imagine a software system where cameras with few people watching them can be identified so that volunteers can know which cameras to watch. Also, motion detectors could alert to the need for humans to watch particular cameras. Though wind blowing bushes and trees as well as animals would generate a lot of alerts.
The cameras will have night vision capacity and activity can be reported at a toll free number.
Texans, as well as those in other states, will be able to watch real-time video streams on the Internet to monitor the borders. The cameras will run 24 hours a day and will have night-vision capabilities.
If these border watchers see something suspicious taking place, they can call an 800 number that will be routed to the appropriate law-enforcement agency.
Unlike his Open Borders predecessor as governor Perry is taking a hard line against illegal immigration.
SAN ANTONIO – Gov. Rick Perry joined his fellow Texas Republicans in railing against illegal immigration Friday, telling the GOP faithful the Bush administration has failed to control a "porous and unsecured border."
"There is no homeland security without border security," Mr. Perry told thousands of delegates to the party's state convention.
The delegates cheered, but the party was at the same time striking a defiant tone against the policies of its leadership in Washington and, in some cases, Mr. Perry himself.
In addition, the delegates object to a state law allowing in-state tuition for illegal immigrants, a bill that Mr. Perry signed into law. Republicans also want federal funds withdrawn from colleges that provide such tuition discounts.
Perry might be in the process of shifting toward a more restrictionist position due to popular anger over illegal immigration.
The Border Patrol chief spoke Spanish as he answered questions about the cameras.
Border Patrol Chief David Aguilar, talking to reporters Wednesday in Mission, suggested the state camera plan was devised separately from the federal camera network.
“It’s important that we take the opportunity to align our forces,” Aguilar said in Spanish. “Regarding the proposal by Governor Perry, we are looking forward to the opportunity to sit down and discuss it with him to ensure that whatever is done will be aligned with the efforts of the Border Patrol.”
Luis Figueroa, an attorney with the Mexican American Legal Defense and Educational Fund, warned that the cameras could lead to racial profiling and vigilanteism.
“This leaves the door open to anyone who has a vindictive state of mind or a racial motive,” Figueroa said. “Anyone down there could easily be mistaken and falsely accused of something they didn’t do.”
MALDEF wants Americans to pay for even more Hispanic immigrants. Me, I don't want to take on the "white man's burden" "To seek another's profit, And work another's gain." and I deeply resent the desire of our elites to make me do so.
A Border Patrol union president does not think the Border Patrol has enough agents to respond to all the calls that will come in.
And T.J. Bonner, president of the union that represents nearly all Border Patrol agents, said the plan could further strain the overworked agency.
“At first blush, it sounds like just another crazy idea that is going to overwhelm the capabilities of the federal government to be able to respond to the number of calls coming in and to the number of reports,” Bonner said. “But there is a silver lining: It might just make legislators aware.”
Bonner said it won’t take smugglers long to figure out where the cameras are.
But cameras placed densely along the entire border and watched by people at home could eliminate the need for patrols. The Border Patrol could spend all their time just responding to calls to go exactly to where illegals are crossing. More funds for local law enforcement to do border enforcement could provide the people needed to catch the illegal crossers.
The border is only 2000 miles long and there's 5280 feet to a mile. So how many feet (or meters if you prefer) between cameras would be acceptable to get good coverage?
Combine the cameras with a barrier layer of fences and walls along the entire length and the number of illegal crossers to even report would go down by orders of magnitude.
Governor Perry is acting like a politician who wants to get reelected.
"A stronger border is what Americans want and it's what our security demands and that is what Texas is going to deliver," Mr Perry said.
The cameras will cost $5m (£2.7m) to install and will be trained on sections of the 1,000-mile (1,600km) border known to be favoured by illegal immigrants.
Web users who spot an apparently illegal crossing will be able to alert the authorities by telephoning a number free of charge.
Mr Perry, a Republican, is running for re-election in November.
