Those of you who have been following my blog over the years, will realise that I started flirting with computer vision and robotics research while studying for my PhD at Tokyo Institute of Technology way back in 2008. This was driven by the fact that more than 70 % of my peers were working on robotics, computer vision, gesture recognition and other related topics.

Most of the talk during our meeting used to centre around these topics. Coming from an English speaking country, my Japanese friends also used to ask me to edit a lot of their research papers. I, however, could not abandon my work on Computational Intelligence based algorithms in favour of computer vision or robotics. This was mainly because at that time it did not look like Africa was ready for computer vision and robotics research.

Nevertheless, my interest in robotics and computer vision still lives on. Nowadays, I am doing some research works in these areas and I am also planning to launch a course on robotics at Honors level at my institution.

In a nutshell, computer vision refers to the extraction of information from images in order to make sense of them. This information is usually used as an input to a machine learning algorithm to detect the similarity of the objects to previous object. This cuts out the use of a human being in the detection process.

In robotics, this trends opens up the field of robotic vision, where computer vision techniques enable the robot to capture visual information from its surroundings. By using this information, the robot makes well-informed decisions about its next actions and movements.

In industrial settings, computer vision techniques, commonly known as machine vision, enable businesses to detect abnormalities and defects in their products before shipping them out to customers. In some cases, machine vision has been used for automotive vision systems. These systems provide vision inspection for a wide range of automotive production phases such component manufacture, assembly and production.

As more and more robust hardware systems and sophisticated AI algorithms continue to hit the market, machine vision systems for industrial settings are becoming more accurate and much cheaper. As a result, businesses stand to benefit a lot for this development. Some of these benefits are as follows:

  1. Accuracy: Computer vision systems are characterised by the highest degree of accuracy. Businesses, that invest in these technologies are guaranteed high levels of accuracy thereby guaranteeing higher levels of product quality.
  2. Reduced production costs: Businesses stand to benefit from reduced production costs as computer vision systems enable them to detect faulty processes well in advance. This enables companies to replace those processes before they finally break down thereby reducing periods of down times. Of course, the initial capital required for acquisition computer vision systems might be huge. But the benefits in terms of cost reduction in the long term are greater.
  3. Faster processes: These systems have the capability to carry out repetitive tasks at a much more rapid rate compared to human beings thereby achieving much higher throughput.
  4. Provision of real-time data and statistics: Some recent computer vision systems in industrial setting provide excellent vision interfaces which provide real-time data and statistics to enable production trends.

Computer vision systems have come here to stay. Young people planning their future careers need to be properly guided as some of these jobs that we see today will no longer be there. This is a story for another day.

0 0 votes
Article Rating