Soon, smart CCTV that knows when you need shopping advicePublished On: Sun, Oct 21st, 2012 | Computer Science | By BioNews
CCTV security cameras, which are a familiar fixture in shops, could soon be used to track and analyse your browsing habits as well.
Mirela Popa and colleagues at Delft University of Technology, Netherlands, are developing a software that can automatically categorise shoppers’ behaviour using video footage from the fisheye cameras.
For example – the system can tell if customers appear to be confused, are looking around for a specific product, or are heading decisively towards a particular section.
According to Popa, who hopes the technology will lead to better customer service, the aim of the software is to help customers and increase retailers’ profits.
“This could be extremely valuable in terms of how to judge when a customer needs help,” New Scientist quoted Patrick O’Brien of retail analysts Verdict Research in London, UK, as saying.
“In a busy shop it’s difficult to know who it would be most productive to help next,” he said.
By installing additional cameras at eye level among shop displays, Popa’s system can also build up a detailed picture of how shoppers interact with particular products, thereby providing instant feedback to retail analysts.
The technology is based on motion-detection algorithms that track people’s movements and then learns patterns based on what it has observed.
This information can also be used to predict future customers’ behaviour, Popa says.
The team conducted an early trial in a supermarket in Netherlands and, according to Popa, people have been positive so far.
“I think that life is changing,” she said.
“You can’t do anything about that,” she added.
The findings of the study will be presented at the International Conference on Image Processing in Orlando, Florida, this month.