Atlas Knowledge Base requires authentication to access.
Dashboard
Image Classifier Options
Recent Pages
Image Classifier OptionsSBN-Media/Media-Processing/image-classifier-optionsjust now
Saved Pages
No saved pages. Click the star on any tab to save it here.
My Collections
No collections yet. Create one to group related pages together.
Image Classifier Options
The Image Classifier is configured in three layers. Service settings under imageclassifier control the service as a whole. Profiles name a model and its detection settings. Groups apply a profile to a set of cameras, add detection areas, and override profile settings where a camera set needs different behaviour. Set overrides in sbn-media.local.yaml, view the current values with ./sbn-media config view -a imageclassifier, and apply changes with ./sbn-media config reload.
How the layers combine
A camera is assigned a classifier group in its device configuration. When the camera is classified, the group's profile is loaded and every setting the group sets replaces the profile's value. Per-class settings (confidence thresholds, motion velocities, class priorities) merge key by key, so a group can change one class and keep the rest from the profile. A camera with no group uses the service settings.
Groups can also be edited from SBN Anywhere with Edit group in the device configuration window. Saved edits stay a draft until published; a published edit is layered on top of the group in the configuration file and takes effect without a reload.
Service settings
Turning classification on
Setting
Default
Description
imageclassifier.enabled
false
Classify attached streams automatically. On-request classification works regardless.
imageclassifier.devices
(none)
Device types to attach to automatically, for example camera and webrtc. An empty list turns automatic attachment off.
imageclassifier.maxConcurrent
4
Streams one instance classifies at once before other instances take the rest. 0 means unlimited.
imageclassifier.activeGroups
(all)
Groups this instance handles for on-request classification. Empty handles every group.
Model and runtime
Setting
Default
Description
imageclassifier.model.path
./models/yolov8n.onnx
Detection model used when no profile applies.
imageclassifier.model.numClasses
80
Number of classes the model detects.
imageclassifier.model.classNames
(COCO names)
Custom class names for a custom-trained model.
imageclassifier.backend
onnx
Inference backend.
imageclassifier.execution.provider
cpu
Where inference runs: cpu, cuda (NVIDIA GPU), tensorrt, openvino, coreml or directml.
imageclassifier.install.dir
./models
Where the AI runtime and downloaded models are kept.
imageclassifier.models
(built-in list)
Models available to ./sbn-media classifier setup.
imageclassifier.model.ftp.*
(set by installer)
Server, credentials and path for downloading prepared models.
Detection
Setting
Default
Description
imageclassifier.confidence.threshold
0.5
Minimum confidence (0-1) for a detection to be reported.
imageclassifier.classes
(all)
Object classes to report. Empty reports every class the model knows.
imageclassifier.nmsThreshold
0.45
Overlap (0-1) above which two boxes for the same class are merged into one.
imageclassifier.frame.interval
5
Classify every Nth frame. 1 classifies every frame.
imageclassifier.frame.ttl
5m
How long frame images from video classification are kept.
imageclassifier.tracking
false
Track objects across frames, giving each a stable identity and direction of movement.
imageclassifier.defaultMotionVelocity
(none)
Speed (0-1 of the frame per frame) above which a tracked object counts as moving.
imageclassifier.publish.changes_only
false
Publish results only when the detections change.
imageclassifier.publish.change_iou_threshold
0.5
Box overlap (0-1) at which two detections count as the same object for change detection.
Always-on classification
Cameras with history enabled can be classified continuously, with or without a viewer.
Setting
Default
Description
imageclassifier.alwaysOn.enabled
false
Classify history-enabled cameras with no viewer attached.
imageclassifier.alwaysOn.maxWorkers
2
Cameras classified this way per instance. They count within maxConcurrent.
imageclassifier.alwaysOn.frameInterval
25
Classify every Nth frame on these cameras.
Alarm classification
These settings classify the video for an alarm, for cameras whose alarm binding has analytics enabled.
Setting
Default
Description
imageclassifier.mediaReady.enabled
false
Classify the clip stored for the alarm once it is available.
imageclassifier.mediaReady.tenants
(none)
Tenants whose stored alarm clips this instance handles.
imageclassifier.alarmStreams.max
4
Alarm video spans one instance classifies at once.
imageclassifier.alarmSignals.enabled
false
Send a signal to the SBN alarm log for each final classification of an alarm, on the camera's bound alarm account and zone.
imageclassifier.alarmSignals.codes
person: BA
Alarm code for each detected class, as class and code pairs.
imageclassifier.alarmSignals.defaultCode
UA
Alarm code for a detected class with no entry in codes.
imageclassifier.alarmSignals.failedCode
UX
Alarm code sent when a classification fails or is inconclusive.
imageclassifier.alarmSignals.describedCode
UX
Alarm code of the alarm-log entry that carries an AI description. The entry goes to the camera's bound alarm account and zone.
imageclassifier.alarmSignals.latchTTL
24h
How long an alarm is remembered as signalled; a repeated delivery within this time sends nothing.
Detection events
Setting
Default
Description
imageclassifier.detectionEvents.enabled
false
Publish one detection event per tracked object that passes its group's filters.
imageclassifier.detectionEvents.pictures.enabled
false
Publish a set of pictures for each reported object when it leaves view. Enable only after every recorder and API proxy runs a release that supports it.
imageclassifier.detectionEvents.pictures.max
8
Pictures per object, 1 to 8.
Re-identification
Re-identification attaches an embedding to each object's best picture, which powers the cross-camera "find similar" feature. Install an embedder first, for example ./sbn-media classifier setup dinov2-small.
