Field Guide to HD & Megapixel Cameras
From compression, to storage, to lenses, and more, project designers and technicians will find this guide to be a useful resource on the desk and in the field.
If you had to identify the topics that seemingly everyone is talking about, high-definition and megapixel video would rank at or near the top of the list. Based on the their proliferation in the consumer market — most notably TVs and digital still cameras — end users have come to expect the same clear, crisp, detailed images from everything. That includes security.
In some cases, this poses a problem for installers and integrators who are on the front lines, dealing directly with end users. On one hand, a “bigger, faster, now” mentality rules the consumer side, as manufacturers continually push the envelope when it comes to resolution. On the other hand, security technologies typically lag behind the consumer market — and rightly so. When it comes to safety and security, the margin for error is slim or nonexistent. So manufacturers have to take their time and ensure that their products and technologies are “just right” before unleashing them in the marketplace.
According to several industry insiders, based on advances in HD and megapixel camera technologies, the age of HD and megapixel video may be just around the corner. IMS Research predicts that 30 percent of worldwide camera shipments this year will be HD or megapixel. The firm predicts that these technologies will continue their rapid ascent and reach a tipping point and surpass 50 percent in 2014. Continued growth should push that number beyond 60 percent in 2015. To Raul Calderon, senior vice president of Arecont Vision in Glendale, Calif., those projections seem realistic.
“In the market, I would say a conservative estimate is a 65-to-35 standard-to-high-resolution split right now. In the next three to five years, it will be the other way around,” he says.
Of course, whether these numbers ultimately prove true remains to be seen, but what’s clear in the industry today, regardless of what the research says, is that demand for HD and megapixel cameras is extremely high. Installers and integrators who aren’t tapping into this lucrative market already need to get started — today.
‘Good Enough’ Is No Longer Good Enough
The obvious advantage to HD and megapixel video is better detail. While past video systems were “good enough,” that’s clearly no longer the case. In some cases, video produced by systems installed within the last few years is too blurry and pixelated to be of much use to anyone. Put them side by side with even the lowest-quality HD or megapixel system available today and the difference is clear.
“It’s kind of a bad word, but customers love that they’re able to get more of a ‘CSI’ thing from their video system,” says Todd Pinnell, product manager – video for Speco Technologies, located in Amityville, N.Y.
Because forensic capabilities are the sweet spot for HD and megapixel, those types of high-quality video have become essential for almost any application, Calderon says. And, he adds, while higher-megapixel cameras are available, in most cases those higher resolutions can actually be a bad thing.
“We’re still seeing that HD — which is 2.1 megapixel — is a good choice. Above that, you can’t resolve all the pixels and you start to see degradation,” Calderon says. “From a forensic perspective, it’s all in the details. If the detail isn’t there, there’s not much you can do with it.”
It’s a universally accepted fact that any conversation about high-resolution video will ultimately circle around to bandwidth. Because they serve as the first stage in any video surveillance system, that goes double for conversations about cameras. It’s one thing to capture video, but it’s another thing altogether to transmit it across an IP network for viewing and/or storage.
HD and megapixel cameras have complicated this issue, based on the sheer volume of data they collect in order to provide high-quality images as promised. What consumers fail to realize, says Scot Jolma, solutions engineer for Park Ridge, N.J.-based Sony Security Solutions Group, is that transmitting security video is a completely different animal than transmitting broadcast video.
“It’s easier with broadcast because you’re dealing with a controlled environment and specific channels for transmission,” he says. “In a security application, you have to transmit video from a less controlled environment across a lot more channels.”
That’s where video compression comes in. However, each of the major compression standards — JPEG, MPEG-4 and H.264 — comes with its own pros and cons, which can make choosing the right one for an installation a challenge, Jolma explains.
While JPEG offers the best quality, it requires the most bandwidth, making it the least network-friendly. For that reason alone, JPEG is often left out of the compression conversation. MPEG-4 requires less bandwidth but offers lower quality. H.264, which has taken the lead in the compression race, marries the two, offering high quality and low bandwidth consumption. Its main drawback is computer processing requirements for compression and decompression. According to Jolma, H.264 requires an additional 15 percent of processing power per video stream.
“If you’re only processing one stream, it’s no big deal,” he says. “If you’re talking about 16 or 32 streams, you’re looking at a significant difference between MPEG-4 and H.264.”
In the end, it comes to weighing the demands on the network against requirements for video quality. Most manufacturers provide some sort of tool for determining the bandwidth requirement based on camera, application and more. Aside from these, some basic know-how also can be helpful.
“A good baseline or rule of thumb for calculating bandwidth requirements is this: A 1.3 megapixel or 720p camera in real-time H.264 can be estimated at 4 Mbps for a high-quality image,” says Aaron Saks, product trainer at Panasonic System Networks Company of America in Seacaucus, N.J. “Bandwidth can be reduced down to 2 Mbps for lower quality, or when there is little change in the image.”
