# Introduction to Computer Vision

There is some strong misconception about computer vision or CV as it is very simple and it is very trivially solved by people, even very young children. A human being can easily pick a specific item from cluttered objects and why can’t a robot? Human being has an innate nature to identify things, but a camera doesn’t have the ability. the camera is not as efficient to identified colors as compare with our eyes. Even after decades of research, we are not meeting at least in terms of human vision.

“If you want a machine to think, You need to teach them to see.” There is the importance of CV. So I am considering this blog as an introduction to computer vision and its prominence in different industrial sectors.

### Why Computer Vision and Image Processing are correlated to each other?

As you all know, an image is a collection of pixels and depending upon storing the information to a computer memory it varies to 8 bit, 16bit or 24bit images. Each pixel can store a piece of information from 0 to 255[ie.. 20-1 to 2n-1, n=8] if it is an 8bit image. An image can be a single channel or multi-channel images. Multi-channel images will be color images and even color images can be of different formats such as RGB, RGBX, HSV, YUV, YCbCr, etc.

Each color formats are important and depending upon the applications color formats are chosen. For Eg: 8bit RGB image is a multi-channel image consisting of 3 Channels Red, Green, and Blue. Each channel has an 8-bit image and the values will be ranging from 0 to 255 and it depends upon the intensity of the three colors and every other color is formed by the combination of Red, Green and Blue values. So if you need to extract a red ball from a cluttered image, you can easily choose red channel [here, Red channel is our composite image] and the objects with red color will be having higher intensity numbers. As RGB is a highly correlated image it is not much recommended for separating color and HSV format can be used where H refers to Hue and it represents the color and each H value represents different colors.

There are many other color formats and different formats are chosen depending upon the application. As color image processing itself is a vast topic and it is one of the most important pre-processing techniques to select a good composite image and then localization of the object or the extraction can be much easier.

There will be future blogs on different color formats, color conversion methods and how it is applied on different applications.

#### Applications of Computer Vision

The application of CV varies to different industries such as

1. Optical character recognition is used in the number plate recognition system.
2. Optical character recognition also helps in the banking sector to reduce human intervention while reading a document or a cheque.
3. Security surveillance such as face recognition, crowd counting.
4. Camera-based CV systems play a major role in Advanced Driving Assistance System [ADAS].
5. In the manufacturing Industry, Finding the defect during the production line using a camera improves the production of the industry.
6. There are many other applications that we use daily in our mobiles such as the filters that are available on Instagram, Snapchat and our mobile devices such as tint, vivid, B/W filters are basically an image processing method to enhance the images.

So as you can see computer vision is happening everywhere around you. In this post, I have briefed a gentle introduction in the field of computer vision.