To configure object tracker, do as follows:
Configure parameters.
Parameter | Description |
---|---|
Enable | Enable/disable detection tool. |
Record object tracking | If you need to enable recording of metadata, activate this parameter. |
Auto Sensitivity | If you require automatic adjustment of the sensitivity of scene analytic detection tools, activate this parameter. |
Camera position | Еo reduce false alarms rate from a fish-eye camera, you have to position it properly. For other devices, this parameter is not valid. |
Decoder mode | Select a processing resource for decoding video streams. When you select a GPU, a stand-alone graphics card takes priority (when decoding with NVidia NVDEC chips). If there's no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. Otherwise, CPU resources will be used for decoding. |
Max. Object height Max. Object width | Maximum height and width of a detectable object as a percent of the FoV height. The values should be in the range [0,05; 100]. |
Min. Object height Min. Object width | Minimum height and width of a detectable object as a percent of the FoV height. The values should be in the range [0,05; 100]. |
Motion detection sensitivity | Set the sensitivity for motion detection tools, on a scale of 1 to 100. |
Frame size change | Analyzed framed are scaled down to a specified resolution (8, 1280 pixels on the longer side). This is how it works:
|
Antishaker | To correct for camera shake, activate this parameter. This setting is recommended only for cameras that show clear signs of shaking-related image degradation. |
Time of Object in DB | Enter the time interval in seconds, during which object's properties will be stored. If the object leaves and enters the FoV within the specified time, it will be identified as one and the same object (same ID). |
Object calibration | Activare, if you want to configure the perspective. |
Leveling rod height | Enter the height in decimeters of the object you want to find. |
Enable filter | If necessary, configure the neural network filter. The neural network filter processes the results of the tracker and filters out false alarms on complex video images (foliage, glare, etc.). |
Moving object filter mode | Select the processor for the neural network — CPU, one of GPUs or a IntelNCS |
Abandoned object filter mode | Select the processor for the neural network — CPU, one of GPUs or a IntelNCS. |
Moving object filter file | Select a neural network for analyzing moving objects. A neural network filter can be used either only for analyzing moving objects, or only for analyzing abandoned objects. You cannot operate two neural networks simultaneously. |
Abandoned object filter file | Select a neural network for analyzing abandoned objects. |
By default, the entire FoV is a scene analytics zone. If you need to exclude parts of the scene prone to false alarms from detection (leaves, water, etc.), you should set privacy masks − zone with scene analytics and metadata generation disabled. To do so, follow the steps:
The detection tool is now configured.