Luckily, says Paul Bodell, executive vice president, Global Business Development, for San Juan Capistrano, Calif.-based IQinVision, concerns about many of these issues are becoming moot.
“Even three years ago, the biggest questions about networked video revolved around whether there would be enough bandwidth and storage available,” he says. “Today, those are mostly gone.”
In the earliest days of IP cameras, end users were particularly interested in on-board storage. Not only was this a less expensive option — particularly in light of decreasing storage costs — than a network video recorder, it was also something that end users were more comfortable with. In most cases, they were dealing with the same kind of memory card they were using in their digital still camera at home.
Perhaps the hottest storage solutions in the industry today combine HD and megapixel video with an up-and-coming concept in the industry: The Cloud. Thanks to Microsoft and others, as was the case with memory cards, the cloud is a concept that’s familiar and comfortable to end users.
“Since HD and megapixel cameras create enormous quantities of video data, using powerful storage technology has been paramount,” says Xiangquin Ying, product manager, for Hangzhou, China-based Hikvision. “Network attached storage can be used for centralized network data archiving, and NVR server can achieve temporary video caching.”
Another way to rein in storage needs, Saks says, is not to swat at a fly with a hammer. In other words, use a camera that will accomplish your security objectives without overdoing it.
“You hear a lot about higher and higher numbers of pixels, but I think the core opportunities for market growth are in the range of 1.3 to 3 megapixel cameras,” he says. “These cameras provide a noticeable improvement in image quality and can be incorporated into systems easily, especially given H.264 compression that helps to keep bandwidth and storage needs low.”
Several new technologies are transforming HD and megapixel from simply image-capturing and storage devices to more complete solutions.
“We can do a lot more at the edge, things like intelligent video systems and trip lines, so these cameras are becoming small computers,” Jolma says.
The intelligence that can be located within the camera includes processing power, lighting optimization, camera and video controls, and third-party software that can optimize a camera’s performance. These are just the tip of the iceberg, Saks says.
In terms of processing alone, he describes, camera intelligence can manage the dynamic range (the span of gradations from lightest to darkest); optimize each pixel’s contrast to faithfully reproduce objects in any area and position; produce better images in extreme lighting conditions; and transform dark areas into natural, high-contrast images. It’s technology innovations like these that are transforming and expanding system capabilities while simplifying system design, Saks says.
Perhaps the most talked-about technology that’s being moved to the camera itself is video analytics, notably facial recognition, motion detection and license plate recognition. Traditionally, most of the analytics algorithms have been best suited for standard-definition video, Calderon says. Newer analytics offerings take full advantage of IP-based megapixel cameras and allow for better decision-making and assessment of a scene.
“Basically, they’re allowing smart systems to become smarter,” he says.
In light of all of these developments, you may be wondering what kinds of specialized equipment, skills and knowledge are required to work with the latest HD and megapixel cameras. The short answer, according to Jolma, is knowledge of networks, bandwidth and how to manage network storage.
While this answer may seem like an oversimplification of the issue, there’s a degree of truth to it, especially when it comes to HD.
“Installing HD is very simple. It’s using the same cable that may already be in place for an analog system,” Pinnell says.
However, the story is much different when it comes to megapixels.
“The technological know-how is coming down, but it’s still not as easy as HD,” Pinnell says. “Is it easier than it was four or five years ago? Yes, but you can’t get away with the way things were done 10 years ago.”
The greatest consideration from an equipment standard is perhaps the most basic element of a video system: The lens. Unfortunately, choosing the correct lens for HD and megapixel cameras is one of the most often overlooked aspects. (For a full discussion about lenses, see sidebar below.)
Keeping Costs Down… to a Point
One of the greatest advantages — and selling points — of HD and megapixel cameras is the promise of reducing the number of cameras. That’s all well and good, Calderon says, but at the same time, installers and integrators also should be looking at increasing resolution. For example, if it takes eight standard-definition cameras to cover a wide-angle scene, two megapixel cameras with wider angles may do the job because the pixel density of the scene is greater. But if you use four of those higher-megapixel cameras, you’ll get twice the resolution while still reducing the total number of cameras.
Bodell says in addition to reducing the number of cameras, a lot of customers want to replace their PTZ cameras, and high-megapixel cameras are the best option for doing that. “The knock on PTZ is that they’re always pointed in the wrong direction,” he says. “With megapixel cameras that cover a wider area with fewer cameras, that’s not an issue anymore.”
With decreased total cost of ownership comes greater opportunities and expansion into the mainstream, Bodell says. “Five years ago, two or three verticals were enthusiastic about high-resolution, but now it’s in the general population,” he says. “It’s great for wide coverage applications or anywhere someone needs detail like cash-counting rooms, cash registers and pharmacy distribution centers.”
