IVISION · Monitoring workflow guide

Use Cases and Capability Planning Guide

Use this guide to identify an observable condition and follow the linked detail page for camera considerations, examples and limitations. Catalog entries describe workflows to assess; confirm detector availability, event creation and notification delivery for your deployment.

Illustrations and website previews are examples, not evidence of detection performance. Human review remains part of the response.

Face Recognition

Evaluate identity-related observations at controlled camera positions. Face detection locates a face; recognition compares an eligible image with an enrolled reference. A possible match, an unfamiliar appearance and an access decision are different outcomes.

Face Detection & Recognition

Locates faces and, with a supported recognition service, compares observations with approved enrollment images.

A match cannot prove identity, intent or permission.

Smart Attendance

Associates incoming attendance captures with employee records for time and presence review.

Camera presence does not establish hours worked or productivity.

multi camera face tracking

Associates candidate face sightings across selected cameras when a supported matcher is available.

Blind spots and false matches can break an inferred route.

Unauthorized-access detection

Combines a person observation with defined area permissions to flag a possible access exception.

An unknown face alone does not prove unauthorized access.

Unauthorized-person detection

Flags a person observation without an accepted association to an approved list for a defined context.

No match does not establish intent or a violation.

Emotion & Sentiment Detection

A legacy catalog label for visible expression cues; facial appearance cannot establish emotion or sentiment.

Cannot establish emotion, honesty, mental health or performance.

Stranger / Unknown Person Alert

Marks a face observation without a confirmed known-profile association for review.

Cannot prove danger or that someone has never visited.

Visitor Registration & Identification

Connects an approved visitor record with arrival evidence and an agreed identity check.

A face match cannot establish an appointment or escort permission.

Blacklist Alerting

Compares candidate observations with an explicitly authorized watchlist for verification.

Cannot prove identity, wrongdoing or intent.

Age & Gender Estimation

A legacy entry for estimated appearance categories, not verified age, sex or gender identity.

Cannot verify legal age, gender identity, sex or eligibility.

Mask / Helmet Detection

Observes visible mask or helmet presence when image detail supports the selected item check.

Cannot confirm certification, fit, fastening or protection level.

Masked or covered face detection

Identifies observations where coverings obscure facial detail needed for the intended check.

Cannot establish motive, threat or identity behind a covering.

Employee working-time analysis

Interprets approved attendance timestamps under agreed work-time rules to support a reviewable time summary.

Cannot infer productive work from camera presence or calculate entitlement without policy.

Employee entry and exit registration and analysis

Registers and reviews employee-associated passage events at defined arrival and departure points.

Cannot establish a work shift or permission solely from a passage.

Known-person and unknown-person classification

Classifies an observation as associated or unassociated with a defined known-profile set.

Cannot establish legitimacy, threat or whether someone has visited previously.

Creation of visit histories for unknown or unidentified individuals

A proposed workflow for linking unconfirmed sightings into a reviewable visit history without assigning a verified identity.

Cannot prove identity, intent or that similar sightings belong to one person.

Blacklist and watchlist management

Maintains an authorized reference list with provenance, validity and review controls for screening workflows.

A list entry cannot itself prove wrongdoing or current access status.

Shift-based attendance analysis

Associates reviewed attendance records with an approved shift schedule for exception analysis.

Cannot establish task completion or staffing competence from attendance alone.

Late-arrival and early-departure detection

Compares reviewed arrival and departure records with approved schedule boundaries.

Cannot establish misconduct or actual work start/end from a capture alone.

Appearance Labels and Gender: Review Boundaries

Documents why a visual appearance label must not be treated as a person's gender identity or a verified demographic fact.

Images do not establish gender identity, sex, personal preferences or eligibility for a service.

People Events

Review visible actions such as a fall, prolonged presence, running or phone use in a defined area. These observations can help an operator find a relevant moment; they do not establish intention, misconduct, injury or a person's mental state.

Phone Usage Detection

Flags visible handheld-phone interaction in a defined area when the object and gesture are discernible.

Cannot determine what someone is viewing or whether they are distracted.

Slip & Fall Detection

Flags a visible movement sequence consistent with a possible fall.

Cannot diagnose injury, cause or loss of consciousness.

Person Down / Collapse Detection

Flags a person remaining in an unusual low or lying posture in a monitored area.

Cannot establish consciousness, injury or cause.

No Food / Drink Zone Compliance

Flags visible food, drink or consumption-related objects in an area with a defined restriction.

Cannot establish substance content, contamination or actual ingestion.

Sleeping / Inactivity Detection

Observes sustained low movement or posture patterns that may warrant a contextual check.

Cannot prove sleep, fatigue, neglect or medical condition.

Earphone and headset usage

Flags visible ear-worn devices where a defined operational policy restricts their use.

Cannot establish audio playback, volume or whether warnings were heard.

abnormal activity detection

Flags a defined visual departure from expected activity in a specific monitored scene.

Cannot establish suspicious intent or explain why activity differs.

Driver and personnel drowsiness

A catalog workflow for reviewing visible fatigue-related cues where suitable close-view analysis is validated.

Cannot diagnose fatigue, health status or fitness to drive.

