Despite a surge in online shopping, brick-and-mortar outlets across the United States are still thriving. Meanwhile, physical stores and virtual counterparts both face their very own challenges. For instance, While their virtual counterparts face fears of cyberattacks and hacking. Whereas, physical stores face the challenging task of preventing customer theft, shoplifting and refund frauds.
It is alarming how much shoplifting and theft costs the retail industry each year. It is one of the largest contributors of inventory shrinkage and loss. Many retailers are enhancing budget allocation for security systems with investments in technology increasing the most.
Retailors Lose Billions from Shoplifting
According to the 28th annual 2019 National Retail Security Survey in the United States conducted by the National Retail Federation in collaboration with Dr. Richard Hollinger of the University of Florida. The shrink rate also known as the rate of inventory loss. Due to various reasons such as shoplifting, theft, error, or fraud — has remained stable over the past few years. A loss of $50.6 billion is revealed for the sector. Although, the average shrink rate viewed in terms of value is 1.38%.
68.2% of professionals surveyed across major US retail chains with loss prevention teams responded that they would allocate additional resources for loss prevention, mostly in technology.
Retailers’ Challenges in Spotting Shoplifters
Traditionally, security staff undergo training to recall faces inside physical stores. Unfortunately, a limit exists for recognising and processing of number of faces for future reference. For instance, imagine all the new faces appearing during a rush hour or day. Additionally, changes in appearance over time, including wigs, beards, caps, hair colour or glasses can easily trick a person’s ability to process and recognise a face during the short time it makes an appearance at the store. Nowadays, pick pockets resemble genuine prospects making things much difficult.
A study conducted by a research team at the University of York attempted to answer the question, “How many faces do people know?”. Did you know a person recalls over 5,000 faces including that of famous people. However, the study found that there is a significant difference between identifying a known face and a face that is never seen before. Additionally, A known face is easily identifiable even with a slight change to the image. But, an unseen face could easily be disrupted.
Computer Vision Helps Efficient Face Recognition
Leveraging of Technology is perfect for such instances. Because, larger number of faces are identifiable with greater efficiency and accuracy. The aim of computer vision technology is automating and replicating the cognitive processes of the human visual system. Having obtained information from images and videos, computer vision uses machine learning to train computers to process and analyse patterns across faces.
As a result, retailers are able to spot suspicious persons in their stores, even if people are not looking straight at security cameras or wearing glasses, growing beards or changing hairstyles. Though this raises several privacy and ethical questions, the technology’s capability is racing far ahead of societal norms and governmental policies.
Through computer vision, retail chains can receive real-time notifications of suspicious persons or behaviours. Through short video clips, retrospective reproductions of actions of identified behaviours can be captured on demand. This solution will allow retail chains to maintain a central database with images and videos of known offenders and deploy them at all stores. An investment in this technology will significantly improve the process of efficiently and proactively discovering potential offenders and create a safer environment for both employees and store customers.
