With multi-streaming, you can designate and track multiple user-defined windows within a single camera frame.
Photo credit: Image courtesy Axis Communications
Fredrik Nilsson is general manager for Axis Communications in the Americas and author of the book "Intelligent Network Video".
When opting for network cameras, users often cite their numerous advantages over analog technology: easier installation, better total cost of ownership and better scalability, among many others. But what's often the most compelling advantage for users is the perception that the higher resolution of network cameras will provide better image quality and greater detail. While this theory is certainly true for megapixel and HDTV cameras, there are a few limitations you need to consider before deciding whether a higher resolution camera is the right choice for your application. Once you've weighed the pros and cons of using higher resolution technology, you will need to select which camera to use among the vast array of vendor models available today. A checklist at the end of this article will help you compare various manufacturer offerings.
The caveats of higher resolution
Though megapixel cameras are designed to deliver higher resolution, the greater detail comes with strings attached. The two main things you need to make a megapixel system work as promised are ample lighting and ample storage capacity.
- Low-light performance. High resolution image sensors contain more pixels than standard-resolution VGA sensors, but oftentimes are of similar physical size. The implication is that the sensor manufacturers simply reduce the size of each pixel to create more pixels per area. For example, a 1.3 megapixel sensor contains four times as many pixels as a VGA sensor, which also means that each pixel is only a fourth of the size of a standard pixel on a similar sized VGA sensor. The drawback of the smaller pixel size is that it reduces a sensor's ability to capture light. In other words, a typical megapixel sensor is less light sensitive than a standard VGA sensor. To achieve optimal performance you need ample lighting in the area under surveillance. In most indoor environments, such as schools, offices and retail stores where sufficient lighting can be guaranteed at all times, megapixel and HDTV cameras will work just fine. In environments with more challenging light conditions, such as a parking lot at night, a megapixel camera might not be sufficiently light sensitive. If it isn't, you may have to add illumination to the scene to get the most from your megapixel or HDTV camera, which will add cost and complexity to the project.
- Accommodating additional storage. Simply put: Greater image detail translates into larger image files. The math is pretty straightforward. If you're capturing four times the resolution of a standard VGA camera, you'll need four times the storage. So if you currently spend X dollars per gigabyte of storage, you'll need to increase your storage budget by a factor of 4X to archive video from each megapixel camera in your surveillance system. Additionally, the network traffic will increase by the same factor, which might be a challenge in some networks.
A comparison checklist
Once you've taken lighting and storage into consideration and decided that a megapixel or HDTV is indeed the right camera for your application, the next step is to evaluate which camera is the best fit for your environment. To ensure you're making an apples-to-apples comparison between vendor cameras, you need to consider the following:
Compression. As previously noted, megapixel or HDTV cameras generate a large amount of data. To avoid geometric expansion of your storage farm, efficient compression is essential. Ask your vendor if the camera you're considering supports the advanced H.264 compression standard. H.264 is a standard that can reduce the size of a digital video file by more than 80 percent compared with Motion JPEG, and by as much as 50 percent compared with MPEG-4 Part 2. Just as important, compare what compression format the camera supports when it is running at the highest resolutions and at high frame rates. Not all megapixel cameras can use H.264 when the camera is set for the highest resolution and highest frame rates; this is because H.264 requires a very highly-capable compression chip inside of the camera.
Frame rate. Some megapixel cameras provide very high resolution, but only at lower frame rates. In some surveillance applications, three or five frames per second (fps) might be sufficient, while others might require 12 or even a full 30 fps. If a high resolution camera is HDTV-compliant (according to SMPTE standards) it will guarantee 30 fps.
Day/Night capability. If you plan on placing the camera outdoors, or in an indoor environment that has poor or varied lighting at certain times of the day, you need to choose a camera that supports true day/night capability. True day/night capability means that the camera is equipped with a removable infrared-cut filter. Near-infrared light, which spans from 700 nanometers (nm) up to about 1000 nm, is beyond what the human eye can see, but most camera sensors can detect it and make use of it. During the day, a day/night camera uses an IR-cut filter to block IR light so that it doesn't distort the colors of images as the human eye sees them. When the camera is in night mode and the IR-cut filter is removed, the camera's light sensitivity is drastically improved and put into black and white to avoid distortion of colors. Keep in mind that some cameras do not have a removable IR-cut filter, but simply switch to a black and white image at night to give the illusion of night vision by minimally increasing light sensitivity. These not true day/night cameras. Other cameras increase their light sensitivity at night by reducing their resolution. For instance, a 1.3 megapixel camera would combine four pixels into one, thereby increasing the light sensitivity, but at the expense of reducing the resolution by a factor of four. This method brings actual resolution down to VGA quality after dark. If nighttime surveillance is a priority for your installation and you're willing to settle for lower resolution at night in order to get greater light sensitivity, a standard VGA camera could be an even better alternative than megapixel.
Multi-streaming. While megapixel and HDTV cameras do provide a lot of image details, it's rare that you use all that information when doing live monitoring, especially when viewing four or more cameras on the monitor at the same time. In those instances, it would probably be best to stream megapixel images to the storage system for forensics, but use a low-resolution video stream for the live monitoring. With cameras that support multi-streaming, you can customize the system for the optimal performance for both live monitoring and archiving. For instance, you can create different compressions and resolutions for each stream, such as one lower-resolution stream for monitoring and a second full frame rate, high-resolution stream for archiving and forensics. Also, some cameras support an even more advanced multi-streaming technology, allowing the field of view from one camera to be divided into multiple, separate, user-defined streams. This allows the user to use a single megapixel camera to cover a wider field, and segment smaller video streams from that image into their own individual windows (see photo at right). Part of the theory behind this format of multi-streaming is that it allows a user to replace several individual cameras with a single megapixel or HDTV camera.
Lenses. Camera optics makes a big difference. With respect to image quality, the camera is only as good as the lens it uses. Check that the lens is rated for a specific resolution to provide the image clarity and quality you expect, or else your investment in megapixel may be wasted. Also check with the vendor that the lens can handle difficult light scenarios and deliver the depth of field you need. [From Axis: A camera with precise-iris control -- often referred to as P-Iris -- is a special lens with in-camera software used to guarantee depth of field and better image quality. P-Iris automatically adjusts the iris opening to the optimal aperture for image sharpness, depth of field and exposure based on the lighting conditions at any given moment.]
Below: Notice the difference in image clarity in an environment with challenging light conditions. The image on the top was captured with a traditional surveillance camera lens, while the image on the bottom was recorded using a surveillance camera equipped with a P-Iris lens.
Are megapixel and HDTV cameras right for you?
Megapixel and especially HDTV are breakthroughs for surveillance, providing you with unprecedented clarity and detail. But to make an educated decision as to whether to deploy the technology, you need to understand how the potential drawbacks will impact your surveillance system. If the benefits outweigh the limitations, then the next important step is to comparison shop among vendors to make sure the megapixel camera you choose supports the features you need. Because of the complexity of these types of cameras you will quickly realize that not all megapixel and HDTV cameras are created equal.
About the author: Fredrik Nilsson is General Manager of the Americas for Axis Communications and author of the book Intelligent Network Video. He is a regular expert contributor on topics of networked video surveillance systems and cameras.