Testing Wide Vs Narrow Camera FoVBy: John Honovich, Published on Nov 24, 2011
Determining the correct Field of View (FoV) for each surveillance camera is one of the most important decisions in designing surveillance systems. It can also be one of the most challenging as a fundamental tension exists: People prefer wide FoVs because the wider the Field of View the more they can see. However, the wider the FoV, the less details they can actually make out in any given area of the scene. While megapixel has improved this situation, significant limits exists in how wide and how far one can see simultaneously.
Testing Wide vs Narrow FoV
One of the most common outdoor use cases is to monitor the perimeter of a building and the parking lot adjacent to the building. Ideally, it would be great to accomplish this with a single camera that captures the details of a person coming up to a building while at the same time capturing what is going on in the parking lot.
The images below provide examples of two different Field of Views. The first, the 'wide' one captures both the sidewalk/entrance to the building and a large portion of the parking lot. The second, the 'narrow' one focuses on the sidewalk/entrance with only a small portion of the distant parking lot.
Both pictures are from the same camera. The only thing that has changed is the optical zoom.
Here is another view contrasting the setup of the two shots, providing another look at the tradeoffs in the scene captured.
In this test, we were primarily interested in seeing the image detail tradeoffs on the sidewalk adjacent to the building. The distance from camera to the doorway where our subject stands is approximately 80 feet. The image below shows that the narrow FoV shot provides far more details than the wide FoV. You can almost make out the face of the subject in the narrow shot while the wide shot has no chance of doing so.
As the subject moves closer to the camera, the practical benefits of the narrow FoV setup decreases. In the image below, at ~45 away from the camera, the 'narrow' FoV is clearly still better but even the 'wide' FoV provides sufficient details of the subject's face.
Finally, at ~12 feet from the camera, the benefits flip. While both cameras provide very clear images of the subject's face, the 'wide' FoV still shows the complete subject while the subject is nearly exiting the Field of View of the narrow setup.
Other Important Factors
While the two most obvious factors are the (1) ability to capture details of a subject at the perimeter and (2) the ability to capture the parking lot, other important factors do exist:
- Ability to handle Bright Sunlight: When we did this test, it was noontime and the sun was moving from left to right in this scene coming across the building. This created a very challenging lighting situation. We wound up only using the premium Sony CH140 camera as the two Axis P series (1344 and 1347) performed badly. This is consistent with our WDR megapixel shootout results. Bottom line, regardless of FoV, lighting issues can play havoc with PPF calculations or projections.
- Night time performance: Night time performance will always be worse. The only question is how bad night time performance will be. This is highly dependent on the level of illumination and how close lighting is to the camera. Any day time results such as presented here are best case scenarios.
- Higher Resolution: This test used a 720p/1.3MP camera. Higher resolutions could offset the tradeoff by allowing a wider FoV and the same captured details as a lower resolution. However, we regularly see two major tradeoffs as noted above: Both bright sunlight and lower lighting reduce or eliminate the advantages of higher resolution cameras. As this is an outdoor, those factors are going to have a major impact regularly.
- Height of camera: In this test, we were primarily interested in seeing the tradeoffs in monitoring a subject walking along the perimeter. Height was not a factor here. However, mounting the camera higher would have provided much greater usability for monitoring the parking lot, especially with the wide FoV.
Even with higher resolution, tough decisions need to be made about wide of an area that a camera can cover. Frequently, users and designers want to do too much with a single camera. This short test should help show some of the negative tradeoffs of going too wide.