Setting
Default
Description
imageclassifier.reid.models
(none)
Named embedder models, each a name and a model file path.
imageclassifier.reid.default.model
(none)
Embedder used when a group does not choose one.
imageclassifier.reid.model.path
(none)
Single embedder model file, for sites that do not use reid.models.
Camera analytics
Setting
Default
Description
imageclassifier.onvif.producer.enabled
false
Report detections made by the camera itself (ONVIF analytics) alongside the classifier's own.
imageclassifier.onvif.metadata.log
off
Diagnostic logging of the camera's analytics: off, detections, all or raw. all writes thousands of lines an hour.
Profile settings
Profiles are defined under imageclassifier.profiles, each under its own name. A profile names a model and the detection settings that go with it.
Model
Setting
Default
Description
modelPath
(required)
Model file.
modelType
(required)
Model family: yolov8, yolov11, yolox, rfdetr or dfine.
backend
onnx
Inference backend.
executionProvider
(service setting)
Where inference runs for this profile.
inputWidth, inputHeight
640
Model input size in pixels.
numClasses
80
Number of classes the model detects.
classNames
(COCO names)
Custom class names.
Detection
Setting
Default
Description
confidenceThreshold
(required)
Minimum confidence (0-1) for a detection to be reported.
classConfidenceThresholds
(none)
Per-class thresholds, for example person: 0.3.
classes
(all)
Object classes to report.
nmsThreshold
0.45
Overlap at which two boxes of the same class are merged.
preprocessing
(none)
Image stages applied before detection. clahe improves contrast in dark or washed-out scenes.
claheClipLimit
2.0
Strength of the clahe stage.
deepDetection
false
Split moving regions into overlapping tiles and classify each tile, which finds small and distant objects at extra CPU cost.
deepDetectionOverlap
0.25
Overlap between tiles (0-1).
deepDetectionPadding
0.1
Margin added around each moving region (0-1).
Frame rate and motion
Setting
Default
Description
frameInterval
(service setting)
Classify every Nth frame.
adaptiveInterval
false
Classify more often while objects are present and less often when the scene is idle.
intervalMin
2
Frame interval while objects are present.
intervalMax
30
Frame interval while the scene is idle.
motionGate
false
Skip classification of frames that show no change from the previous one.
motionThreshold
0.01
Amount of change (0-1) that counts as motion.
motionDownsample
64
Grid size used to compare frames. Smaller is faster and less sensitive.
Tracking and reporting
Setting
Default
Description
tracking
(service setting)
Track objects across frames for stable identities and direction of movement.
minHitsToConfirm
2
Consecutive detections before a tracked object is reported.
defaultMotionVelocity
(service setting)
Speed above which a tracked object counts as moving.
classMotionVelocities
(none)
Per-class speeds, for example person: 0.005 and car: 0.02.
publishChangesOnly
(service setting)
Publish results only when the detections change.
changeIoUThreshold
(service setting)
Box overlap at which two detections count as the same object.
priority
50
Scheduling priority, 0-99. Higher priority cameras are classified first when the host is overloaded.
classPriorities
(none)
Display order of classes. A lower number is shown first.
staticPriorityPenalty
0
Added to the display priority of objects that are not moving.
describeClasses
(all)
Classes that trigger an automatic AI description of the detection. Empty describes every detection.
Group settings
Groups are defined under imageclassifier.groups, each under its own name.
Setting
Description
profile
Name of the profile the group uses. Required.
description
Free-text description of the group.
detectionAreas
Areas of the frame to watch or ignore (see below). A camera's own saved areas take precedence.
includeStaticDetections
Report objects that are not moving. Off by default: with tracking on, only moving objects are reported.
alwaysIncludeClasses
Classes reported even when not moving, for example person.
stopOnClass
Stop classifying an alarm clip at the first frame in the alarm window that contains this class. Empty classifies every frame.
reidDefaultModel
Embedder for this group's cameras. Empty uses imageclassifier.reid.default.model.
reidModelByClass
Embedder per class. An empty name turns re-identification off for that class.
reidDefaultThresholds
Match cut-offs for "find similar": badge (default 0.15) for the badge and panel (default 0.30) for the matches panel. Lower is stricter.
reidThresholdsByClass
Match cut-offs per class.
describe
AI description of this group's alarm clips (see AI description settings).
A group can also set any of these profile settings to override them: confidenceThreshold, classConfidenceThresholds, frameInterval, adaptiveInterval, intervalMin, intervalMax, motionGate, motionThreshold, motionDownsample, publishChangesOnly, changeIoUThreshold, tracking, minHitsToConfirm, deepDetection, deepDetectionOverlap, deepDetectionPadding, preprocessing, claheClipLimit, priority, classPriorities, staticPriorityPenalty, defaultMotionVelocity, classMotionVelocities and describeClasses.
AI description settings
The describe block of a group sets how its alarm clips are described. Which detected classes trigger a description is set by describeClasses.
Field
Default
Description
enabled
false
Describe this group's classified alarm clips.
image
crop
crop sends up to three crops of the detected objects from the best frame. scene sends the whole frame.
prompt
(AI LLM default)
Instructions given to the model with the image.
provider
(AI LLM default)
AI provider used for the description.
model
(AI LLM default)
Model used for the description.
Detection areas
Each area is a polygon of at least three points, in coordinates from 0 to 1 across the frame (0,0 is the top-left corner).
Field
Description
name
Name of the area.
points
Corners of the polygon, each with x and y. The last point joins back to the first.
mode
include (default) reports detections inside the area. exclude ignores detections inside it.
classes
Classes the area applies to. Empty applies to every class.