Regular SIW columnist Fredrik Nilsson, who serves as general manager of Axis Communications, will be contributing a number of "Eye on Video" segments examining current technology and technological transitions affecting the business of network (IP) video s
IMAGE 1: Megapixel video surveillance technology really comes into the spotlight when you need to identify persons/things, and can be applicable at locations like a facility entrance.
IMAGE 2: Aspect ratios affect surveillance camera choices. A wide aspect ratio would be sufficient for this library, where it would be unnecessary to have the ceiling in this 4:3 frame recorded. Using megapixel camera technology gives security users the c
IMAGE 3: Megapixel technology can really shine in the retail environment. In this scene, a megapixel camera could either be used to provide an overview of the cashier register station, or could be used to see fine details that might on occasion be part of
IMAGE 4: In this image from a shoe store, a single megapixel camera could be used by loss prevention and security staff to digitally zoom in on areas of interest, without having to install numerous non-megapixel cameras to cover those areas.
Megapixel technology has realized significant improvements over the years so much so that it is now one of the dominant forces behind buying decisions of many digital devices. Think about it. Snapshot digital cameras purchasing decisions - both by professional and amateur photographers - are driven largely by how many megapixels a camera offers. Buying a better camera for many consumers means buying one with more megapixels.
The same is true for cell phones. Megapixel technology has influenced the development of cell phones, which has profoundly changed social behavior. Thanks to advances in megapixel technology, cell phones are increasingly being used as viewing devices, taking on more important roles in crime prevention, journalism, business applications and for personal use. The crisp images shared today are a far cry from the fuzzy, unreadable images which cell phones captured just a few, short years ago.
Network cameras are also seeing and experiencing the megapixel revolution, a major benefit that cannot be achieved by their analog counterparts. Network cameras are today estimated to be around 10 percent of the video surveillance market and are growing very quickly at a pace of +40 percent a year. Megapixel is one of the technologies further accelerating this growth.
See What You Have Been Missing
Megapixel technology provides higher resolution, which enables organizations with security installations to see more detail. In fact, megapixel cameras offer several times better resolution than analog cameras. Why does this matter? Organizations will have a much easier time identifying people and objects, making their security system more effective (see Image 1, at right).
Megapixel network cameras can also cover larger scenes than non-megapixel network cameras at a given number of pixels per area. For example, one 2.0 megapixel network camera will cover an even larger area than four non-megapixel network cameras combined without loss of image resolution. The result equals better coverage, less cameras and cost savings.
Another benefit to megapixel technology is that it helps define objects when viewed in different aspect ratios. In a conventional TV monitor, for instance, an aspect ratio of 4:3 is provided. Network video can offer the same ratio, in addition to others, such as 16:9. The advantage of a 16:9 aspect ratio is that unimportant details usually located in the upper and lower part of a conventional-sized image are not present and do not take up bandwidth and storage space (see Image 2, at right).
When to Use Megapixel Cameras
Megapixel cameras are a great compliment to non-megapixel cameras optimized for other needs such as powerful optical zoom, extreme light sensitivity or low cost. Together, these cameras provide a lot of punch behind a complete, reliable and cost-effective security system.
When designing a system, an IT or security manager must decide where he/she needs cameras and what cameras work best in a particular setting. After all, some cameras are more suitable for certain environments than others. One consideration that should be made is about what needs to be seen. Does a security manager require a general overview of an area or does he/she need to be able to identify a person or an object within a scene? The answers to these types of questions will help determine what kind of camera is most appropriate for the installation.
If a general overview is the goal, cameras should be installed to watch people and their movements, not necessarily the identification of individuals. Cameras may also be used to assess whether a parking lot is full or has empty spaces, not necessarily to identify cars and/or license plates. In this scenario, sufficient resolution and coverage may be adequately achieved by using a single megapixel network camera or multiple non-megapixel network cameras.
On the other hand, if the goal is to identify persons and/or objects in a scene, cameras with higher resolution will be needed. For example, retail stores may need surveillance cameras for point-of-sale monitoring where managers can see every item a person is purchasing or to identify a face (see Image 3, at right). Managers have the option of using network cameras with telescopic lenses or lenses with zoom capabilities, or they can place the camera closer to the area being monitored. Retailers can also install megapixel network cameras, which will provide even higher resolution images that provide more details than non-megapixel cameras.
In either scenario, managers must follow best practices to ensure maximum security. It is important to note that a conventional CCTV camera providing 4CIF (Common Intermediate Format) resolution offers a resolution of 704x480 pixels (NTSC) or 704x576 pixels (PAL) after the signal has been digitized in a DVR or a video server. This equals a maximum of 400,000 pixels or 0.4 megapixels. For an overview image, 20 to 30 pixels are generally considered enough to represent one foot of a scene. For areas that require more detailed monitoring, the demands can rise to as much as 150 pixels per foot. For example, if a security team wants to be able to strongly identify people passing through an area that is seven feet wide and seven feet high, the camera needs to provide a resolution of 1,050x1,050 pixels, which is slightly more than 1 megapixel.
Other uses of megapixel technology
Megapixel technology can also be used in non-mechanical PTZ cameras. Conventional PTZ cameras have moving parts that enable users to pan, tilt and zoom to see one portion of the room at a time but not the entire scene at once. The wide angle non-mechanical PTZ using megapixel sensors shows the whole scene and lets the user pan, tilt and/or zoom to get more details, and does so without moving parts (see Image 4, at right).
By using the network camera in an overview setting, a security officer can zoom in on any suspicious behavior by just clicking on the part of the image where it is happening. No other PTZ camera can zoom in instantly on off-centered action. Plus, since there is no movement in the lens system, the camera instantly changes the field of view. Zooming will not decrease image quality and the lack of moving parts and users get a wide-angle lens combined with a three megapixel sensor.
When not to use megapixel technology
While megapixel network cameras provide clear benefits by way of image resolution and versatility, it is important to be aware of potential drawbacks with current megapixel technology. For manufacturing and cost reasons, many megapixel sensors are the same size or only slightly larger than sensors with less resolution. This means that each pixel is smaller in a megapixel network camera and consequently less light sensitive than a non-megapixel network camera.
In addition, since megapixel technology produces higher resolution video, it will increase demands on bandwidth and storage space. While bandwidth in modern networks is plentiful, storage will come at a cost and moving from VGA to 1.3 megapixel cameras will quadruple the storage cost at a given frame rate. Compression is important when using megapixel resolution cameras, and MPEG-4 can offer benefits here.
Many megapixel cameras have limited frame rate support while VGA resolution cameras for the most part support 30 fps (frame per second) and some even 45 fps or 60 fps. Many megapixel cameras only support 15 fps or less. For some applications this might be a limitation.
The Future of Megapixel
While VGA/CIF resolution cameras are the standard today, in a few years 1 and 2 megapixel cameras will be the standard. Additionally we can expect improved light sensitivity, higher resolutions and higher frame rates from megapixel cameras.
Higher resolution will make it possible to build even more advanced non-mechanical PTZ cameras, with higher coverage and zoom. This technology may eventually replace the most advanced PTZ dome network cameras of today - but not anytime soon. To cover 360 degrees and achieve 35x zoom a 500 megapixel sensor (yes, 500!) would be required, and that type of product is still many years out. Better compression such as H.264 will reduce some of the storage requirements and will accelerate megapixel deployments.
There is no question that the capabilities of megapixel technology will continue to significantly impact the network security market in addition to having a larger role in influencing mainstream buying behavior. The exciting part will be watching the technological advances unfold and adoption of IP-based technology increase, producing even more efficient and cost-effective systems.