Vehicle-person proximity detection

Flags a person and vehicle appearing within a defined separation criterion in the monitored view.

Cannot prove collision risk or a measured distance without suitable calibration.

Vehicle-person near-miss detection

Flags a close interaction whose visible paths may indicate a near-collision candidate.

Cannot prove fault, inevitable collision or injury avoidance.

Pedestrian-pathway monitoring

Observes pedestrian movement relative to designated walking routes.

Cannot establish why someone departed or whether a deviation was authorized.

Pedestrian entry, vehicle-only areas

Flags visible pedestrian entry into an area designated for vehicle traffic only.

Cannot establish unauthorized status or safe separation by entry alone.

Unsafe crossing detection

Flags a pedestrian crossing outside defined points or against a validated traffic rule.

Cannot determine safe stopping distance or driver awareness from a crossing alone.

Collision-risk estimation

A catalog workflow for evaluating potential path conflicts from visible vehicle and pedestrian motion.

Cannot guarantee prediction, warning lead time or collision prevention.

Running Detection

Flags movement patterns consistent with running within a defined monitored area.

Cannot establish panic, intent or physical speed without further evidence.

Suspicious Behavior Detection

A legacy label for reviewing explicitly defined activity exceptions, not determining suspicious intent.

Cannot infer criminal intent, trustworthiness or threat from appearance.

Fighting / Violence Detection

Flags visible interaction sequences that may resemble a physical altercation.

Cannot establish aggression, intent, responsibility or injury.

Loitering Detection

Flags a person remaining within a defined zone longer than an agreed duration.

Cannot establish intent, unauthorized status or continuous presence during track gaps.

Climbing / Restricted Height Access

Flags visible climbing or entry into a defined elevated area.

Cannot verify structural safety, anchorage or safe working method.

Smoking Detection

Flags visible objects or gestures consistent with smoking in a defined restricted area.

Cannot identify substances or prove smoking from a gesture alone.

Uniform / Dress Code Compliance

Compares visible workwear attributes with a defined site uniform requirement.

Cannot establish role, identity, fabric certification or personal intent.

Gathering / Grouping Detection

Flags multiple people remaining together within a defined area and period.

Cannot establish association, motive or an unsafe crowd from grouping alone.

Detection of unusual employee movement or activity

Flags a clearly defined employee-associated activity exception within an approved operational scope.

Cannot establish intent, misconduct or job performance.

Prolonged Inactivity Observation

Reviews a visible person or area that remains without sufficient observable movement for a configured interval.

Occlusion, a stalled stream and quiet work can resemble inactivity. Stillness does not establish unconsciousness, sleep or misconduct.

Possible Collapse Observation

Reviews an observed transition from an upright posture toward the ground that may warrant a prompt welfare check.

Sitting, kneeling, occlusion and camera motion can resemble a collapse. The cause, injury and medical status cannot be diagnosed.

Worker Safety

Examine visible protective equipment, access to machinery areas and work positioning. Each scenario must be tied to a site safety rule and a camera view that can show the relevant equipment or boundary. A visual indication supports a safety check; it does not certify compliance.

PPE Compliance

Reviews visible protective equipment against the requirements of a defined work area.

A visible item does not establish fit, certification or correct use.

Helmet Compliance

Flags people whose heads are visible but whose required protective helmets are not identifiable.

A camera cannot establish impact rating, fastening or internal damage.

Safety Vest Compliance

Reviews whether a person has a visible high-visibility garment in an area where it is required.

Presence does not establish reflective performance or garment certification.

Gloves Compliance

Reviews visible hands for required protective gloves during a defined task or access check.

An image cannot confirm chemical resistance, sterility or safe glove use near rotating equipment.

Goggles Compliance

Reviews whether required eye protection is visibly present during a defined work-area check.

Appearance cannot establish lens rating, seal or suitability for welding or chemicals.

Harness Compliance

Reviews visible harness equipment on people approaching or working in a designated elevated area.

Presence alone cannot prove tie-off, anchor strength or a safe fall distance.

Staircase Safety

Supports review of predefined visible events on stairways, such as a person down or an obstructed landing.

Video cannot establish injury, traction or every unsafe behavior.

Machine Guarding (LOTO)

Reviews visible machine-guard position or access around a guarded machine; energy isolation remains a separate verified procedure.

Electrical, hydraulic and stored energy may remain invisible.

Mask Detection

Reviews whether a required face covering is visibly present in a defined area.

Video cannot determine infection, filtration, seal or medical status.

Working at Height

Reviews people entering or remaining in defined elevated-work areas visible to a camera.

A perspective image alone cannot establish height, fall distance or safety compliance.

Confined-Space Safety Monitoring

Supports visual review of people at a defined confined-space access point or observable interior area.

Video cannot measure oxygen, toxic gas or hidden occupants.

Unsafe Proximity to Machinery

Reviews people approaching a defined clearance region around machinery.

Pixel separation is not automatically physical distance or proof of machine operation.

Machine Zone Entry

Reviews a person crossing into a defined machine operating or maintenance area.

Entry does not establish unauthorized access or hazardous machine motion.

