What Is Optical Sensor Line Of Sight? (Cone FOV)
Optical sensor line of sight is the clear path between a sensor and the scene it must view. Its field of view (FOV) describes the angle and area the lens can capture. A target can sit inside that viewing cone yet lack enough image detail for the task. Check geometry, obstructions, image resolution, lighting, and privacy before fixing the camera in place.
Seasonal changes can alter a camera view even when the camera stays still. Leaves may block a path in summer, while low winter sun can cause glare or deep shadows. I treat the view as a part of the site that needs checking in more than one condition, not as a one-time aim-and-forget setting.
Plan the Sensor’s View: Scene, Purpose, and Privacy
Before choosing a camera position, define what the sensor needs to show and how much detail the task needs. A wide view may help detect movement across a large area, while a tighter view may be needed to observe or identify a person. The intended use also guides where the camera should not point.
Line of sight means an unobstructed optical path from the lens to the part of the scene you want to view. Field of view, or FOV, is the area captured through the lens, described by horizontal and vertical angles. The “cone” is a useful shorthand, but a camera’s actual view is often closer to a widening rectangular shape.
Start with a site sketch. Mark the sensor position, lens direction, target boundaries, and distance to the nearest and farthest points. Add likely obstructions such as walls, door frames, beams, signs, and foliage. Also mark windows or other areas where recording could intrude on privacy.
Then state the task in plain terms. For example: “Detect movement at the gate” is different from “recognize a face at the gate.” Do not assume that a wide-angle view can do both. A view can include the whole target but spread its pixels too thinly for useful detail.
Check Sightlines, FOV, and Physical Clearances
FOV geometry helps estimate how much scene a lens can cover at a given distance. Use the camera’s datasheet for its lens and sensor details, then compare the calculated view with your site sketch. Geometry is a planning check, not proof that an image will be clear or usable.
For a rectilinear lens, the angle for one sensor dimension is:
FOV = 2 × arctan(sensor dimension ÷ (2 × focal length))
Use the sensor’s width to find horizontal FOV and its height to find vertical FOV. For example, a sensor listed as 6.4 mm wide with a 4 mm lens has a horizontal FOV of about 77 degrees. Do not use an advertised diagonal FOV as though it were the horizontal angle.
At a distance d from the camera, the centered width or height of the view can be estimated with:
Coverage = 2 × d × tan(FOV ÷ 2)
For the example above, at 5 m, the horizontal span is about 7.7 m. This assumes a level, unobstructed view and a flat plane perpendicular to the lens axis. Angled walls, lens distortion, camera tilt, and the shape of the scene can change the result.
A wall or opaque finish blocks the view behind it, even if that area falls inside the calculated FOV. A mullion can split a view, and a door can block it when opened. Glazing also needs a real test: reflections, tint, coatings, dirt, and infrared-blocking glass can reduce or eliminate useful imagery, especially when the camera’s IR lights are on.
| Check | Planning method | What it does not prove |
|---|---|---|
| Horizontal and vertical FOV | Use sensor width and height with focal length, or the maker’s stated angles | That the advertised diagonal angle matches either axis |
| Scene coverage | Calculate span at the target distance | That the target has enough pixels for the task |
| Sightline | Trace the path from lens to each target point | That glass or changing conditions will not degrade the image |
| Pixel density | Divide horizontal image pixels by horizontal scene width | That every point has equal detail or that recognition is guaranteed |
For pixel density, a simple estimate is horizontal image pixels ÷ horizontal scene width in metres. A 1920-pixel-wide image spread across a 7.7 m span gives about 249 pixels per metre at that plane. The estimate assumes the full image width covers that span; real performance can vary with lens design, settings, motion, focus, and light.
Understand Image Detail and DORI Guidance
DORI is a set of common video-surveillance performance levels: detection, observation, recognition, and identification. IEC 62676-4 provides application and performance guidance that is commonly expressed using target pixel densities. These are guidance values, not legal minimums or guarantees of a result.
The commonly stated DORI values are:
| Task | Common guidance | Plain meaning |
|---|---|---|
| Detection | 25 px/m | Notice that a person or object is present |
| Observation | 62.5 px/m | See activity or broad features |
| Recognition | 125 px/m | Distinguish a familiar person or object |
| Identification | 250 px/m | Capture detail intended to support identification |
Check the applicable edition of IEC 62676-4 and the camera’s settings before relying on a threshold. Image density alone cannot guarantee a particular outcome. Focus, compression, shutter speed, motion blur, lighting, and the subject’s angle all matter. A person moving quickly in low light may be harder to see than a still target at the same estimated density.
