Lens Selection and Coverage Geometry Guide
Megapixels do not create evidence. Geometry does. Lens choice, mounting height, and target distance determine whether a scene produces identification-grade detail, usable analytics, and predictable retention. This guide shows how to engineer lens decisions using repeatable coverage math, practical pixel density targets, and deployment patterns that hold up during real incidents.
Why Lens Geometry Beats Feature Lists
Evidence is a math problem
Face and object detail are primarily a function of focal length, distance, and framing stability. If the subject occupies too few pixels, no setting or codec will recover it.
Wide lenses create false confidence
A wide lens makes a space look covered, but identification collapses quickly with distance. Many systems fail at entrances because the lens is optimized for “seeing the door” instead of identifying the person.
Analytics needs stable scale
Detection and classification are sensitive to subject size in frame, angle, and exposure stability. The same analytics can be excellent on one camera and unusable on another because geometry differs.
Storage is downstream of geometry
Bad geometry often leads to over-recording, over-resolution, or over-camera-count to compensate. Correct lens intent reduces waste while improving outcomes.
Lens and Pixel Density Quick Calculator
This is a planning estimator that approximates horizontal scene coverage and pixel density at the target distance. It helps you catch the common failure mode: a lens that is too wide for the evidence objective. Validate with a field test for your exact camera sensor size and scene lighting.
Lens Roles by Zone
The fastest way to reduce design variance is to standardize lens roles. Most commercial systems need a small set of repeatable roles rather than random focal lengths per camera.
Entrance identification camera
- Purpose: reliable faces under motion
- Mounting: 8–10ft preferred
- Lens: 6mm to 12mm depending on depth
- Key risk: backlight and vestibule lighting
General overview camera
- Purpose: situational awareness and event context
- Mounting: 10–14ft typical
- Lens: 2.8mm to 4.0mm
- Key risk: trying to use overview for ID zones
Corridor and choke-point camera
- Purpose: separation, directionality, usable scale
- Mounting: depends on corridor height and angle
- Lens: 4mm to 8mm, corridor mode where available
- Key risk: wide lens makes people small and blends movement
Perimeter lane and parking strategy
- Purpose: capture movement and vehicles with usable separation
- Mounting: avoid extreme overhead angles when possible
- Lens: 6mm+ often required for lanes and longer standoff
- Key risk: “one wide shot” that cannot support investigation
Standardize the lens roles, then scale
Multi-site programs fail when each location invents its own focal length decisions. If you standardize roles (Entrance ID, Overview, Corridor, Perimeter Lane) you reduce installer variance and simplify troubleshooting.
Mounting Height Rules That Prevent Evidence Failure
8–10 feet
Best zone for entrances and identification cameras. Preserves face angle, reduces distortion, and improves motion reliability.
10–14 feet
Typical for general coverage in commercial environments. Pair with correct lens intent so overview cameras do not become accidental ID cameras.
12+ feet for ID zones
Treat as high risk. If you cannot lower height, you often need a tighter lens and a dedicated ID shot at a controlled distance.
Fast self-check that catches most bad systems
- If more than 50% of cameras are over 12ft, expect ID failure in key zones unless compensated by lens strategy.
- If entrances are covered by wide lenses, assume face detail collapses under motion.
- If lighting changes rapidly (glass doors, headlights, vestibules), treat exposure stability as part of the design, not an afterthought.
Process Diagram: The Correct Lens Decision Sequence
This sequence prevents the most common “looks covered, fails in court” outcome. The key is to define evidence intent before lens choice.
Lens Decisions Connect to the Entire Stack
Retention and storage sizing
Over-wide coverage often leads to “more cameras at higher resolution” to compensate, which expands storage cost and increases retention risk.
Network and PoE planning
Camera count and resolution drive bandwidth and PoE budgets. Correct lens roles often reduce unnecessary camera count while improving evidence outcomes.
Recording platform selection
Lens strategy affects motion behavior, bitrate patterns, and how searchable the footage is. Match recording platform capability to the evidence plan.
Analytics performance
Analytics accuracy is strongly correlated with subject scale and stable exposure. If the lens is too wide or the angle too steep, analytics becomes noise.
