According to Wikipedia, Computer vision is an interdisciplinary scientific field that deals with how computers can gain a high-level understanding from digital images or videos. From the perspective of engineering, it seeks to understand and automate tasks that the human visual system can do. Computer vision means when we consider image-based information (even video) as input data, and we start something with that data. This can be data collection and evaluation, but very often by analyzing the image, we put a computer in front of some decision situation, and the decision results in some mechanical solution.
Elements of computer vision
1. Imaging – this is typically a camera with the right accessories (microscope, ultrasound, LIDAR, etc.)
2. Signal processing unit – target hardware used to process visual data, such as a video card or similar device
3. Software – special programs required for the data interpretation phase. This is the soul of the process.
4. Communication – user interface (UI) that can be used to display and diagnose the process.
What can Computer Vision be used for?
What approach board do I see? Pedestrian recognition, dynamic sensing of distances. It is clear that self-driving cars are perhaps the most well-known problem in the field of computer vision today.
Computer Vision is used to evaluate and check for quality control. Is the diameter of the hole exactly what it should be? Is the drug mixture or soft drink exactly the colour it should be? These are simple “yes/no” decisions that allow a decision to be made based on a particular input image.
Realistic modelling and reconstruction of objects and buildings in a 3D virtual space are possible automatically through Computer Vision.
Computer vision answers questions like; Is the box at the right angle on the assembly line? Does the automatic robot arm put the part in the right place?
Traffic counting on the roads, entry counting in the office building or supermarket, making a heat map based on the routes travelled.