Agent Vi’s image processing over IP partitions detection algorithms so it can process video data in both the camera and the central server. This is said to minimise hardware and network bandwidth requirements and reduce installation and operational costs for enterprise-class video analytics.
The processing technology on the edge of the network is called the universal feature extraction. This process is performed by the part of the algorithm that extracts information from the image pixels of incoming images when they are at their highest quality, and when no data has been lost due to image compression. Using a combined local and server approach is said to retain the strengths of both architectures, but without any of their limitations.
The IPoIP feature that differentiates it from the competition in operational cost savings is its ability to update algorithms to all the devices on the network with a single download,” said Frost & Sullivan analyst, Jason Halverson. “Other systems typically require massive upgrades involving manual updates at each device, making this feature a significant cost saver.”
The company’s distribution strategy is a two-tiered approach that focuses on both OEM and integrator customers in vertical markets.