=== Laser dispersion lens ===
Fit a high powered laser diode (880nm) to a laser dispersion lens. Fly the quodro-copter to a specific hight and pulse the laser to flood the ground at night with a brief Infrared dispersed beam, taking a photograph at the same time with a 0.003lux CCD camera. A quad can patrol a large area in this manner fly back to its base and have automated [[ImageProcessing]] done as per [[UavInTheNews?highlight]] -
http://www.theengineer.co.uk/Articles/308178/A+drone+of+your+own.htm.
".....Essex University's Prof Owen Holland, a consultant for one of the 11 competitors, Team Swarm, explained the benefits of using a swarm approach:
'Swarming has many proven advantages. Most importantly, it can survive unexpected events. If one vehicle is suddenly lost, the swarm reconfigures itself to complete the task.' Team Swarm used eight quadrotors called Owls. Their operator used three-dimensional planning software to swiftly plan and rehearse routes for the Owls over Copehill Down village. The craft then flew those routes, taking hundreds of high-resolution images. On their return, these images were processed by a cluster of 10 powerful multi-processor PCs, which analysed each image using the University of Surrey's threat recognition software. Within minutes, the operator had produced a map of Copehill Down showing the location and type of all recognised threats. The winner of the Grand Challenge was Team Stellar, which entered Saturn — an integrated system with one high-flying and one mini-UAV and an unmanned ground vehicle, with a control station fusing data from visual, thermal and radar sensors. It was awarded the highest score by a panel of judges after its vehicles successfully identified a range of threats planted in the village, including actors dressed as militia.........."
Create barrier beams:
Criss-cross a large area with long range [[InfraRedLeds]] as per the
http://www.lirc.org design using a TSOP1838 IC with a magnification lens on Rx and Tx allowing distances over 100m. Populate a farm with hundreds of such beams. On the beam breaking the quad automatically flies to the intrusion point taking a picture during night and daytime hours. Send the picture to India or china for manual verification. All farmworkers carry a GPRS-GPS system to minimize false alarms.
Automated person tracking using facial recognition
A PHD student and his professor recently completed a design that does automatic facial recognition of thousands of people in a CBD area. The computer can tell you exactly who disembarked from a taxi, what shops he visited , if he left his suitcase behind and when he left , what vehicle and number plate. The only reason why this can't be implemented tomorrow in Pretoria and JHB is because there is no bandwidth network controlled in a decentralized manner. With unlimited bandwidth we can place a 0.003 lux camera with wide lens every 50m connected to an Mpeg4compression system streaming hundreds of faces to a central server. For the software to work a certain resolution image needs to be captured, thus more cameras, the less expensive they have to be with less expensive lenses. We can easily stop the crime, the difficulty is in convincing people who have been convinced by Naspers that South africa is doomed that bandwidth is the problem, not more policemen running around like off-head chickens. But AI computer systems doing behavior analysis on thousand of people loitering about the streets of JHB. It is going to take a paradigm shift in thinking and I don't seem to get any support from the tech geeks..... don't really understand what is going on .... the ANC has 1Bil Rands unspent for telecoms for rural areas. Use that money to create Ronjo FSO nodes rather in JHB..... stop the crime..... get more economic growth...... more taxes......
Sign up with Google
Here is another idea: Embed the latest breakthrough in facial recognition(