When we talk about imaging systems, we mean using cameras to see, identify, analyse and understand images. A machine vision approach can be applied to any visual content: images, video – anything with pixels.
In the same way that a sensor system is more than just sensors, an imaging system is more than just a device that captures an image.
It encompasses a range of overlapping techniques and technologies – software and hardware, integrated systems, data analytics and engineering expertise.
The integration of these elements allows us to analyse images in detail and extract the most important information to solve multi-layered challenges.
The computational processing of images has been around since the 1960s, but in recent years, advances in electronics, data storage and cloud computing have made machine vision pretty much accessible to everyone.
One of the reasons for this is that machine learning (the ability of systems to automatically learn and improve from experience without explicitly being programmed) has created new opportunities.
Additionally, the market is awash with affordable, high-quality camera devices that, coupled with tiny credit card-sized computers like the Raspberry Pi and the right engineering expertise, can generate incredible detail at very low cost.
Machine vision is probably most well known as the key element of autonomous technologies such as those used in surveillance drones or robotic manipulators in factory assembly.
However, the accessibility of lower cost systems is now driving a huge range of applications outside these specific areas and now object recognition, object tracking, image classification or 3D reconstruction can enhance our understanding of an environment (usually one that is remote to us), so we can improve quality, eliminate error and reduce costs.
Navigation, route planning, obstacle detection, traffic sign detection
The human face
Accessing personal devices, biometrics, photo tagging
Gesture recognition and control, vision-based user inputs, eye tracking
Industrial automation, robot guidance, package inspection
3D visualisation and animation, 3D image segmentation, tumour detection
Crop detection, atmospheric visibility, pollution monitoring
Object detection, people counting, crime prevention, home security
Pedestrian detection, driver monitoring, parking assistance, road toll charging