People–Equipment Near Miss

Identifies candidate close interactions between people and moving equipment for contextual review.

Video alone may not establish distance, intent, braking or causation.

Missing Safety Equipment

Reviews absence of an explicitly defined visible safety item from a person or an assigned location.

Absence from the image does not prove loss or lack of protection.

Unsafe Work Practices

Reviews a narrowly defined observable work action against an approved task procedure.

Video does not establish intent, competence or every procedural step.

Lockout/Tagout Compliance

Supports review of visible steps and access context associated with an approved lockout/tagout procedure.

Video cannot prove zero energy, lock ownership or completion of hidden steps.

Confined-space entry detection

Flags visible passage through a defined access point to a designated confined space.

Cannot assess atmosphere, isolation, rescue readiness or safety inside the space.

Unsafe climbing detection

Flags a defined climbing action at an object or access point where the site rule requires review.

Cannot establish equipment stability or complete work-at-height compliance.

Machine-guarding compliance monitoring

Reviews visible guarding or barrier conditions against a defined machine-specific visual criterion.

Cannot verify interlocks, energy isolation, internal hazards or safety performance.

Vehicle Safety

Bring together vehicle observations, loading-area movement, parking activity and forklift interactions. Plate reading, counting, pathway review and speed estimation require different views and validation. A vehicle event is a prompt to examine the route and context, not proof of an offence.

Vehicle Detection

Identifies candidate vehicle objects in an assessed camera view.

Presence does not establish identity, authorization or a complete movement history.

Vehicle Tracking

Links observations of a vehicle through an assessed sequence or camera view.

A track does not automatically establish plate identity or continuity through blind spots.

Vehicle Counting

Counts validated vehicle passages across a defined line or through an assessed region.

A passage count is not necessarily unique vehicles or site occupancy.

Vehicle Classification

Assigns detected vehicles to broad supported visual categories such as cars, vans or trucks.

Appearance cannot establish payload, legal vehicle class or exact specification.

Vehicle-Type Identification

Assesses more specific supported vehicle appearance labels beyond a broad vehicle category.

Exact model, year and mechanical specification are not established without separate evidence.

Vehicle Information Storage

Organizes supported vehicle-related observations and associated metadata in reviewable records.

Storage does not prove recognition accuracy or unlimited search coverage.

Vehicle Entry and Exit Logging

Records supported vehicle crossings at designated entry and exit points with their available timestamps.

Missed crossings or uncertain identity can prevent complete visit matching.

Vehicle Blacklist and Watchlist Alerts

Compares supported vehicle identifiers with an authorized site-maintained list for human review.

A list match does not establish driver identity, intent or criminal conduct.

License Plate Recognition

Extracts candidate license-plate characters from a sufficiently detailed vehicle image.

A reading does not establish ownership, driver identity or registration validity.

Automatic Number Plate Recognition

Assesses automated plate capture and reading as vehicles pass through a defined camera approach.

Plate recognition does not automatically authorize or physically admit a vehicle.

Vehicle-Traffic Heatmaps

Summarizes supported vehicle-position observations as a spatial concentration view over a defined period.

Density does not by itself measure danger, unique vehicles or physical coordinates.

LPR and ANPR-Based Vehicle Search

Retrieves supported stored vehicle observations using reviewed plate text or available plate-search criteria.

Search cannot recover unread plates, unindexed footage or deleted records.

Parking-Area Monitoring

Reviews supported vehicle occupancy and activity indications within an assessed parking area.

Visible occupancy does not establish permit status or total site capacity.

Authorized-Parking Detection

Reviews whether a vehicle’s observed parking position meets an explicitly defined permitted-zone rule.

Location alone cannot establish permit validity or driver entitlement.

Unauthorized-Parking Detection

Flags candidate parking positions or durations that differ from an explicitly defined site rule.

An image does not establish legal liability or every permission exception.

Vehicle Pathway Monitoring

Reviews vehicle movement relative to a designated site driving route.

Camera-relative paths do not automatically provide precise physical location.

Vehicle Stop-Sign Compliance

Reviews a vehicle’s movement at a defined stop position using validated stop-event criteria.

Video does not automatically establish a legal traffic offense.

Restricted-Zone Vehicle Entry

Reviews vehicles crossing into a defined area where vehicle access is restricted.

Crossing alone does not establish permission or driver intent.

Forklift Detection and Tracking

Identifies candidate forklifts and links their visible positions through assessed camera coverage.

A track does not identify the driver or prove continuity through blind spots.

Forklift Safety

Brings selected forklift-related visual observations into a defined supervisor review workflow.

Camera observations cannot certify safe operation or prevent all collisions.

Forklift Speed Monitoring

Assesses forklift movement speed only where a suitable measurement and calibration method is verified.

Uncalibrated apparent motion cannot establish physical speed.

Forklift Driver Detection

Reviews whether a person is visibly present in a forklift’s designated operating position.

Presence cannot establish identity, training or actual control state.

Driver and Non-Driver Classification

Assigns tentative operational roles from a person’s visible spatial relationship to a forklift.

A visual role does not establish employment, permission or who drove earlier.

Forklift Maximum-Height Monitoring

Reviews a forklift mast, fork or load position against an explicitly defined height reference.

