Quick answer: Use three practical groups to compare the printed analytics rows: target detection, rule-based classification and a wider function list. These are planning labels, not official feature packages guaranteed by a series badge. For each complete SKU, confirm firmware, rule options, simultaneous limits and the event path to the actual recorder/client. Test the required day/night scene; a human/vehicle label does not promise zero false alerts, recognition or identification.
On this page
Why does the AI row matter more than the series badge?
The catalog uses Value, Sharp and Intelligent series labels, but a badge is not an analytics package. CS-I4KS L-2.8mm lists Tripwire, Perimeter and Motion Detection; a human/vehicle filter is not stated in that row, which does not prove its absence in every firmware. CS-I4KS L-2.8-12mm lists Smart Alarm (Human & Vehicle Detection). CS-I8XQ I-2.8mm lists Tripwire, Perimeter and Human & Vehicle Classification, while CS-I8XQ II-2.8mm lists additional functions. Compare the complete SKU and actual records instead of treating a suffix as a promise.
Write the complete SKU and the required task on the schedule. Keep catalog-listed functions separate from confirmed firmware settings and tested results. Detection, classification, identification and recognition are different tasks. Camera analytics, AI by NVR and VCA Search also have separate capabilities and limits; AI NVR Explained covers the recorder side.
Level 1: What does Smart Alarm (Human & Vehicle Detection) actually do?
Smart Alarm (Human & Vehicle Detection) is listed on the named example SKUs. Treat it as a target-detection capability to verify, not a guarantee that every alert is a person/vehicle or that headlights, weather and other causes are eliminated. Confirm the actual trigger options, target sizes, lighting and event output on the proposed firmware.
Do not infer undocumented features or their absence from a short catalog label. Check the actual manual and firmware for rule configuration, metadata and client search. Likewise, Tripwire, Perimeter and Motion Detection without a classification row do not establish the precise filtering behavior. These entry-level candidates may fit:
- Small sites that want an after-hours alert for a person or vehicle in the yard, viewed in CASE Connect; app features and remote access options should be confirmed by model, software version and project configuration.
- A defined gate, lane or door may be a candidate starting view. Test the actual moving traffic, occlusions, weather and background instead of assuming nothing else moves.
- A supported camera/AI-NVR combination may be a candidate after its stream, rule, channel and workload limits are confirmed and bench-tested.
Level 2: What changes with Tripwire, Perimeter and Human & Vehicle Classification?
The named Sharp examples list Tripwire, Perimeter and Human & Vehicle Classification; CS-I5PS S-2.8mm also lists Face Capture. Confirm which combinations, directions and target filters the actual firmware exposes. Metadata delivery and searchable target tags require a compatible, tested recorder/client workflow; the classification label alone does not prove that they travel with every event. VCA Search covers those questions.
CS-I5KS SD-2.8mm lists 3072×1728@30fps, 30m IR, 10m white light, 120dB WDR, IP67/IK10, Metal+Plastic, −35°C to 65°C and a 512GB slot on physical catalog page 3. Those rows help shortlist it; they do not establish the usable analytic distance. Test target detail, motion, angle and illumination in the actual rule zone.
Face Capture on CS-I5PS S-2.8mm is printed, but it is a project-review trigger, not a quote line: confirm use case, model, software, region and legal requirements first, as set out in Facial Recognition: When Not to Promise It.
Level 3: What is in the full AI suite on Intelligent cameras?
CS-I8MS I-2.7-13.5mm and CS-I8SS I-2.7-13.5mm list a wider set, including Tripwire, Double Tripwire, object rules, counting, Heatmap, Safety Helmet Detection and Human & Vehicle Classification. The table gives questions for selected functions, not guaranteed definitions or results. Confirm actual metrics, thresholds, concurrent limits and event mapping from the exact manual. An I/O count does not establish output ratings, an external-load interface or permission to control a gate, strobe or safety function. Use the IP camera user manual for the actual revision, and request any missing rule/interface documentation.
| Printed function | Question it answers for the customer | Typical position |
|---|---|---|
| Double Tripwire | Which documented two-line sequence/direction can be configured and tested? | Fence line, dock apron |
| Loitering | Can the required zone and dwell threshold be tested without asserting identity? | ATM vestibule, rear door, fuel island |
| Object Abandon / Object Lost | Which documented object-left/object-missing event fits this scene and can be tested? | Lobby, display, tool crib |
| People Counting / Heatmap | Which counting modes and heatmap metrics are actually supported and usable here? | Store entrance, sales floor |
| Safety Helmet Detection | Which helmet-related event is documented and works in the approved view? | Site gate, plant floor entry |
What the second-generation II turrets add
The II turret pages list additional functions including Visible Light Smoke & Fire Detection, Fire Exit Occupancy Detection, Face Capture and classification categories. Confirm each required function and limit for the actual model/firmware. A video alert or obstruction rule supplements the approved response process and does not replace a required fire-alarm, exit-safety or other life-safety system. Face functions require the applicable use-case and privacy review before use. The named light/audio/storage rows do not prove those acceptance outcomes.