I like the idea of states operating the cameras rather than the federal government. The states can move more quickly without federal regulations and without federal level lobbyist organizations trying to torpedo immigration enforcement initiatives.
Governor Perry's Virtual Border Watch Program is part of a larger initiative to put more local law enforcement officers along the border.
- Texas will dedicate $20 million in available state funds to sustain and expand Operation Rio Grande, a comprehensive border security strategy ordered by the governor in Feb. 2005, through the current fiscal biennium. These funds will be used to pay for officer overtime, needed equipment like four-wheel drive vehicles, body armor and night-vision goggles, and technology upgrades such as electronic fingerprint booking stations. Funding will not only continue to flow to border sheriffs, but will also go to local police departments and law enforcement agencies up to 100 miles from the border.
- Perry will ask the Legislature in the next session to authorize $100 million to sustain Operation Rio Grande until "the federal government fulfills its responsibility of securing the border." With these funds, said the release, Texas will be able to increase law enforcement's presence on the border by the equivalent of 1,000 additional officers. "Putting more officers on the ground has always been the best strategy for reducing all types of crime, from misdemeanors to drug trafficking and human smuggling, and this new commitment will make Texas safer," Perry said.
Drug smuggling and general crime will also go down as a result of a system of cameras combined with many more officers available to catch crossers.
Imaging software these days is capable of discerning bushes and coyotes moving from humans. Since the cameras are networked, there is no need for a local computer per camera for this analysis, but powerful workstations that can handle many cameras each. The latest Sun and SGI machines should do the job.
This is pretty dumb. Think of the bandwidth. A processor strong enough to encode images or video is strong enough to do motion detection.
They should at the very least have robust motion detection working before any human sees a thing. I work in computer vision. This is exactly what I do. I could prototype the necessary software in a day.
There is a great deal of money spent on automated site surveillance by the military and various other government braches. It isn't perfect, but it is good enough for a border this large.
Then again, a wall is much cheaper and more effective solution than an array of cameras. They aren't mutually exclusive. Quite the opposite.
How do you know a wall would be cheaper?
What would an array of cameras cost to cover 2000 miles?
"How do you know a wall would be cheaper?"
Well, a camera system is pretty ineffective without proper enforcement or actual barriers, so the price comparison is a bit useless. I meant just to say that building a wall without a camera system would work pretty well. A camera network without a wall wouldn't.
Good cameras can cost thousands - especially IR. A regular PC would work for the computation, but there is cost in making each unit rugged enough to last in the field for, hopefully, years. There is also the network infrastructure to get video from any source.
An omnicam is a mirror/camera system that can see the entire 360 degree horizon. That would let you see more than a typical 60 degree field of view camera, but you take a hit in resolution. Lower resolution means that you can't reliably detect objects at a distance. Higher resolution -> greater effective distance -> fewer cameras needed.
If an omnicam could see things reliably N meters away, at a cost of M thousand dollars, and 2000 miles is 3,218,688 meters, a automated surveillance network would cost M*3,218,688/N.
My guess is that the lower bound on cost is $1K and upper bound is $30K. A system with a megapixel camera probably can't see more than 100M away reliably. That makes the end price between $32M and $960M.
That's probably on the high end of distance reliability. It might be closer to 50m.
Either way, twice the high end at $1.9B is cheaper than the wall, right? You've given numbers before on cost/mile that I don't recall.
A wall with 32K automated surveillance cameras would be the best solution for border security. It's amazing it isn't being pursued faster.
I'll post again if I get a more reliable estimate of detection distance for omni-directional cameras.
How are they going to keep the Mexicans from stealing the cameras and selling them?
I very much appreciate your calculations. Border barrier costs: It depends on what we build. From another post, Just a wall would be $3.2 billion.
