Imaging technology | CCTVImage Vision-ary perspective
Capital investment and on-going revenue costs are always significant, so what can you do to cope with the unexpected? As climate change prompts unusual environmental conditions, CCTV Image investigates how to avoid being caught out by this and other similar issues in future
THE SCENE is set. Your cameras have been specified, purchased, installed and commissioned. Tied-back trans- mission links are up and running, with recording ready to go. Operators have familiarised themselves with the new set of ‘eyes’ on the ground, while on-going maintenance arrangements are in place to ensure the equipment is kept up to scratch in future. Other system interfaces have been sorted out, with testing completed and software gremlins ironed out.
So the surveillance units begin operation, the mayor has his photo taken in the control room, and the local papers duly publish the news story. This is sometimes the stage when tree foliage begins to obscure the camera views, as the onset of spring brings out the leaves and blow a hole in the winter trials!
However, the wrong kind of leaves aren’t the only operational headache to befall hapless system managers, for a variety of other hazards lie in wait to potentially foil optimised surveillance views.
Before and after images, showing Halo-3 capabilities
Strong winds can set up significant vibration for camera mounts including poles, blurring the picture and prematurely tiring the opera- tors who suddenly have to cope with it. Equally, poor scene lighting, both day and night, and hazardous weather conditions, can intervene at the most unexpected times to ruin the crucial moment an incident occurs or an intruder is spotted.
Low light camera capabilities, and ambient lighting levels in town and city centre areas, are all well and good. But how do you cope with strong backlighting or the sun’s glare when low in the sky? What hap- pens when white light is not available, environmentally unacceptable, or simply too costly to install and/or run? Equally, when snow, fog or even mist descends, you’ll need con- tingency arrangements. Moreover, surveillance in out-of-town areas where intruders can hide from view behind bushes, trees and other natural camouflage demands additional measures. So how can all these challenges be resolved?
Common problems
The common problems faced by image-based surveillance systems, in applications such as fixed and mobile installations, covert CCTV moni- toring and unmanned aerial vehicles, include poor raw sensor imagery and challenging lighting and atmospheric conditions.
These issues have led to increasing demand for infrared imaging and the need for adaptive contrast enhancement. Capitalising on the emergence of high performance computing platforms with low size, weight and power, cou- pled with compact visible and IR sensors, Isle of Wight based elec- tronic systems designer RFEL has devised a low power solution for portable use.
Launched at Counter Terror Expo last month, Halo is an off- the-shelf video enhancement solu- tion that’s said to deliver high
The Halo box of tricks
frame rate HD video for human or machine interpretation. Its capabili- ties are claimed to include intelligent fusion of multi-modal imagery such as from a visible and IR sensor, image stabilisation, contrast enhancement in marginal lighting conditions, and noise reduction in low ambient light.
Halo uses Field Programmable Gate Array (FPGA) technology to achieve the high processing bandwidths needed while maintaining low power consumption. At the system’s heart is a proprietary image fusion ‘engine’: differing imaging wavebands, typically daylight and IR (though different input combinations are possible), provide comple- mentary information and a composite image is formed based on select- ing the best features from both images on a per-pixel basis. Key image features from both modalities are retained, with poor quality and replicated information discarded.
RFEL’ s technical sales engineer & project manager, Wayne Cranwell, told CCTV Image that image stabilisation is important, since presenting the maximum available image information to the operator is largely wasted if the viewing platform is subject to random movement due to wind or vibration. Stabilisation also significantly improves the efficiency of standard compression algorithms and considerably reduces operator fatigue, improving detection rate accuracy. “Operators can cope with vibration, but it’s an extra job they have to do on top of their other tasks and removing this distraction makes the job of recognition and identification easier,” says Cranwell. Cranwell adds that Halo’s sophisticated image stabiliser processes and extracts all useful image information and therefore works in the most challenging circumstances, unlike the computationally similar approach of feature tracking.
“There are various image stabilisation techniques, either based on mechanical correction or image processing,” Cranwell says. “Mechanical methods tend to be more expensive, consume more power, and are physically larger and heavier. Those carried out in-cam- era are generally cheaper and more compact, but can have performance limitations such as an inability to correct for roll, and may operate over a restricted range of unwanted camera movements. “Additionally, such integrated camera techniques are less well established for infrared cameras and units using interchangeable lenses. Mechanical stabilisation corrects for movement associated with the camera, but doesn’t correct for other effects such as atmospheric scin- tillation – ie visual aberrations such as moisture, smoke or heat haze.”
Spring 2013 24
Issue sponsor
www.securitynewsdesk.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36