pFlux Technology

 

pFlux is a state-of-the-art object detection and tracking capability performant regardless of camera view or scene type. By deemphasizing irrelevant objects and spurious background movement found in scene variations, pFlux focuses only on the objects and events of interest. This results in more effective attention cueing for security and surveillance operators (Security Solutions) and improved modeling for AI/ML engines.

 

Solutions built on pFlux are mathematically driven, making them flexible, edge-compatible and incredibly fast. The underlying algorithms screen out background noise stemming from ambient environmental motion, occlusions, glare, reflections, camera movement, etc., resulting in fewer false positives and reduced nuisance alarms. 

Complex video

pFlux eliminates background noise and nuisance motion for efficient object and event detection. In the example to the left, the pFlux output, represented in the lower right (labeled FTSG), reduces foreground nuisance motion, isolating objects and motion associated with vehicles.   

Weather

Nuisance motion caused by weather negatively affects machine and human-based video surveillance operations. From left to right in the example to the right is raw RGB video, the pFlux video analysis and an overlay of the analysis on the raw video. Human objects are clearly identified and isolated from ambient motion related to falling snow.

 

Light and reflection

Surveillance video with light and reflection can create distractions, obscure objects and confuse AI/ML models. To the left, pFlux isolates objects of interest (vehicles) effectively even as these objects move through glare and reflection.  

Privacy (PII/GDPR)

The pFlux output supplements video content with spatial and temporal analysis that is compliant with PII/GDPR/CCPA privacy standards. The image to the right is a colorized output of pFlux, representing objects in motion (blue) differentiated from persistent objects (red).