Which level does the job need?
Choose the event the operator needs, the action they can take and the evidence they must retrieve. Set acceptance tests for missed events and unwanted alerts under representative day/night conditions; do not promise a business outcome or nuisance-free monitoring from a feature list. The job examples are candidates for review, not proof that a planning level supplies every direction, metric, trigger or response shown.
| Job | Alarm the customer needs | Level | Example SKU (printed row) |
|---|---|---|---|
| Small yard, after-hours alert | Human or vehicle present | Level 1 | CS-I4XS L-2.8mm: Smart Alarm (Human & Vehicle Detection) |
| Fence line or lane with direction | Human crossing inbound only | Level 2 | CS-I5KS SD-2.8mm: Tripwire, Perimeter, Human & Vehicle Classification |
| Store door with traffic reporting | Approved counting direction/mode and documented heatmap metric; no dwell metric assumed | Level 3 | CS-I8MS I-2.7-13.5mm: People Counting, Heatmap |
| Blocked fire exit, back of house | Documented exit-occupancy alert, supplementary to the required safety systems | Level 3, II | CS-I8XQ II-2.8mm: Fire Exit Occupancy Detection |
| Rear lot, intervention wanted | Light, voice and alarm on a classified target | Level 3, II with EW | CS-I8XQ II-2.8mm: white light 98 ft (30 m), Red & Blue Light, speaker |
A suitable AI NVR may process supported incoming streams, subject to the exact recorder’s rule/channel limits and workload. Confirm that combination rather than assuming every camera can gain every rule. For a required ONVIF conformance claim, verify the full model and firmware in the official ONVIF database. Then test the required streams, recording, playback and events with the actual client/version. A profile list or brand name is not a complete integration statement.
How do you commission analytics so they still work in month six?
- Bench first. Check whether the required documented options are exposed on the actual camera/recorder firmware and note the versions. Record missing or unverified functions as pending, not as a passed test.
- Mount for the rule. Follow the actual rule-specific target-size, angle and scene requirements. Test the intended targets and paths; one mounting rule does not fit every analytic function.
- Place rules where the target is sufficiently visible in the actual image. Configure available target type, size, direction and schedule, then test legitimate traffic, exclusions and day/night conditions. A printed illumination maximum is not a guaranteed analytic range.
- Walk-test at night and record. Use authorized test participants and representative intended/excluded paths under the agreed conditions. Log received, missed and unwanted events and the tested settings; one pass does not guarantee future detection.
- Hand over the alert path. Where supported and approved, verify the actual client notification, recipient permissions and response workflow. Document delivery observations and limits; the CASE Connect name does not guarantee every camera event reaches every device.
What to tell your customer
We will confirm each specified function on the actual model and firmware, then test its event path and representative day/night scene. The proposal will state the tested scope and limits. Face, fire/smoke, exit and app workflows need their separate required reviews; camera analytics do not replace a required life-safety system.
FAQ
Is Human & Vehicle Detection the same as Human & Vehicle Classification?
They are different catalog labels, but they are not enough to infer every trigger, metadata tag or searchable event. Verify the exact model/firmware options and test the desired event through the actual recorder/client. Do not translate either label into recognition, identity or zero false alerts.
Can an AI NVR add tripwire and perimeter to a Level 1 camera?
CASE AI NVRs print their own AI by NVR row, which on CS-N220 P16H-AI includes Tripwire, Double Tripwire, Perimeter and Smart Motion Detection, applied by the recorder to the incoming streams. Which cameras and how many channels can use it at the same time is confirmed per NVR model and firmware, so test it on the bench before you promise it for a camera position.
Does the full AI suite work at night?
Test the actual night scene: target size/angle, motion, exposure, illumination, weather and rule settings all matter. The published IR or white-light maximum does not guarantee a usable analytic distance. Record representative results and limitations for each required rule and position. The IR-range guide distinguishes illumination from the required recorded and analytic task.
Can I promise smoke and fire detection from a camera?
Describe a function only where the exact model lists it and the proposed configuration has been confirmed and tested. Video fire/smoke or exit-obstruction analytics are supplementary alerts, not a substitute for a required certified life-safety system or its inspections. Confirm the response workflow and responsible authority’s requirements before relying on such an alert.
References
- Anabon current model documents and manuals
- ONVIF model and firmware conformance database