To Infinity and Beyond
As the IMS Research data shows, the sky’s the limit for HD and megapixel cameras’ potential. Among the developments that should drive continued growth are smarter features, higher resolution, continued advances in compression and storage, and lower prices.
Advancements in wide dynamic and low-light technology will be a key focus of the HD market, Ying says. “Even though HD and megapixel can offer more image details, when the cameras are used in reality, how to adapt into the real environment becomes a challenge,” he says. “Therefore, wide dynamic and low-light technologies can be used to solve the problems. For example, wide dynamic technology is required in applications with extreme lighting contrast environment.”
In fact, Bodell says, light sensitivity has come a long way in recent years. “Three or four years ago, standard-resolution cameras were 30 percent better than low-light megapixel cameras,” he says. “Today that difference is negligible and future CMOS in high-megapixel cameras will surpass that in the next year or so.”
The continued growth and evolution of smart features, especially video analytics, also will be a major factor, Jolma says. “We’re going to see more behavioral-type intelligence like certain types of motion and certain objects,” he says. “The Holy Grail is learned behaviors. Right now, no one can do it for general applications. Some specific applications? Yes. So the company that gets it right will have a leg up.”
Jolma’s measuring stick for how rapidly advanced technologies will be available is the consumer video segment, which he says tends to be two or three years ahead of security.
“Right now, 3D imaging is big on the consumer side. In security, that would allow you to tell the distance to an object,” he says. “That’s going to be where the end user wants to go, and companies that are already working on that technology in their consumer sections have the advantage of knowing how sensors, capture, transmission and other factors work.”
But when you boil it down to the basics, what exactly do the two terms mean, and what are the differences between them? And if all megapixel cameras are HD (or higher) resolution, but not all HD cameras are megapixel, what time does the 10:15 train arrive at Penn Station?
Here’s a quick “tale of the tape”:
Q&A on Lenses
What role do lenses play in delivering HD and megapixel video, and what factors have to be considered in choosing the right lens for the right camera?
Scott Jolma, Sony: Lenses are one of the most critical components in HD and megapixel video, and most lenses aren’t designed for high-resolution sensors. Even for megapixel, you’re limited to 3 megapixel resolution for image clarity through the lens. At 5 to 10 megapixel, you can really struggle to find lenses that not only catch up to the sensor, but do it well. Digital SLR lenses work, but they’re not designed for security. For example, there’s no auto iris. Also, they’re very expensive. Lenses will play catch-up, so that limitation is going away soon.
Xiangqun Ying, Hikvision: The focal length and the selection of HD lens would differ according to different camera view and coverage, as well as the requirement of details in the view. Besides clarity, color rendition, relative brightness and distortion should be taken into consideration when choosing a HD lens. Blooming and ghosting are also critical for scenes with multiple light sources, such as traffic monitoring. A bad lens suffering from either blooming or ghosting could affect the image quality and obscure the license plate, interfering with the accuracy of license plate recognition.
Aaron Saks, Panasonic: Lenses are a critical component of megapixel cameras, much more so than in the case of a standard-definition camera when lens quality is less likely to be a limiting factor. Make sure that the lens is megapixel rated to match the camera. In addition, a camera with ABF eliminates the need for a day/night or IR-corrected lens, ensuring 24x7 focus. Otherwise, a day/night camera would need a special IR-corrected or day/night lens to ensure focus throughout. There is actually a great deal of science to choosing the right lens, and this is where knowledge can help.
Paul Bodell, IQinVision: With HD and megapixel cameras, the “gotcha” is lenses. We’ve traditionally lived in a world where you could simply pick any lens, but now you really have to marry the resolution of the camera with the resolution of the lens. However, a select number of lenses are affordable, around $50 or $60. At 10 megapixels-plus, you’re talking about a totally different lens. You plug in what’s basically a 35mm lens, and in most cases, those are as expensive as, or more expensive than, the camera itself. You also have to try before you buy because there are no standard qualifications for lens quality.
Raul Calderon, Arecont Vision: Not all lenses are equal. Optics and lighting are paramount. Light is HD/MP camera’s best friend. The darker the picture, the grainier it becomes. In the past, optical grade wasn’t limiting factor in production. Standard CCTV lenses traditionally offer higher resolution than the camera itself. Now, we’re seeing the opposite. And there is a cost associated with using the right lens; HD and megapixel lenses are two or more times more expensive than CCTV lenses, but they allow you to resolve what a camera is able to produce. It’s like having glasses: Your eye has the ability to see, if the glasses allow.
Todd Pinnell, Speco Technologies: Lens selection is extremely important. In our case, we used 2.8-12 lenses at first, but noticed that around the edges, we would get degradation. We migrated to HD, but had to move by trial-and-error to determine just what kind of lenses to use with which sensors. The sensor is also huge, but I don’t know where the happy medium is between lenses and sensors. Sometimes the difference between the two can be a bit of overkill.