The image alone cannot establish load stability or structural clearance.

Raised-Fork Safety Monitoring

Reviews visible forklift travel while the forks appear above an assessed travel-position reference.

Video cannot establish load stability, rated capacity or exact height without calibration.

Forklift Pathway Monitoring

Reviews forklift movement relative to a designated materials-handling route.

Image overlap with a boundary does not always mean physical encroachment.

Forklift Stop-Sign Compliance

Reviews forklift movement through a defined stop position using validated stopping criteria.

A visual indication does not prove intent or a legal offense.

Forklift Restricted-Zone Monitoring

Reviews forklift crossings into an area where forklift access is restricted.

Entry alone does not establish permission or operator intent.

Forklift Near-Miss

Identifies candidate close interactions involving forklifts for sequence-based human investigation.

Apparent proximity does not establish collision risk, blame or exact distance.

Forklift Operation Zone

Reviews forklift presence and movement within an explicitly defined permitted operating area.

Zone presence does not establish safe capacity, clearance or operation.

Entry into Pedestrian-Only Areas

Reviews vehicles crossing into a region explicitly reserved for pedestrians.

Image overlap alone may not prove physical intrusion or a near miss.

Vehicle Dwell Time

Measures the duration of a supported vehicle presence or visit within a defined observation region.

Track loss or incomplete coverage can distort duration.

Unauthorized Vehicle

Reviews a vehicle observation against current authorized access information where that comparison is supported.

No match is not proof of wrongdoing or driver identity.

Wrong Direction

Reviews vehicle movement against a defined permitted travel direction in an assessed route.

A short reverse or turn does not necessarily violate the traffic plan.

Overspeed

Reviews calibrated vehicle-speed estimates against an agreed site threshold.

Uncalibrated image motion cannot establish speed or legal enforcement evidence.

Vehicle speed estimation

Estimates vehicle motion speed from a calibrated view and validated timing method.

Cannot guarantee exact speed or legal measurement validity without validation.

Authorized and unauthorized forklift-operator detection

Compares a verified operator association with current authorization for a specific forklift task.

Cannot verify training, fitness or identity from seat presence alone.

Unsafe forklift parking detection

Flags a forklift remaining in a location or visible configuration that conflicts with a defined parking rule.

Cannot verify parking brake, energy isolation or complete safe shutdown.

Detection of stopped or improperly parked vehicles

Flags a vehicle remaining stationary in a defined location beyond an agreed period.

Cannot determine driver intent, breakdown status or legality from stillness alone.

Detection of vehicles or equipment approaching restricted areas

Flags a vehicle or supported equipment object entering a defined approach area near a restricted boundary.

Cannot predict entry intent or guarantee warning time.

Area Monitoring

Define the part of a camera view that matters to an operational rule: a restricted doorway, equipment zone, boundary or occupied space. The relationship between an observed object and the configured area is the starting point for review; authorization must come from separate policy or records.

Restricted Zone Monitoring

Observe supported activity inside a camera-defined restricted area.

Presence does not establish identity, access permission or malicious intent.

unsafe entry into hazardous areas

Surface candidate entry into an area designated hazardous by the site.

Video cannot determine all hazards, actual exposure or whether entry is safe.

Area Intrusion

Identify a candidate crossing into a monitored area under a validated entry rule.

A crossing does not prove criminal intent or establish access permission.

Virtual fence monitoring

Use an image-defined line or boundary to review supported object crossings.

An image line is not a surveyed physical fence or guaranteed perimeter protection.

electronic boundary monitoring

Review supported camera observations associated with designated boundary segments.

This workflow does not imply a physical sensor fence, complete coverage or cross-camera identity.

Time Limited Zone

Review zone observations against a validated time condition such as a schedule or permitted dwell interval.

Image presence does not establish a booking, permission or continuous dwell across unseen gaps.

Time Lapse Zone Analysis

Compare retained observations of a defined zone across selected times or intervals.

Sampled views can miss short events and cannot establish continuous activity.

people approaching prohibited areas

Observe a person moving toward a site-defined prohibited-area boundary.

An approach does not prove intent to enter or predict the person’s next action.

Pedestrian entry, vehicle-only areas

Identify candidate pedestrian presence or entry within a designated vehicle-only image area.

A camera observation does not establish collision risk, stopping distance or safe vehicle control.

Restricted-height access detection

Review a visible person or object against a defined height-restricted access condition.

Uncalibrated image size does not establish physical height, clearance or safe passage.

Occupancy Metrics (Real-Time)

Present a current estimate of visible occupancy from a supported counting workflow.

Visible occupancy is an estimate and may exclude people behind obstructions or outside coverage.

Zone occupancy views

Present visible occupancy by defined camera zone where the underlying observations are supported.

A displayed zone state is not proof of complete coverage, permission or safe capacity.

Notifications through dashboards, mobile applications, email, SMS, APIs or messaging platforms

Notification-channel configuration defines where an event response should be delivered, subject to a connected and tested delivery service.

Outbound delivery is not verified in the reviewed source. Restricted-zone SMS is marked as coming soon; channel labels are not availability guarantees.

