What is Computer Vision

Reading Time: 2 minutes

Edge Computing and AIoT is an exciting space since it is an opportunity to use technology in new and innovative ways. Yet technology must deliver value to the stakeholders. Helping people and companies understand and unlock the value of the edge and AIoT and helping to deliver solutions that realize this value is what I spend most of my time doing these days.

Computer Vision (often referred to as CV) seeks to have a computer “see” and “understand”, in the same way that human vision does.

It includes:

Components of a Computer Vision System

There are four primary components in a Computer Vision System:

Data acquisition

In a computer vision this is primarily some sort of image capture device, such as a camera, that is able to output a digital representation of what can be seen in the physical world.

The acquired data includes color, brightness, intensity and light scatter.


Transform data so that it can be processed. This often includes changing the format of the data and extracting low-level features. The output is a transformed representation of the data from the data acquisition phase.



The processing phase is where the magic happens and patterns in the data are detected, classified and tracked. The output is high-level features or components of the data.

Patterns include


The postprocessing stage often relates to forming some sort of decision based on the high-level features and/or components of the data. Sometimes this includes further processing of the data.

In summary