Sensormatic has expanded its computer vision portfolio to improve shopper behaviour analytics, detect theft and maintain in-store safety. Part of Johnson Controls, Sensormatic’s computer vision products are designed with proprietary AI algorithms designed for retail store analysis.

Built in partnership with Intel and introduced in 2021, Sensormatic’s computer vision offering are deployed using a smart hub device in conjunction with existing camera infrastructure. It is designed to provide streamlined adoption of AI in retail environments. Computer vision automates tasks and derives meaningful information from video footage in real time. It can address challenges such as customer engagement, stocking, labour allocation, loss prevention, and shopper satisfaction.

Capabilities include real-time shelf sweep detection, monitoring shelf activity for large-scale removal of items so that in-store personnel can take pre-emptive measures to stop theft. Alerts assistants of low stocks and tracks the movement of high-value items that are taken from shelves. It can also monitor loitering, helping retailers prevent or stop organised retail crime events (ORC) by identifying individuals lingering in low-traffic areas after business hours, when criminal activity is more likely to occur. In addition, it can provide group detection alerts, monitoring groups entering or forming within stores and alerting assistants to prevent or stop ORC events.

The system also provides information on shopper behaviours, including traffic pattern insights, audience measurement and dwell time measurement. The system’s traffic pattern analysis observes paths to purchase and shopper movements throughout the floor to allow design of more effective layouts and stocking. The audience measurement features provide insights into shopper demographics and sentiments so that retailers are equipped with the necessary data to tailor marketing plans, promotions and offerings to their particular customer base. Dwell time measurement provides analysis of the length of time customers spend engaging with different displays to help retailers evaluate effectiveness of their campaigns and promotions.

The Sensormatic computer vision system also provides slip-and-fall detection, monitoring sales floors for shoppers who may have injured themselves on the premises. The system also monitors car parking areas to to identify unauthorised or abandoned vehicles and to monitor customer wait times in order to detect potential crime, improve safety and improve customer experiences.

“We’re working side-by-side with incubator customers to identify and design timely solutions that target the most challenging retail problems; our computer vision offering allows us to solve these problems and streamline operations using existing video infrastructure,” said Subramanian Kunchithapatham, chief technology officer at Sensormatic Solutions. “The technology architecture allows retailers to add and remove computer vision analytics capabilities as needed to complement their current solutions. The latest enhancements can help retailers to learn about customer behaviour, mitigate losses and develop better shopping experiences, while also allowing them to look beyond loss prevention for video integration.”

A recent study conducted by Sensormatic Solutions and Coresight found that 43% of retail leaders surveyed currently use AI to analyse in-store video, and a further 31% expect to adopt these solutions in the next 1-3 years.