Multi-site and multi-camera management

Multi-site and multi-camera management organizes camera records, operational context and authorized access across an agreed deployment structure.

Cross-site aggregation, federation, scale limits and centralized health monitoring need confirmation. Tenant isolation is not evidence of every multi-site workflow.

Intelligent area and zone monitoring

Defines a camera-visible area and the specific observations that should be reviewed within it.

Cannot observe hidden conditions or determine every activity within a boundary.

Perimeter intrusion detection

Flags a supported boundary-crossing observation at a defined site perimeter.

Cannot see through barriers or establish intent and permission from motion alone.

Zone occupancy and activity monitoring

Reviews presence and selected supported activity within a defined camera-visible zone.

Cannot establish productivity, intent or complete activity from presence alone.

Custom rules for each area, camera or time period

Stores camera-relative zone geometry, schedules and selected event criteria for an area-specific monitoring workflow.

Saving a rule does not prove detection or notification execution.

People Counting

Use entry, exit, presence and dwell observations to examine how spaces are used. An entrance count, a current occupancy estimate and a queue estimate answer different questions. Interpretation depends on camera coverage, counting direction, occlusion and the period being compared.

People Counting (Entry / Exit)

Counts observed passages across defined entry and exit boundaries.

Cannot identify unique visitors or total occupancy without additional accounting.

Occupancy Metrics (Real-Time)

Estimates how many people are present in a defined monitored space at the latest observation.

Cannot guarantee a complete headcount or locate everyone in an emergency.

Maximum-capacity monitoring

Compares a validated occupancy estimate with a defined operational capacity threshold.

Cannot determine legal capacity or guarantee safe evacuation.

Min and max occupancy alerts

Flags occupancy estimates outside an agreed lower or upper operational range.

Cannot establish staffing adequacy or safety solely from total count.

Min and max worker-count alerts

Compares an estimated worker count with a defined staffing range in a monitored area.

Cannot establish qualifications, productivity or safe staffing from count alone.

Visitor Occupancy Tracking

Estimates visitors present in a defined space using an agreed, validated classification method.

Cannot infer visitor status from appearance alone.

Employee occupancy tracking

Estimates employee presence within selected monitored areas using a verified staff association method.

Cannot establish hours worked, productivity or complete whereabouts.

Crowd-density mapping & analysis

Shows estimated concentrations of people across a defined visible area.

Cannot certify safe crowd pressure or visibility of every person.

Gathering / Grouping Detection

Identifies a cluster of people remaining together within a defined counting area.

Cannot establish social relationships, shared identity or intent.

Queue Length & Wait Time Estimation

Estimates people waiting in a defined queue and, where tracking supports it, the time spent waiting.

Cannot infer customer satisfaction or measure unseen waiting accurately.

Dwell Time Analysis

Measures observed time spent within a defined zone when tracking remains sufficiently continuous.

Cannot establish interest, satisfaction or continuous presence during gaps.

Heatmap & Traffic Flow Analysis

Aggregates observed movement or presence into spatial patterns over a defined period.

Cannot explain motives or represent unseen areas.

Desk Occupancy

Estimates whether a defined desk or workstation is visibly occupied.

Cannot determine productivity, booking ownership or hours worked.

Workspace utilization analysis

Combines defined occupancy and duration measures to describe how monitored workspaces are used.

Cannot measure employee productivity or explain preferences by itself.

Time Lapse Zone Analysis

Summarizes observed activity in a defined zone across an extended review period.

Cannot prove that unrecorded intervals were inactive.

Overcrowding detection

Flags a monitored area whose occupancy or concentration exceeds a defined operational criterion.

Cannot measure crowd pressure or guarantee safe conditions below the threshold.

Understaffing detection

Flags estimated worker presence below an agreed task-specific requirement.

Cannot determine competence, workload or safe staffing solely from video.

Visitor-activity heatmaps

Visitor-activity heatmaps summarize where observed visitor movement or dwell accumulates within a defined camera view and reporting period.

Visitor classification and map generation need validation. Occlusion, repeat visits and blind areas affect interpretation; heat does not measure satisfaction or purchasing intent.

Workplace-activity maps

Workplace-activity mapping aggregates observable movement or presence across defined work areas for space and circulation review.

Off-camera work, stationary people, obscured areas and different task types limit interpretation. Map availability requires confirmation.

Time-lapse occupancy analysis

Summarizes occupancy estimates at defined intervals over a selected period.

Cannot prove exact headcount or continuous absence between samples.

Employee presence-duration analysis

Summarizes supported periods of employee-associated presence within a defined observation scope.

Cannot establish continuous presence, productivity or paid hours from gaps.

Minimum and maximum worker monitoring

Monitors estimated worker presence against a defined minimum and maximum for an active task area.

Cannot prove competence, workload or safe staffing from numbers alone.

Employee dwell-time analysis

Measures qualified employee-associated time within a particular defined zone.

Cannot establish productivity, avoidance or continuous presence through gaps.

Employee-presence and workforce heatmaps

Aggregates validated employee-presence observations into a spatial summary for an approved period.

Cannot establish productivity, preferences or complete attendance.

