We live in a culture of more and are inundated with offers for value-sized meals, bigger TVs and buy-one-get-one-free deals. Clearly more is better, right? If you apply this philosophy to surveillance cameras, then resolution would be even more important and the megapixel race would be just beginning. But in surveillance, it is important to look at the whole picture — and resolution is just one factor when determining image quality. More is not always better.
There is no direct correlation between resolution and image quality in physical security, despite some industry claims. With the latest technology advancements in surveillance cameras, it is more valuable to discuss image usability than image quality.
When the technology first hit the market, selling megapixels often meant megabucks — until end users realized they weren’t getting the image usability they thought they would. That led to one of two conclusions: the camera wasn’t as good as advertised or the integrator didn’t know what they were doing. Both conclusions reflect poorly on the integrator and long-term partnership.
Just ahead, we’ll examine the importance not selling solutions solely based on pixels. Rather, it’s all about defining operational requirements first as the basis for successful implementation of video surveillance. Doing so will not only earn goodwill but, even more importantly, build a good reputation by knowing when not to recommend the megapixel camera.
Defining Image Usability
Image usability can be defined as meeting the operational requirements of a scene, enabling the security professional to either respond or stand down. Successful video surveillance hinges on the delivery of imagery that is actionable — and that starts with proper camera selection.
Savvy integrators understand that the optimal resolution required for a scene is derived from the operational requirement the customer is trying to achieve. In other words, asking the question, “What do you need to get out of your surveillance system?” More specifically, this question could be asked for each and every camera in the system.
This goal-setting question is typically answered by one or more of the following objectives: detection (.5 pixels/inch); recognition (2.5); identification (12.5).
Simple detection of a person entering a room, for example, dictates a different set of criteria than being able to identify who that person is. In the case of needing to detect for a specific surveillance application, the customer would be better served with a standard resolution camera with all other factors being equal.
So then what criteria are needed to determine when megapixel cameras are a requirement?
It all comes down to the number of “pixels on target” needed to accomplish the primary intent: detection, recognition and/or identification. Each of these designations has a corresponding pixel requirement that refers to the minimum pixel density of an object or person based on the designation defined.
There is no industry standard for pixel density requirements, but rather a set of references and guidelines from various organizations. For the purposes of this article, we will use the Swedish National Laboratory of Forensic Science (SKL) reference, which is one of the most stringent by definition.
As expected, the most demanding designation is identification, which is defined by SKL as a width of 80 pixels across a person’s face. Though it may seem counterintuitive, you can actually cover more with less if the combination of camera selection, lens pairing and installation location is done properly. Customers can deploy fewer cameras with a wider field of view while still maintaining the appropriate number of pixels on target to satisfy the objectives of the scene.
In most cases, a standard definition camera will meet customer requirements when paired with the appropriate lens and camera placement when considering the size of the scene and distance from the camera. That being said, to attain the required pixels on target, a lower resolution camera will have a narrower field of view, which could compromise situational awareness. In the earlier example, the customer could identify who walked into a room but not be able to observe their actions when they moved out of the field of view once inside.
With standard resolution cameras, you either sacrifice detail by setting a wide field of view or lack situational awareness by narrowing it. When deployed correctly, security practitioners can deliver both with higher resolution cameras.
High Definition vs. Megapixel
The Society of Motion Picture and Telecommunications Engineers (SMPTE) created a high definition video quality standard commonly known as HDTV. The standard can also be referred to as 720p or 1,080p, which creates some level of confusion in the market as both directly refer to the vertical resolution delivered by the respective video stream. Remember, resolution alone does not equate to quality, which is why SMPTE included in its standard:
- Resolution (720p or 1,080p)
- 30 frames per second
- Aspect ratio of 16:9
- Broader palette to render higher color fidelity
A “megapixel camera,” on the other hand, only refers to the number of pixels produced by the image. When compared to megapixel cameras, 720p is equivalent to about 1 megapixel, while 1,080p is equivalent to 2 megapixels. That is where the similarities end for megapixel vs. HDTV.