Versions Compared

Old Version 1

changes.mady.by.user Alina Luchkina

Saved on

compared with

New Version 2

changes.mady.by.user Alena Kniazeva

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Tip

Hardware requirements for neural analytics operation

General information on metadata

The standard detection sub-tools are triggered by certain actions of objects in the camera FOV:

Name of the detection sub-toolDescription
Line CrossingDetection tool triggers when an object track crosses a virtual line
Loitering In AreaDetection tool triggers when an object stays in a specified area for a long period of time. The triggering stops when an object leaves the area or disappears inside the area
Stop In AreaDetection tool triggers when an object stops moving in a specified area. The triggering stops when an object leaves the area or disappears inside the area
Abandoned ObjectDetection tool triggers when there is an abandoned object in a specified area
Motion In AreaDetection tool triggers when an object starts moving in a specified area. The triggering stops when an object stops in the area, leaves the area, or disappears inside the area
Entrance In AreaDetection tool triggers when an object enters a specified area
Exit From AreaDetection tool triggers when an object leaves a specified area
Multiple ObjectsDetection tool triggers when the number of objects in a specified area exceeds the set value
Move From Area To AreaDetection tool triggers when an object leaves one area and enters another area of the frame

Tooperate the standard detection sub-tools, you can use the metadata from the sources:

  1. Category: Trackers:
    1. Object Tracker.
  2. Category: Face (reduced set of detection sub-tools: Line Crossing, Loitering In Area, Entrance In Area):
    1. Face Detection VA.
    2. Face Detection TV.
    3. Face Detection VL.
  3. Category: Vehicle:
    1. License Plate Recognition BRS.
    2. License Plate Recognition IV.
    3. License Plate Recognition RR.
    4. License Plate Recognition RR—Search In Archive.
    5. License Plate Recognition RR—Parking.
    6. License plate recognition (VT).
    7. Vehicle Recognition RR.
  4. Category: Poses:
    1. Pose Detection.
  5. Category: Production Safety:
    1. Equipment detection tool (PPE).
    2. PPE detection (VL).
  6. Category: Retail:
    1. Barcode Detection.

The standard detection sub-tools can operate on the basis of all of the above objects

Axxon One uses two different trackers for moving objects detection and metadata calculation:

  1. Object Tracker is a primary tool. 
  2. Neural Tracker is essentially the same but operates through neural networks. The neural one is more accurate and detects even static objects, but it requires more computing resources.

The Scene Analytics detection tools work with both trackers.

Note
titleAttention!

The abandoned objects detection tool works only with the primary tracker.

When created, both Object Tracker and Neural Tracker objects are enabled by default. Tracked objects' parameters (relative width and height, color) are displayed in the camera window.

Info
titleNote

Up You can track up to 25 objects can be tracked at the same time.