Crowd Density Map

Presents spatial estimates of observed crowd concentration across a defined camera area.

Map color does not establish a legal capacity limit, individual identity or the cause of crowding. Perspective and occlusion affect interpretation.

Queue Waiting-Time Estimation

Estimates an observed interval between defined queue-entry and service or exit observations.

Hidden joins, lane switching, occlusion and early departures can break association. An observed interval may exclude waiting outside coverage.

People Activity Heatmap

Aggregates observed people activity by area over a specified time window to show relative spatial intensity.

A hot area may reflect repeated passages or long presence depending on the unit. Heat does not prove preference, intent, a sale or productivity.

Fire & Emergency

Assess visible indications of possible fire or smoke and the condition of emergency routes or equipment locations. Cameras may provide useful context where a supported detector has been validated, but they do not replace certified fire detection, alarms, inspections or emergency procedures.

Fire Detection

Reviews visible flame-like patterns as candidate indications of fire.

Hidden combustion remains unseen; cameras do not replace certified alarms.

Smoke Detection

Reviews visible smoke-like plumes or haze in an assessed scene.

Video does not measure concentration or invisible gases.

Smoke and Fire Detection

Reviews smoke-like and flame-like indications while retaining the distinction between the two cues.

Absence of one cue does not rule out fire.

Emergency Situation Detection

Brings explicitly defined visual incident cues into an emergency review workflow.

Video cannot establish every emergency or diagnose medical conditions.

Missing Fire Extinguisher Detection

Reviews whether an expected extinguisher is absent from its visible assigned position.

Presence cannot establish pressure, service date or suitability.

Fire Extinguisher Presence Monitoring

Supports recurring visual checks that an extinguisher remains at an assigned visible station.

A visual record does not certify readiness or completed inspections.

Blocked Emergency Exit Detection

Reviews objects occupying a defined emergency-exit approach or clearance region.

An image cannot certify exit width, door operation or the full escape route.

Emergency Exit Accessibility

Supports review of visible conditions along an emergency-exit approach and doorway.

The view cannot establish full accessibility or evacuation capacity.

Unsafe Door Status Detection

Reviews a door’s visible position against an explicitly defined expected condition.

Position does not establish lock, latch or fire performance.

Emergency Event Recording

Supports review of available clips or snapshots associated with a recorded emergency-related event.

Gaps, clock drift and limited views leave incomplete evidence.

Hazards

Review visible floor changes, obstacles, spills and plume-like appearances in areas where the camera can show them. Images can suggest a surface or scene change, but cannot reliably establish a substance, gas concentration, toxicity or whether a surface is safe to walk on.

Visible Hazardous-Material Leaks

Reviews visible material release near an assessed vessel, pipe or storage point.

Video cannot establish chemical identity, concentration or exposure.

Liquid-Spill Detection

Reviews visible liquid accumulation on a monitored surface.

Video does not establish liquid type, depth or slip resistance.

Gas, Vapor or Smoke-Like Leak

Reviews a visible plume near an assessed release point without identifying its chemical composition.

Invisible gases and concentrations are outside ordinary visual evidence.

Wet-Floor Detection

Reviews floor appearance that may indicate a wet surface.

A camera cannot measure friction or determine that a floor is safe to walk on.

Obstacle Detection

Reviews objects occupying a defined passage or work-area clearance region.

Video alone cannot determine every required clearance or object hazard.

Unsafe Walkway Detection

Combines selected visible route-condition checks within an explicitly defined walkway.

Structural strength, traction and full accessibility are not established visually.

Hazardous-Area Entry Monitoring

Reviews people crossing into a defined area that the site has designated as hazardous.

Entry alone does not establish permission, exposure or actual injury.

Spill and Leak Location Identification

Adds camera and scene-location context to a reviewed spill or leak indication.

A visible pool does not necessarily identify the upstream source.

Facility Monitoring

Check visible changes around doors, storage positions, aisles and unattended items. Useful rules depend on the normal state of the specific space: a pallet awaiting collection differs from a pallet blocking an exit, and a closed-looking door may not be locked.

Unattended-object detection

Identify a visible object remaining in a monitored place without an observable nearby attendant.

Absence of a visible attendant does not prove abandonment or a security threat.

Abandoned-object detection

Surface an object that persists after an apparent separation from nearby activity for further review.

Video does not establish ownership, deliberate abandonment or hazardous contents.

Unattended pallet detection

Identify a pallet remaining in a monitored area without visible handling activity.

Video does not establish inventory ownership, contents, load stability or dispatch status.

Unattended box detection

Surface boxes remaining in a monitored place without visible handling activity.

A visual box category does not reveal contents, recipient or delivery completion.

Obstruction detection

Identify visible objects occupying a defined passage or clearance area.

Image overlap alone does not prove legal clearance, accessibility compliance or emergency-route safety.

Door-open and close monitoring

Observe the visible open or closed position of a camera-covered door.

Visible closure does not establish that a door is locked, latched or secure.

Facility cleanliness monitoring

Review visible litter, residue or disorder against agreed facility appearance criteria.

Video cannot establish pathogens, chemical contamination, hygiene certification or an unseen spill.

