Technical10 min read

    DORI in CCTV

    Master the four DORI levels to design CCTV systems that meet your security objectives. From detection to identification, understand what each level requires.

    What is DORI?

    DORI is an acronym for Detection, Observation, Recognition, and Identification—the four progressive levels of visual clarity in CCTV systems. Each level defines how well you can see details of people, objects, and activities in a monitored area.

    Choosing the right DORI level for each camera is crucial. Over-specifying wastes money; under-specifying fails to meet your security objectives. Understanding DORI helps you find the balance.

    Detection Level

    Detection is the basic level: you can see that something or someone is present, but nothing more. You know there's movement, but you can't determine what it is.

    Detection Requirements:

    • Pixel size: 25+ pixels per meter
    • What you see: Motion, shapes, presence
    • Typical use: Wide-area surveillance, perimeter
    • Camera type: Standard 2MP camera

    Observation Level

    Observation provides more detail. You can determine basic characteristics: is it a person or a vehicle? What color are they? What's their general direction of movement?

    Observation Requirements:

    • Pixel size: 62.5+ pixels per meter
    • What you see: Type, color, direction
    • Typical use: Parking lots, building entrances
    • Camera type: 4-5MP camera

    Recognition Level

    Recognition lets you identify specific individuals or objects if you already know them. You can recognize your employee across the parking lot, or identify a specific vehicle.

    Recognition Requirements:

    • Pixel size: 125+ pixels per meter
    • What you see: Facial features, clothing details
    • Typical use: Retail stores, entrances
    • Camera type: 8-10MP camera

    Identification Level

    Identification is the highest level. You can identify individuals or objects you've never seen before. Facial features are clear enough to identify someone, or license plates are fully legible.

    Identification Requirements:

    • Pixel size: 250+ pixels per meter
    • What you see: Facial details, license plate characters
    • Typical use: License plate capture, facial recognition
    • Camera type: 12MP+ specialized camera

    DORI Comparison Table

    LevelResolutionTypical DistanceUse Case
    Detection2MP40+ metersPerimeter, wide areas
    Observation4-5MP20-40 metersParking, entrances
    Recognition8-10MP10-20 metersRetail, entrances
    Identification12MP+5-10 metersLicense plates, faces

    © 2026 CCTVplanner. All rights reserved.