Computer Vision
Computer vision is the branch of AI that interprets images and video. It lets machines identify objects, read text in a photograph, detect faces, spot defects on a production line, or guide a self-driving car. Where a camera simply records light, computer vision works out what that light actually shows, turning raw pixels into useful understanding of a scene.
To a computer, an image is nothing more than a grid of numbers describing the colour and brightness of each pixel. There is no built-in notion of “dog”, “road sign”, or “tumour”. Computer vision is the effort to bridge that gap, taking those bare numbers and working out the meaningful things a human would notice at a glance. It is a task our brains perform effortlessly and constantly, which is part of why it took computers so long to get good at it.
The breakthrough came from deep learning, using many-layered neural networks that learn visual patterns from huge collections of labelled images. Early layers pick out simple features such as edges and textures, later layers combine these into shapes and parts, and the top of the network recognises whole objects. Shown enough examples of cats, the system learns for itself what makes a cat look like a cat, without anyone writing a rule.
Computer vision now sits behind an enormous range of everyday and industrial technology: unlocking a phone with your face, checking passports at the airport, reading number plates, analysing medical scans, and helping robots and vehicles navigate the world. In the UK, it underpins work across healthcare, security, agriculture, and manufacturing, and it is one of the most commercially active corners of AI we cover.