Improper object placement detection

Identify a visible object outside a defined permitted placement area.

The image does not establish ownership, load limits or precise physical clearance without calibration.

Housekeeping and Workplace Organization Monitoring

Review visible workspace arrangement against agreed housekeeping criteria.

Video does not certify 5S compliance, employee performance or overall workplace safety.

Blocked-walkway detection

Flags a visible object occupying part of a defined pedestrian walkway.

Cannot certify legal clearance or safe passage from the image alone.

Restricted-door monitoring

Reviews visible door-state or passage observations at a designated restricted doorway.

Cannot verify lock engagement, access permission or fire rating visually.

Storage-area compliance monitoring

Compares visible object placement with a defined storage-area arrangement.

Cannot verify load capacity, substance compatibility or concealed conditions.

Production Monitoring

Evaluate visible product, component and process conditions at defined inspection points. Counting, orientation, assembly checks and surface inspection need product-specific acceptance criteria, representative samples and controlled imaging. A general surveillance view is not automatically an industrial inspection system.

Production-line monitoring

Review visible activity and selected observable states along a production line.

Video alone does not establish internal machine state, root cause or production output.

Production-line counting

Count supported unit crossings at a defined production-line checkpoint.

A crossing count is not automatically a count of good output or unique finished products.

Product counting

Estimate the number of supported visible products in a defined inspection or staging area.

Hidden, stacked or visually merged items may not be counted; inventory completeness is not implied.

Component and part counting

Count supported visible components or parts before a defined assembly or packing step.

Counting does not establish part identity, dimensions, material or hidden contents.

Visual production analysis

Interpret selected visible production observations alongside operational context.

Images do not establish hidden process variables, causal relationships or complete performance metrics.

Production-flow analysis

Review visible movement and waiting between defined production stages.

Video does not establish material genealogy or hidden movement outside coverage.

Production-cycle monitoring

Measure intervals between validated visible start and end events for a production cycle.

A visual interval may differ from machine cycle time and does not establish worker productivity.

Production-speed analysis

Estimate observed production cadence from validated unit movement or crossing intervals.

Uncalibrated video does not establish exact distance, conveyor speed or maximum machine capacity.

Bottleneck detection

Identify visible accumulation or persistent slow transfer as a candidate production constraint symptom.

Accumulation alone does not prove a bottleneck or identify its root cause.

Missing-component detection

Check whether a specified visible component is present at an inspection point.

Not visible does not necessarily mean absent; hidden parts and electrical function remain unverified.

Incorrect-component detection

Compare a visible component with a validated reference for the expected component type.

Visually identical parts may differ internally; appearance does not establish material or electrical specification.

Defective-part detection

Identify candidate visible anomalies on individual parts before assembly or further processing.

Video does not reveal every internal defect, dimensional error or functional failure.

Defective-product detection

Flag candidate visible anomalies on a finished or substantially completed product.

A visual pass does not establish internal condition, function or full specification compliance.

Surface-defect detection

Identify candidate visible marks or finish irregularities on an inspected surface.

Video does not establish defect depth, material integrity or hidden subsurface flaws.

Product-quality inspection

Evaluate specified visible product criteria at a defined quality checkpoint.

A visual inspection does not certify regulatory compliance or complete product conformity.

Assembly-error detection

Compare a visible assembled state with the expected configuration at a defined step.

Video does not establish hidden connections, torque, internal fit or functional performance.

Packaging-error detection

Review visible packaging configuration against approved packing criteria.

An image cannot establish hidden contents, seal strength or transport fitness.

Label and barcode verification

Compare a readable label or code with expected data using a separately validated reading method.

OCR, barcode decoding, grading and database integration are not assumed available.

Incorrect-position detection

Compare a visible product or component location with a defined permitted position.

Image coordinates do not establish physical tolerances without calibration.

Product orientation verification

Compare a visible product’s facing or rotation with the expected orientation.

Symmetric or hidden features may make orientation impossible to determine from one view.

Production-Line Fault Detection

Surface visible symptoms of an abnormal production-line condition for investigation.

Images do not diagnose electrical, software or hidden mechanical faults.

Operational-efficiency analysis

Interpret validated visible activity and timing observations alongside production inputs and records.

Camera activity alone does not establish OEE, ROI, labor productivity or causation.

Rejected-product analysis

Review confirmed reject records and their available images to investigate recurring quality issues.

Images alone do not establish that an item was rejected or identify the cause of a defect.

Production-line activity maps

Production-line activity maps summarize observed movement or presence around configured line segments and work zones.

Camera activity is not a calibrated production counter, quality measure or machine-state signal. Mapping and object support must be validated.

Abnormal production-condition detection

Flags a defined visible departure from an expected production condition.

Cannot diagnose machine faults or internal product quality from a broad anomaly label.

Tracking & Movement

Organize observations by camera, location and time to examine visible movement. A within-camera track, a mapped observation and a cross-camera association are different data products. Gaps, identity uncertainty and stale timestamps must remain visible when interpreting a route.

Real-time person tracking

Associates visible person detections across frames within a monitored view when a tracking provider is connected.

Cannot identify a person or track through blind spots by itself.