Camera Categories Where Lens Choice Matters Most
These product classes are often deployed incorrectly without geometry planning. Use the calculator above and validate in the field for the target distance and lighting behavior.
Panoramic cameras
Great for overview. High risk for ID claims unless you add dedicated ID shots.
PTZ cameras
Best as an operator-assist layer, not a replacement for fixed ID zones.
LPR cameras
Requires correct angle, lane control, and exposure tuning. Not “zoom and hope.”
Thermal cameras
Excellent detection. Not identification. Pair with visible-light ID cameras.
People counting cameras
Height and angle must match the counting model. Geometry errors cause drift.
Radar sensors
Great for reliable triggers and classification support, especially outdoors.
Lens and Coverage Geometry FAQ
Wider lenses increase field of view but reduce subject scale rapidly with distance. For evidence-driven zones, use dedicated ID shots instead of trying to make one wide camera do everything.
Frequently Asked Questions
How do I choose the right focal length for a camera?
Focal length determines field of view (FOV) and effective coverage distance. Use 2.8mm for wide overview of rooms 10-25 feet wide. Use 4mm for general-purpose coverage at 20-40 feet. Use 6mm for hallways and narrow areas at 30-60 feet. Use 8-12mm for parking lot perimeters at 40-100 feet. Use 25-50mm for long-range identification at 150-500 feet. Varifocal lenses (2.8-12mm or 4-40mm) let you tune focal length on-site or remotely. Always calculate pixel density at the target distance rather than just choosing a lens by feel.
What's the formula for calculating field of view?
Horizontal FOV = 2 x arctan(sensor width / (2 x focal length)). For a 1/2.7-inch sensor (5.37mm wide) with a 4mm lens: FOV = 2 x arctan(5.37 / 8) = 67 degrees. At 20 feet distance: scene width = 2 x 20 x tan(33.5) = 26.5 feet. Most camera vendors publish FOV at common focal lengths in their spec sheets. Use IPVM, JVSG, or vendor-specific lens calculators to get accurate FOV, coverage, and pixel density for each camera model and scene geometry.
How do I figure out how many cameras I need?
Map your facility to scale in PowerPoint, CAD, or IPVM Design. Identify coverage goals (detection, recognition, or identification) at each location. Draw each camera's FOV cone and ensure no blind spots for critical zones. Typical density: one camera per 500-1,500 square feet for general coverage, one per 100-300 square feet for high-value retail or high-security areas. Add 10-20% more cameras than the bare minimum to cover future changes in layout. Document each camera's purpose, position, and expected coverage before ordering hardware.
What mounting height is best for IP cameras?
8-10 feet high gives best facial identification balance (subject faces are visible but camera is out of reach). 12-15 feet high works for parking lots and large open areas where overview matters more than face identification. Above 15 feet, facial details drop sharply and pixel density on target falls unless you use long focal lengths. Below 8 feet, cameras become reachable for tampering. For high-vandalism areas, use IK10 vandal-rated cameras at 10-12 feet to deter tampering while maintaining identification angle.
How do I avoid camera blind spots?
Use overlapping FOV: each camera should see into the next camera's FOV so no gap exists along critical paths (entrances, corridors, perimeter lines). For 360-degree corner coverage, use 2 cameras at right angles to eliminate the 'over the shoulder' blind spot. For long straight corridors, use one camera facing each direction rather than one in the middle. Ceiling-mounted fisheye or multi-sensor cameras cover blind spots that would otherwise require 2-4 fixed cameras. Walk each area with a test camera before final installation to verify coverage.
What's the difference between detection, recognition, and identification distances?
Detection (can see something there): requires 25 pixels per foot on target (ppf). Observation (see what they're doing): 63 ppf. Recognition (recognize a known person): 125 ppf. Identification (identify a stranger from video): 250 ppf. These increase geometrically, so identification requires 10x the pixel density of detection. Plan your camera layout around these standards per use case: perimeter detection needs far fewer pixels than cash counter identification. Use scene calculators to verify pixel density before ordering cameras.
Want a lens plan that produces consistent evidence?
Share site type, mounting constraints, priority zones, and target distances. We will map lens roles, validate pixel density targets, and connect the plan to retention and network requirements.