Specifying Surveillance Video Quality
Ensuring high quality surveillance video can be difficult. The challenge starts with specifying what level of quality is needed. There are a number of specification methods but important tradeoffs exist between them. In this report, we are going to present a thorough approach to doing so. Undoudtedly, it is fairly time consuming but it provide the highest probability of delivering high quality surveillance that meets the expectations of surveillance users.
Two common approaches exist today in specifying surveillance video quality:
- Specify Resolution (e.g., 4CIF or 720p, etc.): This is the 'classic' way of specifying surveillance cameras. The plus side is that it is easy to list and to verify as resolution is prominently stated in almost all products. The downside is that it is extremely inexact and may be overkill for some scenes and provide terribly poor coverage in others. A slightly more sophisticated approach is to specify higher resolution cameras for scenes covering large areas (e.g., only 4CIF for entrance cameras but 3MP for parking lots, etc.). All in all, this is a quick and dirty way that can be roughly accurate.
- Specify Pixel Density (e.g., 40 pixels per foot (PPF), 120 pixels per meter (PPM), etc.): The 'new school' way of specifying surveillance cameras is to specify specific levels of pixel density - 40 pixels per foot for your entrance camera, 15 pixels per foot for your auditorium camera, etc. The benefit is that it is more exact than simply specifying resolution. The most immediate downside is the complexity or confusion about what pixel density is.
While we see specifications of pixel density (e.g., PPF or PPM), these specifications are almost always incomplete and therefore dangerous. It leads to a false sense of security yet omits crucial details that are vital to accurately achieving a system's video quality goals. Without these details, users are likely to still receive deficient quality and unmet expectations. Here are related attributes that must be specified to properly use pixel density:
- Lighting Conditions in Scene
- Horizontal Width of FoV Required
- Distance from Camera
- Vertical Range of Coverage
Pixel density is a very useful starting point but demands other related specifications. If you are just going to specify pixel density, you might as well just specify resolution. It will likely deliver the same quality levels but with less confusion. However, if you really want to accurately specify and obtain video quality matching your expectations a full specification of pixel density and the above 4 attriutes must be made. Inside, we provide videos and tutorials to better explain the issues involved and how to specify properly.
This report is the 3rd feature in our trilogy on video surveillance quality. For background, review our:
- Pixels Per Foot / Video Quality Report: Here we did a series of field tests to determine how much quality was delivered at various pixel density and lighting conditions. We show videos and screenshots so you can see the tradeoffs yourself.
- Surveillance Camera Specification Template: A comprehensive guide to specifying 27 aspects of IP cameras including surveillance video quality.
- Pixels Do Not Equal Image Detail
- Pixel Density Falls With Distance
- Pixel Density Declines Vary by Camera Angle
- Determining Pixels Per Foot Desired (using our Pixels Reference Chart)
- Determining Horizontal Field of View Width
- Determing Distance from Camera
- Determining Vertical Range to be Covered
- Matching Camera Characteristics (using our Camera Finder)
Our first screencast covers the fundamentals of using pixels including 3 key issues:
In Action - Putting it Together
The 30 minutes video screencast below shows you how to plan out specifications in common scenarios including entrances, rooms, retail and parking lots. We review: