Welcome back everyone for computer vision lecture series.

This is lecture 2 part 2.

In this lecture we are going to talk about colors, what they are, how human vision is

susceptible to colors and how other animals perceive it as well as what it means to capture

color.

Okay, so let's begin.

So what is color really?

Like when you see this image you see red, blue, green and their intersection.

But if I want to just see the red or the green or the blue can I do it?

Yes.

So, you see images have been saved on machines in digital format using three color channels.

And when you independently see each color channel you will see that they are just intensity

values for that particular channel.

And in combination they give a perception of color for us.

This is a traditional notation on how we represent the colored images in MATLAB.

Let's say we have an image of n x m dimension, an RGB image.

In this the first location is the top left pixel value in the R channel.

This is a notation where y pixels down, x pixels to the right basically and the Bth

channel.

So this value will give you the location, this notation will give you the value y pixels

down, x pixels to the right and for the Bth channel.

Similarly this notation will give you the bottom right pixel in the B channel.

B channel is the blue channel in this case.

And if you read an image file it returns an unsigned integer value between 0 to 255.

If you want to convert it you can convert it using this am2double format.

This is basic MATLAB stuff.

If you have background in MATLAB you know what these values mean.

Similarly when you want to look at in a matrix form you will see that these are the rows

along the height, these are the columns along the vertical sorry horizontal direction and

each channel R, G and B are stored in this sequence in MATLAB.

However in OpenCV the notation is reversed for blue and red channels.

Everything else remains the same.

This is important for you because in the exercises we are using python and we are using OpenCV

library and you are going to know and if you have done already the 0th exercise you already

know that this is the standard notation followed by OpenCV library.

Keep in mind the change in order because this has been one of the main problems occurring

or issues for a lot of programmers when they are using different image libraries.

And when you start using a different image library the first thing you have to find out

what is the sequence of color channels that the library stores it.

So what is color really?

I mean when you think of it, it's all we can see right?

But do we all we see is it same for everyone or not?

We want to find it out.

So for that we want to see how our human eye works.

Let's check out the anatomy of our eye.

Here in this image you see that there is an iris which is like a radial muscle which contracts

and expands and controls the light that enters through pupil.

The pupil is this hole that has been formed naturally by contraction and expansion of

the iris.

And together they control the light intensity entering the eye.