CNBC reportsthat an estimated one billion surveillance cameras will be in use worldwide by 2021, which is a massive increase from estimated 770 million surveillance cameras installed in 2019. Artificial intelligence enables new groundbreaking applications for computer vision and is likely to further boost the triumphant progress of smart video technology - especially in “analog” industries like retail. In addition, COVID-19 accelerates the digital transformation trends that have already begun.
At Security & Safety Things we support this exciting development with our open IoT platform for security cameras. Our camera OS brings the latest AI applications built by a growing community of developers from around the world to retail camera systems. The number of camera apps in ourApplication Store designed to take retail business to the next level is growing almost daily.
Brief introduction to Computer Vision & AI
“Computer vision is a field of artificial intelligence that trains computers to interpret and understand the visual world. Using digital images from cameras and videos and deep learning models, machines can accurately identify and classify objects — and then react to what they ‘see.’” - SAS Institute Inc.
The global artificial intelligence (AI) in retail market size is estimated to reach over USD 23 million by 2026, exhibiting a CAGR of 33.7% during the forecast period. The ongoing shift from traditional retail to AI-driven business is one of the key factors driving AI growth in the retail market. However, computer vision affects much more than just video surveillance.
How smart computer vision can take retail to the next level
Smart IoT cameras in stores and malls will help make retail business more efficient and resilient to online competition. AI empowered video analytics turns powerful cameras into business all-rounders that can perform a variety of business tasks beyond video surveillance.
Source: IHS Markit
Tip: Many of the capabilities security cameras need to deliver more than just video you can easily get by installing AI camera apps from our Application Store.
The technology behind: data analytics moves from cloud to the edge
By 2023, there are expected to be around1.3 billion IoT subscriptions. Each new device in IoT increases the amount of data that to be transmitted, stored and processed to add value to private, public and economic life. Machine-generated data will account for over 40% of internet data in 2020, tendency rising so that the centralized internet could reach its limits in the long-term.
Edge computing is a promising technology to overcome the limitations of cloud computing by processing data in the networked devices where it’s generated. Security cameras equipped with powerful processors and AI technology not only deliver video data, but also analyze it and deliver results in one go. As the amount of raw data exchanged between systems decreases, the share of refined data increases and with it overall IoT profitability.
The ROI of smart video technology in retail
Security & Safety Things aims to help retailers get more out of IoT security cameras and thus increase the return on investment (ROI) of video technology. Security cameras that detect gaps in stock and on the shelves in real time, support sales floor optimization with customer insights and help reduce waiting times at checkouts make a significant direct contribution to growing revenue.
Source: IHL Group
By improving customer experience in your store and increasing store profitability, smart camera systems are no longer just cost factors, but become full-fledged profit centers.
Case Study: Kroger uses AI to cut down self-checkout errors
Self-checkout in stores is used worldwide to save resources and time at counters. On the other hand, customers aren’t trained in scanning products as employees are. This leads to errors in billing and friction in checkout operations. That’s why Kroger Co., which is the largest grocery supermarket chain in the United States, uses AI Computer Vision to assist customers with self-service in 2,500 stores.
AI video analytics for object detection enables IoT cameras to identify goods being scanned at checkouts. The system recognizes items and checks whether they have been correctly billed. If an error occurs, the camera system automatically alerts or calls staff for assistance.
In ourApplication Store you find a wide range of camera apps that enable your security cameras to automatically detect products and support customers in self-service. The screenshot below shows how Neurala Brain Builder Checkout automatically detects fruit at a scale. The system has learned to distinguish bananas from apples or other fruits and objects. The apps can be easily installed on security cameras with a few clicks.
Object recognition is just one of many capabilities security cameras can get from AI. Equipped with video analytics for face or motion detection, video technology can help to better understand customer behavior to improve customer experience based on video data.
Learn more about how you can use smart video technology in our blog articles