DORI Explained
DORI for CCTV Applications, Performance-Based CCTV Design
Executive Summary
Video surveillance performance should be specified by operational outcome, not by camera resolution alone. The legacy DORI model, formalized in EN/IEC 62676-4, gave the industry a practical way to connect surveillance objectives to measurable pixel density thresholds, allowing designers to predict whether a scene would support detection, observation, recognition, or identification (CCTV Design Tool Team, n.d.).
The 2025 revision of IEC 62676-4 expands that older model into a broader visual-performance framework with seven operational levels: overview, outline, discern, perceive, characterize, validate, and scrutinize. This update reflects higher expectations for evidentiary quality, facial comparison, analytics-driven workflows, and modern IP surveillance deployments (Axis Communications, 2025; JVSG, 2025).
For end users, consultants, and integrators, the practical implication is clear: camera design now requires more disciplined alignment between operational intent, pixel density, field of view, lighting, scene geometry, and post-event investigative needs. For Backstreet Surveillance, that shift favors a consulting-led approach that combines standards literacy, design simulation, operational interviewing, and field experience to reduce overdesign, underperformance, and avoidable change orders.
Introduction
For years, many surveillance projects were specified around broad statements such as “cover the parking lot” or “watch the gate.” In practice, those instructions were often too vague to guarantee usable video evidence, because a camera can view an area without capturing sufficient detail for the actual task required.
DORI helped solve that problem by giving specifiers and designers a common language for image quality requirements. Instead of debating whether an image looked “good enough,” project teams could anchor camera placement and lens selection to standardized pixel-density thresholds tied to operator tasks (CCTV Design Tool Team, n.d.; Smaling, n.d.).
The newer IEC/EN 62676-4:2025 framework keeps the same performance-based philosophy but adds more granular operational levels and higher thresholds for certain use cases. That change matters for organizations that expect video to support not just awareness and deterrence, but investigation, validation, and evidentiary scrutiny in increasingly complex operational environments (Axis Communications, 2025; JVSG, 2025).
DORI Explained in Detail
DORI stands for Detect, Observe, Recognize, and Identify. In the EN 62676-4 framework, each level corresponds to a minimum image-detail threshold expressed in pixels per meter, giving a measurable basis for surveillance design instead of relying on subjective judgment (CCTV Design Tool Team, n.d.; Smaling, n.d.).
Detect
Detection is the lowest practical DORI threshold and is generally associated with approximately 25 px/m. At this level, an operator can determine that a person or object is present, but cannot rely on the image for meaningful distinguishing detail (CCTV Design Tool Team, n.d.; CCTV Planner, 2026).
This level is appropriate for perimeter awareness, motion verification, and broad scene coverage where the primary objective is to know that an event is occurring. It is not sufficient for facial recognition, positive identification, or detailed behavioral interpretation.
Observe
Observation is typically defined at approximately 63 px/m. At this level, the image begins to support situational interpretation, including visible body posture, activity, clothing contrast, and other general characteristics (CCTV Design Tool Team, n.d.; JVSG, 2019).
Observation is often suitable for entrances, loading zones, and circulation areas where the operational question is not “Who exactly is this?” but rather “What is the person doing?” or “What general attributes can be seen?” This distinction is critical because many systems are mistakenly expected to identify individuals when they were only designed to observe them.
Recognize
Recognition is generally associated with approximately 125 px/m. At this threshold, viewers can determine with a high degree of certainty whether a person shown is the same person they have seen before, which supports familiarity-based recognition rather than courtroom-grade proof of identity (CCTV Design Tool Team, n.d.; JVSG, 2019).
This level is often useful for employee-only areas, school access points, and commercial facilities where staff may need to determine whether a person is known or unknown. Recognition can be operationally valuable, but it should not be confused with positive identification.
Identify
Identification is generally set at approximately 250 px/m. At this level, the image should support establishing identity beyond reasonable doubt through clear facial detail and other fine characteristics (CCTV Design Tool Team, n.d.; Axis Communications, n.d.).
Identification is the proper design target for high-risk portals, transaction points, custody-transfer positions, narrow choke points, and other places where evidence quality matters most. Because 250 px/m must be achieved on the target, not merely somewhere in the scene, true identification coverage usually demands disciplined camera placement, tighter fields of view, and careful control of mounting height, angle, lighting, and scene depth.
Why DORI Still Matters
DORI remains valuable because it gives owners, consultants, and integrators a common operating language for outcome-based design. It helps translate vague expectations into measurable requirements and allows camera, lens, and placement decisions to be made before installation rather than corrected later through expensive redesign (CCTV Design Tool Team, n.d.). DORI is also effective because it exposes the tradeoff between area coverage and forensic detail. A wide field of view may be useful for awareness, but it spreads horizontal pixels over a larger scene width, reducing pixel density at the target and making identification less likely at distance (CCTV Planner, 2026).
In practical terms, DORI prevents a common specification error: assuming that more megapixels alone guarantee better evidence. In reality, usable forensic performance depends on where the target is, how many pixels cover that target, and whether environmental conditions preserve those pixels in a meaningful image.
The New IEC/EN 62676-4:2025 Standard
The updated IEC/EN 62676-4:2025 standard expands the older DORI model into a seven-level visual-performance structure intended to better reflect contemporary surveillance tasks. According to industry summaries of the new edition, the operational levels are overview, outline, discern, perceive, characterize, validate, and scrutinize (Axis Communications, 2025; JVSG, 2025). Reported pixel-density thresholds for the seven levels are 20 px/m for overview, 40 px/m for outline, 80 px/m for discern, 125 px/m for perceive, 250 px/m for characterize, 500 px/m for validate, and 1500 px/m for scrutinize (Axis Communications, 2025). These thresholds indicate a shift toward finer-grained specification and materially higher expectations for advanced evidentiary and analytical use cases.
