Computer Vision: A Promising Technology Of Future

We capture everyday beautiful moments in our cameras. Long back, we used to make pictures and statues as a mark of memory, but science and technology have pushed us to next-generation technology, where capturing a moment is just about a matter of seconds.

Older day cameras used to come up with reels which need to be reconstructed from negatives, but today digital cameras come up with photosensitive diodes and receivers which are very flexible and highly responsive.

Like normal cameras, a digital camera works on the same principle of photosensitive sensors which convert optical input to electrical output. Every image is a superimposed matrix of R, G, and B pixel matrix. And based on those matrix values, illumination takes place which shows up as a digital formed image.

After going through basics of a digital image let's see how computer vision works and how is it a blessing for mankind.

Computer Vision is the smart algorithmic operation of these digital matrices in order to obtain desired outcomes. For example identification of a ball, or counting the total number of coins and their net worth or reading a barcode and what not! Computer vision is everywhere.

Though training a computer to perform a particular visionary task can be hard, but they perceive the world in a very similar way we do. In fact, computers can see invisible things which humans can't using infrared technology. They can compute the data at a much faster rate, they can focus on more than one point at a time and so many other amazing features.

Today computer vision is extensively used in defense, robotics, medical, automation and lot more. Programming languages like python, c++ and Matlab are one of the leading providers embedded with computer vision libraries and toolboxes to help you understand the subject well and implement accordingly to your application.

Let's take a very simple example of how to implement a computer vision algorithm to identify a green ball in a video feed.

Just imagine yourself as standing next to a place and you're asked to detect a green ball. What will you do? The first thing which may come to your mind is that you have to look for a round object once you spot all the round object then you will try to find a green round object which probably fits the best optical description of a ball isn't it?

In the above image as you can see, I used a colour filter which operates under the green range. Since there is so much greenery out there so in order to ensure only the ball is selected a circle filter is also applied. Now any portion of the image which meets these two parameters would be detected as a ball and so are the results.

To our conclusion, we can say that computer vision is a highly disruptive technology with lots of application in real-world problems. It is a nice skill to learn and can be implemented to innovatively interpret many unsolved issues.

Image Source: Google Images