CCTV Field of View: Focal Length, Sensor Size, Height and Tilt
The field of view of a camera decides how much of a scene it can see. Together with its resolution, it decides how much detail each part of the scene gets. This guide shows how sensor size and focal length set the horizontal and vertical field of view, how that becomes scene width and pixel density at a given distance, and what camera height and tilt change. One worked example runs through all of it, and you can open the same example in the calculator.
Open the worked example in the calculator →
- A shorter focal length gives a wider field of view. A larger sensor gives a wider field of view for the same lens.
- Resolution does not change the field of view. It changes how many pixels cover the scene.
- Pixel density at a distance is the horizontal pixels divided by the scene width at that distance.
- Camera height and tilt do not change the angles. They decide where the camera sees the floor.
1. What field of view is
The horizontal field of view (HFOV) is the angle between the left and right edges of the image. The vertical field of view (VFOV) is the angle between the top and bottom edges. Both are measured at the camera, so on their own they say nothing about how large an area is covered. That depends on the distance: a camera with a 68° horizontal view sees a scene roughly 20 m wide at 15 m and roughly 40 m wide at 30 m.
2. The two formulas
Sensor width and focal length are in millimetres. The active sensor height is the sensor height cropped to the aspect ratio of the recorded image. A 4:3 sensor recording 16:9 uses its full width and part of its height, so the VFOV is smaller than the full sensor height would give. The HFOV always uses the full sensor width.
| Format | Width (mm) | Height (mm) |
|---|---|---|
| 1/4" | 3.2 | 2.4 |
| 1/3" | 4.8 | 3.6 |
| 1/2.9" | 5.12 | 3.84 |
| 1/2.8" | 5.37 | 4.02 |
| 1/2.7" | 5.37 | 4.02 |
| 1/2.5" | 5.7 | 4.28 |
| 1/2" | 6.4 | 4.8 |
| 1/1.8" | 7.18 | 5.32 |
| 1/1.7" | 7.6 | 5.7 |
| 1/1.2" | 10.67 | 8 |
These are the generic sizes the CCTV Design Tool designer uses. The format name is a legacy designation, not a measurement, so two sensors with the same name can differ by a few tenths of a millimetre. Use the width and height from the camera datasheet when you have them.
3. From angle to scene width and pixel density
Distance is measured along the direction the camera points, to a flat target facing it. Pixel density is in pixels per metre (px/m): the higher it is, the more detail each metre of the scene gets. The DORI levels (Detect, Observe, Recognize, Identify) are defined by the pixel density a task needs. The DORI and pixel-density guide explains which level suits which task.
4. Worked example
A 4 MP camera (2560 × 1440) with a 1/2.8" sensor (5.37 × 4.02 mm) and a 4 mm lens, aimed at a target 15 m away. It is mounted 4.5 m high and tilted 15° below horizontal.
| Result | Value |
|---|---|
| Horizontal field of view | 67.7° |
| Vertical field of view | 41.4° |
| Scene width at 15 m | 20.1 m |
| Scene height at 15 m | 11.3 m |
| Pixel density at 15 m | 127 px/m |
| Level reached | Recognize |
Working the width through by hand: HFOV = 2 · arctan(5.37 / 8) = 67.7°, so half of it is 33.9° and the scene width is 2 · 15 · tan(33.9°) = 20.1 m. The pixel density is 2560 / 20.1 = 127 px/m. That clears the 125 px/m needed for Recognize, but only just: Recognize is reached up to 15.3 m and lost beyond it. Identify (250 px/m) is reached only up to 7.63 m.
5. Changing the lens or the resolution
| Change from the example | HFOV | Scene width at 15 m | Pixel density | Level |
|---|---|---|---|---|
| None (4 mm, 4 MP) | 67.7° | 20.1 m | 127 px/m | Recognize |
| 2.8 mm lens | 87.6° | 28.8 m | 89.0 px/m | Observe |
| 6 mm lens | 48.2° | 13.4 m | 191 px/m | Recognize |
| 8 MP (3840 × 2160), same 4 mm lens | 67.7° | 20.1 m | 191 px/m | Recognize |
The 8 MP row shows why resolution does not change the field of view: the scene is exactly as wide, so the extra pixels only raise the density. The aspect ratio can change the vertical field of view. The same sensor recording 4:3 (2592 × 1944) keeps the 67.7° horizontal view but sees 53.4° vertically instead of 41.4°.
6. Camera height and tilt
Tilt is measured below horizontal. The upper and lower edges of the image leave the camera at the tilt angle minus and plus half the VFOV. For the example, half the VFOV is 20.7°, so the upper edge points 5.7° above the horizon (15° − 20.7°) and the lower edge points 35.7° below horizontal (15° + 20.7°).
The lower edge meets the floor 4.5 / tan(35.7°) = 6.27 m from the camera, so the floor closer than that is not visible. That is the near blind zone. Because the upper edge points above the horizon, there is no far limit. Tilting further down, so that the upper edge points below the horizon, gives a far limit: at 8 m high and 30° down the floor is visible from 6.55 m to 48.8 m.
The direction the camera points also meets the floor. In the example it does so 17.4 m along that direction. A target further along than that would be below the floor, so the calculator warns when a combination of height, tilt and distance puts the target underground. For example, at 3 m high and tilted 25°, the direction meets the floor after 7.10 m, and a target 15 m along it does not exist.
7. Where a flat calculation stops being accurate
- The target is a flat surface facing the camera. A camera looking down at the floor sees a trapezoid, not a rectangle, and the pixel density on the floor changes with distance and angle. The designer works this out for the ground, a vertical plane or a standing person.
- Sensor sizes are nominal. Use the datasheet dimensions.
- Wide lenses distort. The pinhole formula overstates what a wide lens covers near the image edges. The calculator warns above 120° horizontal.
- Real lenses are usually less sharp toward the edges of the image. A pixel count does not show that.
- Digital zoom, stabilisation and cropping in the camera reduce the field of view that is actually recorded.
- Obstacles are ignored. Walls, glass and furniture block the view. The blind-spot guide covers this, and the designer models walls and openings.
- Lighting, motion and compression decide whether a person can be recognised. Pixel density is a planning figure, not a guarantee.
8. Choosing a lens for a coverage requirement
Work backwards. Suppose the scene must be 20 m wide at 15 m, on the same sensor. The horizontal field of view needed is:
A 4 mm lens is the closest common choice, which is why the example uses it. Check the choice by entering the lens in the calculator and reading the pixel density at the far end of the area, then compare it with the level the task needs.
Assumptions and sources
- A pinhole (rectilinear) lens, the generic sensor sizes above, metres and degrees.
- The DORI levels are those published in EN 62676-4 guidance; the standard's own text is not reproduced here. The methodology page lists the values the software uses and the status of IEC 62676-4:2025.
- Results are planning estimates and are not a statement of compliance with any standard.