Color & Detail Benefit in Big Way
Enhanced color in low-light applications is probably the most significant beneficiary of these noted advancements and improvements. CMOS imagers are one area where camera manufacturers have put a good deal of effort and monetary investment. Through several advancements in chip technology, CMOS imagers are currently capable of providing usable color images at lower light levels than ever before.
“We use low-light cameras — no thermals as there is no demand for it — especially in self-storage [facilities applications]. We normally install standard day/night cameras, which means you have a normal picture and during the night the IRs come on. Most of them perform well in black and white during nighttime as they illuminate the area very well,” says Steve Gumpher, vice president with Automated Security Corp. of Goodville, Pa. “Other than IR, there are times when we rely on the new LED high efficiency lighting that some of the newer self-storage facilities install.”
The other key part of this success story involves major advancements in data compression and general signal processing. For example, today it’s possible for end users to view color in environments where the light level is as low as 0.017 lux, 0.0057 in monochrome (depending on whom you talk to) at 60 frames per second (fps), using a CMOS megapixel camera.
However, in low-light situations CMOS-based MP cameras can produce a good bit of digital noise. This can be misinterpreted by the encoder as motion, causing the signal processor in any average IP camera to update the base image inside the system by sending this noise up the cable as video data. What this essentially does is drastically increase network bandwidth requirements as well as storage capacity.
How do we tell what’s noise and what’s real data? To get cleaner, crisper images, we use software to do things with these images to improve them. We can fill in the gaps and holes in the image as to what should be there — we backfill to get a higher quality image.— Sean Murphy, Bosch
Because 50% or more of the cost of a digital video surveillance system is in data storage, you can see how digital noise can quickly result in an increase in operating costs. Thanks to the introduction of improved signal processing features and capabilities, along with traditional DNR (digital noise reduction), these cameras reduce the likelihood of this of happening.
“We have intelligent algorithms that can differentiate between noise patterns and scene motion, which helps us to remove noise much earlier in the image processing chain. Therefore we can apply more image enhancements in low light without adding extra noise to the images,” says Murphy. “How do we tell what’s noise and what’s real data? To get cleaner, crisper images, we use software to do things with these images to improve them. We can fill in the gaps and holes in the image as to what should be there — we backfill to get a higher quality image.”
These sophisticated algorithms also allow the user to specify areas of interest in an image, setting priorities for each one. This results in more system resources being devoted to creating especially clear, crisp and identifiable attributes within those spaces while paying less attention to the unspecified areas of the picture. Murphy makes reference to Bosch’s stable of signal processor/noise reduction software as Intelligent Dynamic Noise Reduction (IDNR). Other manufacturers have similar advanced technologies, each having their own special sauce and name.
CHECK OUT NEXT: When Upselling Security, Offer the Customer More Protection When the Case Calls for It
Improved signal compression also is a factor in the advancement of low-light performance. One of the ways that H.264 deals with optimizing the details-to-bandwidth ratio is to fill in some or most of the missing components in the video signal.
“However, depending on the scene complexities and motion contents, the difference in scene or prediction that algorithms can do can be low,” adds Murphy. “That’s why H.264 has dynamic bitrates depending on the scene. However, Bosch has some more intelligent algorithms [e.g., IDNR and encoder regions, as mentioned]. This technology is especially very helpful when we talk about high resolution like 5MP or above.”
Thermal Cameras for Total Darkness
There are applications where there is barely enough reflective light for today’s new and improved CCD and CMOS cameras to derive even a poor B/W image let alone one where the perpetrator(s) of a crime can readily be identified. Think back to the example at the beginning of this article — regarding the border between the U.S. and Mexico; this is precisely the challenging situation that U.S. Border Patrol Agents face as they go about their mission of border protection.
“Well-organized criminals carry out raids on poorly protected storage areas. Misguided fanatics driven by nationalism, ethnic hostility or religious extremism are willing and able to travel across countries and across borders to commit acts of sabotage and terror,” according to SightLogix CEO John Romanowich and Danny Chin, authors of “Smart Video Security Handbook.” “They regard buildings, public spaces, airports and infrastructure as convenient targets of opportunity. And while less dramatic than terrorism, theft and vandalism at outdoor facilities are more common and lead to millions of dollars of lost revenue.”
Instead of using reflective light to create a usable picture, thermal cameras actively see in total darkness using infrared radiation, which is emitted by the human body. A single thermal camera provides people detection for miles — in some cases for 10, 15, 20, 30 or more — thus sounding an alert for security to follow up on. Thermal cameras are not for every application, however.
“Where thermal shines is where you have someone monitoring it in real-time like security guards,” says Brian Young, an engineering manager with Integrated Security Solutions Inc. (ISS) of Kalispell, Mont., which has established itself in the area of protection of high-profile government and corporate assets. “It’s a critical infrastructure application where you have critical assets that need protection.”
Thermal cameras also are extremely helpful where environmental conditions are such that normal cameras may not produce usable, identifiable images. One example includes a seaport where the harbormaster is likely to encounter fog, rain and other adverse elements that can make it highly difficult to see at any distance using normal low-light, reflective light cameras, and this includes daylight applications.
“Thermal is especially suited for intrusion assessment where you have a really long perimeter — say an airport, seaport or something similar — and you need to be able to look out several miles. There are a variety of thermal cameras that can easily look out that range and assess a 15-mile perimeter,” says Young. “If you tried to do that same thing with standard IR cameras, you’re going to need 200 or 300 of them to have the same coverage.”
The type of application might mitigate the budgetary concerns for deploying thermal cameras too. “Even though thermal is expensive as an individualized camera, it may be more cost effective in the long run than trying to put out such a huge number of standard low-light cameras,” adds Young, “in addition to massive amounts of electrical infrastructure, fiber-optic cable and everything else that goes with it.”
Page 2 of 2 pages <