A border barrier similar to the Israeli barrier with the West Bank would cost well under $10 billion dollars or less than 2 months costs of the war in Iraq. Or we could look at highway construction for construction costs for a wall. The materials that are used to build sound barriers along highways in populated areas would cost about $3.2 billion for a 5 meter high wall 2000 miles long (see my comment below the original post where I calculate out the numbers). There'd be additional costs for barbed wire, sensors, and additional fencing layers as well as an access road. But we could easily afford all this. It'd be similar scope to building an interstate highway along the border.
From yet another post (I've done many posts with border barrier cost info) Senator Jeff Sessions (R AL) says the triple layer fence will cost $3.2 million per mile.
The fence would be built in areas "most often used by smugglers and illegal aliens," as determined by federal officials. Sen. Jeff Sessions, R-Ala., estimated the cost at roughly $3.2 million per mile, more than $900 million for 300 miles.
At that cost we could build a barrier fence along the entire US-Mexico border for about $6.2 billion dollars.
From yet another post I used Israeli costs as a basis: A July 7, 2004 article puts the cost of the Israel West Bank barrier at $1.56 billion.
In all, the complex of fences, concrete walls, trenches and razor wire is to run 425 miles and is one-fourth complete. It will cost Israel about $1.56 billion. Work had been scheduled to be finished by the end of next year.
That works out to $3.67 million per mile. To build an equivalent high security barrier along the almost 2000 mile US-Mexican order would therefore cost in the neighborhood of $7.34 billion. But I found another estimate of the cost of the Israeli barrier that worked out to $4.15 million per mile.
So do we want a wall and 3 fences all in parallel? We could spend $10 on a wall and 3 fence layers with access road. Then 1 or 2 billion more for a very deluxe implementation of cameras. For at most $12 billion we could have 3 fences, access roads, a wall, and cameras on the internet for the entire border length.
If the cameras are on high poles they'd be hard to reach. But the idea is to have enough Border Patrol around to swoop in on Mexicans not long after the Mexicans are spotted.
"If the cameras are on high poles they'd be hard to reach. "
A good defense strategy would be to place the camera posts every 10 meters and only have live cameras every 10 +/- 2 posts.
Note that a system of microphones would also allow gunshot localization at a great distance. iRobot makes something called REDOWL, which is a head for the PackBot line of robots. It automatically detects different types of gunshots, aims it's pan-tilt-zoom camera head at the location (known in 3D), and reports the event. Systems like this will be coming to your friendly neighborhood traffic cam soon -- without the robot mobility base though.
But the idea is to have enough Border Patrol around to swoop in on Mexicans not long after the Mexicans are spotted.
I would agree with that if I though the Bush administration would do it. I don't think they will. What is likely to happen with a camera only option is that you get a lot of last picture frames of Mexicans as they steal the cameras.
I think the best options are a) crack down on employers who hire illegals b) build a great big wall c) have enough border patrol to arrest those who somehow get past the great big wall d) as an option, add cameras for the few who make it past a-c.
A camera only option means that not only do we have people smugglers and drug smugglers, but now we have camera thieves.
A camera only option seems to me to be like a gift to the illegals. They would now have even more reason to cross the border, so that they could steal our unguarded cameras.
But Paul, you are not seeing the big picture. With all those extra cameras flooding the market, just think how cheap it will be to buy replacements from, uh, suppliers in Mexico. ;-)
Isn't one of the reasons they want to put the cameras on the net instead of motion detection for public relations? The citizens can now see that no one is getting accross!
Is there a danger that the authorities could fake the camera images to show nothing all the time just to shut down the border issue?
People would stand in front of these cameras so that their friends could see them on the internet. Private people would test the cameras in various ways. The Minutemen could organize systematic tests with volunteers.
Note that state of the art detection and tracking algorithms still have false alarms. 1 false alarm a day means 30K images for people to see. Every real person is likly to be detected. Thats hundreds of thousands more.
With a great percentage of images on the web those containing actual motion, the PR will be stronger. Rather than have the vast majority be empty, people will see real events.
Whats the addres to the site... it has started..
mail me at
Advertise here. Contact randall dot parker at ymail dot com