Today more and more shops where goods do not need to pay at the checkout, everything is built on self-service and shopping there’s not a single seller. However, this can attract such stores thieves. Of course, nobody canceled the surveillance system, but it greatly improved, teaching the AI to calculate the offender before he commits theft.
For the development of very interesting meets company Standard Cognition. In the system currently used 27 cameras, which are managed by a special algorithm. It logs almost all the indicators buyers: the speed and length of steps, the direction of gaze and even the goods on which the buyer looks at the moment. The program calculates how long a person is considering labels, monitors the movement of goods throughout the store and so on. But most importantly, from all the information received, the program concludes that the “potential unreliability” of the buyer.
Five of the seven founders of the Standard Cognition previously worked in the Commission on securities and stock exchanges of the USA, where he developed algorithms to detect fraudulent schemes, therefore, need the experience the company has. As stated by one of the leaders of the Standard Cognition Michael Aswel,
“Actually, it’s pretty simple: the potential thieves more than a wide step, they are constantly looking for a way out and this is just a couple of examples of behaviour that indicate bad faith. If you suspect someone in an attempt of theft, it will notify the security guard of the store and he will go to man to clarify the situation.”
The company has already received investments in the amount of 11.2 million U.S. dollars and opened its first test store, which will identify system deficiencies and implemented improvement. By 2020, the company plans to equip its system of not less than 100 shops every day.
This and other news you can discuss in our telegram chat.