Real-time employee tracking

Associates movement observations with approved employee records where identity and tracking inputs are verified.

Cannot establish productivity or movement outside coverage.

Real-time visitor tracking

Associates permitted visitor observations across a defined monitored route where registration and tracking support are available.

Cannot infer visitor identity from appearance or see unmonitored movement.

Real-time vehicle tracking

Links visible vehicle detections across frames to describe movement within a monitored scene.

Cannot establish plate identity, exact speed or movement outside view.

Real-time equipment tracking

Associates visible equipment observations across frames for a defined, supported object class.

Cannot guarantee a unique asset identity or locate hidden equipment.

Last-seen location display

Shows the location and time of the latest available observation for a selected tracked entity.

Cannot establish current position after the observation.

Live location maps

A catalog workflow for placing recent, associated observations on a calibrated facility map.

Cannot guarantee exact coordinates, continuous coverage or current position.

Interactive facility maps

Provides navigable facility-plan context for selected areas and associated camera information.

Does not itself detect people, route movement or certify safe access.

Camera-location maps

Places known camera positions and labels within a maintained facility plan.

Cannot prove current coverage, stream availability or precise object location.

Movement-path visualization

Displays a sequence of observed positions as a path within the supported view or coordinate system.

Cannot establish unseen paths or the reason for movement.

Cross-camera route reconstruction

Orders associated sightings from different cameras into a candidate route for review.

Cannot establish unobserved movement or identity from similarity alone.

Historical route visualization

Displays available movement observations from a selected past period as a reviewable sequence.

Cannot recover missing footage or prove movement between observations.

Tracking across selected intervals

Restricts movement review to observations within an explicitly chosen time range.

Cannot establish movement before or after the selected evidence.

Entry and exit point visualization

Shows observed passages at identified entry and exit points within a facility layout.

Cannot establish total occupancy or access permission from a passage marker.

Zone transition analysis

Summarizes observed movement from one defined zone to another when tracks and boundaries support the link.

Cannot establish movement through unobserved intermediate areas.

Frequently visited area analysis

Ranks areas by a defined aggregate visit measure over a selected period.

Cannot establish preference, importance or unique people without additional validation.

High-traffic and low-traffic area

Compares observed movement intensity across defined areas using a consistent metric and period.

Cannot prove unused space, safety or the cause of movement differences.

Natural-language video search

A proposed search workflow that turns a plain-language request into supported video or event criteria.

Cannot infer intent, recover unseen events or guarantee semantic understanding.

Voice-based video search

A proposed workflow that transcribes an authorized spoken request before applying supported search criteria.

Cannot guarantee transcription, supported intent or hands-free safety.

Reference-image search for people, vehicles and objects

A proposed search workflow that compares an authorized reference image with indexed observations.

Cannot prove identity, ownership or unseen presence from similarity.

Search by clothing color or appearance

A proposed search workflow using supported visible attributes such as clothing color or object appearance.

Cannot establish identity, ethnicity, religion, character or intent.

Search by location, camera, date or time

Event filtering narrows accessible records by their camera, location and recorded time before an operator examines the evidence.

Metadata filters do not search the visual contents of arbitrary recordings or prove an event did not occur.

Search for specific activities or incidents

Activity and incident search selects candidate records associated with a defined event type or supported visual-search query.

General visual activity search is not verified in the reviewed code. Incorrect labels and undetected events limit event-only searches.

Search across live and recorded camera feeds

Live and recorded search is a proposed workflow for reviewing newly received events alongside retained records within a defined camera and time scope.

Unified semantic search is unverified. Live camera playback alone does not supply searchable detections.

Fast retrieval of matching events from large video archives

Archive event retrieval uses retained metadata to locate a matching record and open associated evidence when it remains available.

Archive scale, arbitrary-video indexing and speed are not established. Metadata cannot reconstruct missing clips.

High-activity and low-activity area visualization

Activity-intensity visualization displays relative levels of a defined visual measure across configured areas and time windows.

Relative colors can exaggerate small differences. Missing observations must not be displayed as confirmed zero activity.

Live traffic-flow maps

A live flow map is a proposed view of recently observed movement directions or paths within configured camera coverage.

The reviewed source does not verify a live tracking provider. Occlusions, identity switches and transmission delays can interrupt paths.

Historical movement heatmaps

Historical movement heatmaps aggregate supported observations over a selected past period to compare spatial patterns.

A historical map cannot be recreated from missing observations without a supported reprocessing service. Color changes do not establish causation.

Camera-based analytical views

A camera-based analytical view places supported measurements and event context alongside the relevant camera image.

Displayed overlays are interpretations, not ground truth. Current images and older event annotations must not be presented as the same moment.

Person re-identification across different cameras and locations

Associates candidate person sightings across views using a validated visual matching method.

Cannot establish legal identity, intent or unseen movement.

Historical person and employee search

Searches retained person-related observations within an authorized scope using supported identifiers or criteria.

Cannot retrieve deleted evidence or establish every movement.

Animal Detection

Review visible animal presence and movement near boundaries, sensitive facilities or production areas. Classifying an animal, following its path and deciding how to respond are separate tasks. Species support and visibility must be evaluated for the local environment.