What is Unusual Activity Detection?

Avigilon Unusual Activity Detection (UAD) is an advanced AI technology designed to enhance security by identifying and flagging atypical behaviors of specific classified objects, such as people and vehicles. This technology is embedded in Avigilon’s H5A camera line and leverages edge-based intelligence to detect unusual activities based on the type of object and its expected behavior patterns.

Key Features of Avigilon Unusual Activity Detection

Object-Aware Detection:
UAD is capable of recognizing unusual activities by analyzing the behavior of specific objects like people and vehicles. For instance, it can detect if a vehicle is traveling at an unusually high speed or if a person is in an unexpected location.

Advanced AI Technology:
The system uses advanced AI to continuously learn and adapt to the camera’s field of view, ensuring that it can identify and flag unusual behaviors effectively. This learning process allows the system to improve its accuracy over time.

Focus of Attention Interface:
UAD utilizes the Focus of Attention (FOA) interface, which visually highlights unanticipated activities, providing operators with an overview of anomalous events. This helps in quickly verifying and determining the appropriate response to flagged events.

Filtered Recorded Timeline Search:
The technology enables operators to quickly review large amounts of video by filtering the timeline to highlight periods where UAD events have occurred. This significantly reduces the time required to sift through recorded footage.

Integration with Avigilon Control Center (ACC):
UAD is integrated with the Avigilon Control Center (ACC) software, which supports features like timeline search, event search, and rule triggers. This integration enhances the overall effectiveness of the security system.

Comparison with Unusual Motion Detection (UMD):
While UAD focuses on the behavior of specific objects, Unusual Motion Detection (UMD) is designed to detect anomalies in motion within a scene without predefined rules. UMD continuously learns what typical activity looks like and flags deviations from this norm. Both technologies serve different purposes and are suited for different security needs.