The standard revision is also described as responding to modern surveillance realities, including higher-resolution digital imaging, improved low-light performance, AI-enabled analytics, and broader cybersecurity expectations for networked video systems (Axis Communications, 2025). In other words, the new framework is not merely a renaming exercise; it reflects a different performance ceiling and a more mature view of how video is used operationally and investigatively.
DORI Versus the 2025 Framework
The older DORI structure remains intuitive and highly usable, particularly in early design conversations. Detect, observe, recognize, and identify are easy for end users to understand, and they map well to many common commercial and industrial requirements.
The 2025 framework adds specificity between the old task bands and introduces much more demanding upper tiers. In practice, perceive and characterize are broadly comparable to the upper middle and top end of legacy DORI expectations, while validate and scrutinize extend performance requirements far beyond traditional identification thresholds for use cases where stronger facial comparison or close evidentiary examination is needed (Axis Communications, 2025).
Performance comparison
The most important takeaway is that the new standard makes it harder to hide weak design assumptions behind generic language. A project team now has stronger grounds to ask whether the goal is basic scene awareness, characteristic description, known-person recognition, facial characterization, validation, or high-scrutiny evidence capture.
Design Implications for Real Projects
The movement from DORI to the 2025 visual-performance framework has immediate implications for design methodology. Surveillance projects should begin with operational interviewing that identifies specific decision points: what must be seen, by whom, at what distance, under what lighting conditions, and for what downstream use.
Once those questions are answered, the camera should be designed backward from the target task rather than forward from the camera spec sheet. That means selecting the sensor, lens, mounting geometry, and scene coverage necessary to achieve the required pixel density at the required point of action instead of assuming that a nominal resolution figure guarantees results (CCTV Planner, 2026; CCTV Design Tool Team, n.d.).
This is also where simulation tools become valuable. DORI and related visual-performance overlays on plans help reveal whether a field of view truly supports the intended task at the intended distance, reducing the risk that a camera provides only general awareness where positive identification was expected (CCTV Design Tool Team, n.d.).
Backstreet Surveillance Approach
Backstreet Surveillance is well positioned to help clients move from informal camera planning to standards-based, evidence-aware system design. A consulting-led approach matters because compliance with a standard does not come from quoting a camera resolution; it comes from aligning operational intent, scene geometry, and technical design to measurable performance outcomes. Backstreet’s design methodology can credibly emphasize several differentiators:
- • Standards-based planning rooted in DORI and the newer IEC/EN 62676-4:2025 visual-performance concepts.
- • Practical use of design tools to simulate coverage, pixel density zones, and operational effectiveness before equipment is installed.
- • Consulting expertise that translates client goals into camera placement, lensing, and infrastructure decisions instead of relying on generic “one camera covers all” assumptions.
- • Field-informed design experience spanning commercial security, surveillance systems, and performance-driven
Specification Work
For clients, this approach reduces the gap between expectation and delivered performance. It also supports better budget discipline, because the system can be engineered to deliver identification or validation only where needed, while using broader overview or observation coverage in less critical areas.
Author Qualifications
This paper is authored by Rob Oldham & Scott McQuarrie, CPP, PSP, PCI, CPTED, a security professionals with substantial practical experience in surveillance system design, camera performance specification, mobile security platforms, power systems, and technical solution development. That combination of commercial leadership and hands-on technical fluency is especially relevant in a standards discussion, because surveillance performance failures often originate in the gap between procurement language and operational reality.
Rob Oldham’s perspective is also aligned with the consulting needs of end users, integrators, and channel partners who must balance evidentiary goals, budget constraints, deployment complexity, and long-term maintainability. In that environment, standards such as DORI and IEC/EN 62676-4:2025 are most useful when interpreted by practitioners who understand both the theory and the field application.
Conclusion
DORI changed CCTV design by replacing vague image-quality expectations with measurable operational thresholds. That legacy remains important because it still provides a practical framework for matching surveillance goals to camera performance.
The 2025 edition of IEC/EN 62676-4 pushes the industry further by introducing a broader visual-performance model with additional task levels and more demanding upper thresholds. For manufacturers, consultants, integrators, and end users, the result is a more rigorous basis for specifying what a surveillance image must actually enable.
For Backstreet Surveillance, the opportunity is to lead with design intelligence rather than commodity hardware language. A standards-based consulting model built around DORI, modern IEC/EN guidance, simulation tools, and real-world deployment experience creates a stronger foundation for better system outcomes, better client expectations, and better evidentiary performance.
References
Axis Communications. (2025, November 19). From DORI to visual performance in IEC 62676-4:2025. Axis Communications Newsroom. https://newsroom.axis.com/en-us/blog/iec-62676-4-video-surveillance Axis Communications. (n.d.). Pixel density and DORI [PDF]. https://www.axis.com/dam/public/b2/d9/29/pixel-density-en-US-403691.pdf
• CCTV Design Tool Team. (n.d.). DORI standard for CCTV image quality. CCTV Design Tool.
• JVSG. (2019, August 13). How to design video surveillance system. Part 2: EN62676-4 DORI zones
• JVSG. (2025). IEC/EN 62676-4: 2025 OODPCVS support.
• Smaling. (n.d.). EN 62676-4 (formerly EN 50132-7) — CCTV design, image quality and storage guidance.
Quote #
