Computer vision and artificial intelligence
Computer vision is a rapidly advancing field of artificial intelligence. The goal of computer vision is to give machines or software the ability to see and recognise objects and actions in images and videos.
Essentially, computer vision is all about enabling machines to “see” the world and extract meaningful information from visual data. This can involve everything from simple image recognition tasks, such as identifying objects in a photo, to more complex tasks like tracking the movements of a person or vehicle in a video stream.
Today, computer vision is used in a wide range of applications, including robotics, security and people & vehicle counting and classification.
Computer vision algorithms must work reliably in a wide range of lighting and weather conditions
When counting and classifying vehicles, AI software analyses camera footage and learns to identify different types of passing vehicle.
Similarly, computer vision is used when counting people. Here it is important to determine which shapes are people, without identifying any individual. The algorithms can learn, for instance, to distinguish between staff and customers, to detect people even if they are mostly obscured from the camera’s vision, to ignore shopping trolleys and pushchairs and to generate demographic data like age and gender.
The people counts are useful for, amongst other things:
- Retailers gaining valuable key performance indicators and learning about their customers
- Smart cities monitoring pedestrian flow to help the city authorities serve people effectively
- Businesses looking to open a new shop or restaurant gain precise figures for footfall past their proposed location to help them assess the potential for their new venture
- Buses counting passengers for real-time occupancy, route planning and fare reconciliation
AI systems are clever enough to detect people and vehicles without identifying them, fully compliant with GDPR.