To use the figures for a rough planning check, divide the camera’s horizontal pixel count by the desired density. A 1920-pixel-wide image at 125 px/m covers about 15.4 m across that plane in theory. For 250 px/m, the span is about 7.7 m. This does not mean the camera will deliver that level of detail everywhere; test the actual view.
Mount and Aim the Sensor Without Compromising the Assembly
Mounting height and direction shape the useful view as much as lens choice. Follow the manufacturer’s instructions and product listing, and do not cut or weaken structural members, damage a rated assembly, block required access, or use an unapproved wiring method. There is no universal mounting clearance or camera angle that fits every installation.
I plan a mounting point only after checking the full path from lens to target. In a hallway, for instance, a door closer or open door may block a low-mounted camera. In an outdoor view, a new branch may enter the scene as plants grow. A temporary test position helps reveal these issues before permanent holes are made.
Use the following sequence:
- Survey the sensor location, lens axis, target edges, distances, lighting, and likely obstructions.
- Calculate horizontal and vertical coverage from the camera’s datasheet and the site dimensions.
- Check pixel density at the farthest point that needs detail, not just at the nearest edge.
- Aim the camera at the intended scene and keep privacy-sensitive areas out of view where possible.
- Install it as directed, protecting structural elements, rated assemblies, access routes, and listed wiring.
- Verify the image in place before closing any finish or treating the position as final.
For powered equipment, follow the product instructions and locally adopted electrical and fire codes. Building codes do not set one generic camera FOV or mounting clearance. Requirements depend on the product, the assembly, and local rules, so check with the authority or a qualified trade professional where the installation affects regulated work.
Verify Coverage, Image Detail, and Operating Conditions
A calculated view is a starting point. The final check is the live image from the installed position, tested at the near and far edges of the target. Confirm that the scene is visible, that detail suits the stated task, and that reflections, glare, or changing light do not defeat the view.
I use two practical worked examples to show why FOV and detail must be checked separately. These are planning scenarios, not claims about a specific installed product.
Wide indoor view: A 6.4 mm-wide sensor with a 4 mm lens gives about 77 degrees of horizontal FOV. At 8 m, its idealized horizontal span is roughly 12.8 m. With 1920 horizontal pixels, the simple estimate is about 150 px/m at that plane. That is above the common recognition guidance value, but below the common identification value, before accounting for real-world image losses.
Tighter target view: With the same sensor and a 4 mm lens, a target 5 m away falls within an idealized span of about 7.7 m. A 1920-pixel image across that span yields roughly 250 px/m by the simple calculation. This is near the common identification guidance value, but it still does not guarantee identification. Test focus, movement, light, and the target’s position in the image.
During verification, check:
- Near and far limits of the intended area, including corners and edges.
- Daylight and the expected night or low-light mode.
- Glare, shadows, reflections, and any IR illumination.
- Whether doors, people, or seasonal growth can block key sightlines.
- Whether the displayed detail matches the task, rather than just the camera’s overall resolution.
- Privacy masks and the final aim, then document the settings and view.
If glass lies in the optical path, test it with the actual camera, lighting, and viewing angle. A window that looks clear to the eye may reflect the camera’s own IR light or reduce the image through its tint or coating. Do not count the view as usable until the live image confirms it.
Frequently Asked Questions
These short answers clarify the difference between line of sight, FOV, and image detail. Use them as a final planning check, then confirm the camera’s own specifications and the installed image.
What does optical sensor line of sight mean?
It is the clear path from the lens to the scene. An opaque obstacle blocks the view behind it.
Is FOV the same as line of sight?
No. FOV is the lens’s angular viewing range. Line of sight describes whether anything blocks the path to a point within that range.
Is a camera’s FOV cone a true cone?
Not always. “Cone” is a simple description; many cameras produce a widening rectangular view based on horizontal and vertical angles.
How do I calculate horizontal FOV?
Use the sensor width and focal length: 2 × arctan(sensor width ÷ (2 × focal length)). Confirm the sensor dimensions and lens details in the datasheet.
Does a target inside the FOV have enough detail?
Not necessarily. Check pixel density at the target distance and test the live image for focus, lighting, motion, and compression effects.
Can a camera see through glass?
Sometimes, but do not assume the image will be usable. Reflections, tint, coatings, dirt, and IR behavior can degrade it. Test the actual setup.
Does IEC 62676-4 set a legal minimum FOV?
No generic FOV follows from the DORI guidance. Check the applicable standard edition, project needs, product documentation, and local rules.
How should I confirm the final view?
Test the near and far edges in expected lighting, check obstructions and image detail, review privacy masks, and record the final aim and settings.
(This article was written by one of our staff writers, Garrett Vance. Visit our Meet the Team page.)
