Why Megapixels Lie: The Truth About Camera Resolution and Forensic Value

The surveillance camera industry has spent the last decade in a megapixel arms race, and it has done a disservice to everyone who buys cameras for actual security purposes. Marketing departments plaster "4K" and "32MP" on every box, and end users have been conditioned to believe that more megapixels automatically means better security footage. It does not. A 2-megapixel camera with a properly selected lens, adequate sensor size, and appropriate compression settings will produce more forensically useful video than a 12-megapixel camera with a cheap varifocal lens, an undersized sensor, and H.265 encoding cranked to minimum bitrate.

This article explains the metrics that actually determine whether your camera system can identify a suspect, read a license plate, or provide evidence that holds up in court. We will cover pixels per foot, the DORI standard, sensor physics, compression tradeoffs, and the bandwidth and storage implications of resolution choices. If you specify or install cameras, this is the technical foundation you need to stop chasing megapixels and start designing for forensic value.

Pixels Per Foot: The Metric That Actually Matters

Pixels per foot (PPF), also expressed as pixels per meter (PPM), is the measure of how many pixels cover a given linear distance in the scene at the target area of interest. It is the single most important specification for determining whether a camera can perform a given identification task. A camera's megapixel count tells you the total number of pixels on the sensor. PPF tells you how those pixels are distributed across the scene you care about.

A 4K (8.3MP) camera with a wide-angle lens covering a 200-foot-wide parking lot delivers approximately 19 pixels per foot horizontally (3840 pixels / 200 feet). That is barely enough to detect that a person is present. The same 4K camera with a narrower lens covering a 40-foot-wide entrance delivers 96 pixels per foot, which is enough to recognize a face under good lighting conditions. The megapixel count did not change. The PPF changed by a factor of 5, and with it, the forensic utility of the image.

Calculating PPF is straightforward: PPF = Horizontal Resolution / Horizontal Field of View (in feet). For vertical PPF, use vertical resolution divided by vertical field of view. The number you care about depends on the task: detecting a person requires far fewer PPF than reading a name badge or matching a face to a photo ID.

The DORI Standard: Detection, Observation, Recognition, Identification

The DORI standard, codified in EN 62676-4 (Video surveillance systems for use in security applications), provides an objective framework for classifying camera performance based on the operator task. It replaces subjective descriptions like "good image quality" with measurable PPF thresholds.

Designing a camera system without PPF targets for each camera position is like designing a lighting system without lux targets. You might get lucky, but you cannot demonstrate that the system meets any defined performance criterion. Every camera in a professionally designed system should have a documented PPF value at the target area of interest, and that value should correspond to the DORI level required for that camera's assigned task.

Sensor Size: Why It Outweighs Megapixels

A camera sensor converts photons into electrical signals. Larger individual pixels collect more photons, which produces a cleaner signal with less electronic noise. This is why sensor size matters more than megapixel count, particularly for security cameras that must perform 24/7 including nighttime hours when light levels may drop below 1 lux.

Consider two cameras, both 4MP. Camera A has a 1/3-inch sensor. Camera B has a 1/1.8-inch sensor. Camera B's sensor has approximately 3 times the surface area, which means each pixel is roughly 3 times larger. In low light, Camera B produces a dramatically cleaner image with less noise, better color accuracy, and higher contrast. Camera A may technically have the same resolution, but the image degrades so severely in low light that its effective forensic resolution is far lower.

The premium 1-inch sensor cameras (Sony Starvis 2 and similar) can produce usable color images at 0.001 lux, which is near-total darkness. A 1/2.8-inch sensor camera, the most common in the industry, typically needs 0.01-0.05 lux for acceptable color and switches to black-and-white with IR illumination below that. For critical identification cameras at building entrances, the sensor size premium pays for itself the first time you need to pull a color image of a suspect at 2 AM.

Compression: Where Forensic Detail Goes to Die

Video compression is a necessary evil. Uncompressed 4K video at 30 fps generates approximately 1.5 Gbps, which is impractical for recording and network transport. H.264, H.265 (HEVC), and the emerging H.266 (VVC) codecs reduce this to manageable bitrates by discarding visual information that the codec's algorithm deems redundant. The problem is that the information the codec discards may be exactly the forensic detail you need.

H.265 achieves approximately 50% bitrate reduction compared to H.264 at equivalent visual quality. H.266 promises another 30-50% reduction over H.265. But "equivalent visual quality" is defined by perceptual metrics (SSIM, PSNR) that optimize for how the image looks to a human viewer at normal viewing distance, not for whether a forensic analyst can read a license plate number or identify a facial feature in a zoomed-in crop.

Aggressive compression manifests in several forensically destructive ways. Macroblocking turns faces into pixelated grids during motion. Smearing eliminates fine text on clothing, badges, or tattoos. Color banding reduces the tonal range needed to distinguish skin tones. Temporal artifacts cause moving objects to leave ghost trails that obscure adjacent details. Every one of these artifacts gets worse as you reduce the bitrate, and every camera manufacturer's default setting prioritizes storage savings over forensic fidelity.

Wide Dynamic Range: The Overlooked Specification

Wide Dynamic Range (WDR) is the camera's ability to simultaneously capture detail in both the brightest and darkest areas of a scene. It is measured in decibels (dB), with true WDR cameras using multi-exposure sensor technology to achieve 120-150 dB of dynamic range compared to 60-70 dB for standard sensors.

WDR is critical for any camera facing a door, window, or vehicle entrance where backlight is a factor. Without adequate WDR, a person walking through a sunlit doorway appears as a black silhouette with no facial detail whatsoever. All the megapixels and PPF in the world are worthless if the subject's face is a featureless shadow. For entrance/exit cameras and any position with mixed lighting conditions, specify a minimum of 120 dB true WDR. Do not confuse true multi-exposure WDR with digital WDR (DWDR), which is a software-only processing trick with far inferior results.

Bandwidth and Storage: The Hidden Cost of Resolution

Higher resolution cameras generate exponentially more data. A 2MP camera at H.265 high quality produces approximately 2-4 Mbps. A 4K (8MP) camera at equivalent quality settings produces 8-16 Mbps. An increase from 2MP to 8MP quadruples both the network bandwidth requirement and the storage consumption. For a 64-camera system retaining 30 days of continuous recording, the difference between 2MP and 8MP across all cameras can be 50-100 TB of additional storage, which translates to thousands of dollars in additional NVR capacity and ongoing disk replacement costs.

The engineering approach is to right-size resolution for each camera's task. Perimeter detection cameras covering large areas might need 4K resolution to achieve adequate PPF at distance. Interior hallway cameras at 15-foot width may achieve Recognition-level PPF with just 2MP. Lobby identification cameras may need 4MP with a narrow field of view. Designing every camera at maximum resolution because "more is better" wastes bandwidth, fills storage, and does not improve forensic outcomes.

Conclusion

Forensic video quality is the product of PPF, sensor performance, lens quality, WDR capability, compression settings, and bitrate allocation. Megapixel count is one input to the PPF calculation and nothing more. A camera system designed around DORI performance levels, with right-sized resolution, quality lenses, appropriate sensors, and properly configured encoding, will outperform a system with double the megapixels and none of the engineering discipline.

Zimy Electronics designs camera systems using PPF-based analysis for every camera position, ensuring that each camera delivers the forensic performance level its assigned task requires. We specify sensor sizes, lens focal lengths, WDR ratings, and compression parameters based on engineering calculations, not marketing specifications. From retail loss prevention to critical infrastructure protection, our designs deliver video that is genuinely useful when you need it most: during an investigation.