Hi! Welcome back!
After a few months of lots of work and deadlines, I finally found some quality time to continue with the theoretical posts about digital images. I hope you find it interesting!
Then, today I will be talking about color. It’s a topic with plenty of potential so I would consider this post as an introduction or first contact. Let’s go!
As I mentioned in previous entries (link to post), images are composed by pixels and each one of them represents the basic unit that creates them. This means that each pixel has a limited amount of information about the image in that point-position. When these units are in group you can see all their information together, creating the final image.
In grayscale, the information of every pixel consists in their tone/luminosity, this is, its variation from value=0 (black) to value=255 (white). If we are talking about color pixels they also content information about uniform color, brightness, saturation, …
However, computers and electronic devices store this information in binary code. These digits are “just” 0 and 1 that on its own language, create other numbers as a potential of 2, 8, 16 or 32. Every single 0 or 1 is called a “Bit” and a group of 8 bits is called a “Byte”.
Sounds familiar? 🙂 I’m not going to go deeper into this but it’s very important to know that, depending on the amount of information that every pixel contains we can get more or less color depth in our images. In short, as much information we have, larger color range we can get from our images. This is very good for color transitions and color gradients.
Grayscale photos, also known as “black and white” images, are smaller files since the amount of information they contain focuses on light and doesn’t include any color data. This means that when we transform our photos into b&w version we are “deleting” information from the picture. In fact, this technique is used in order to increase the power and relevance of the lighting in the scene by removing colors from the composition.
Color images provide much more graphic information. In fact, they can be classified depending on the amount of data they content in each pixel. This means that color depth is defined by the bits of information that an image contains. Then, a pixel with lots of bits of information will offer a large amount of colors. In the same way, the file will be bigger since it will need more space for data store.
As we studied in previous posts (link to post), digital images are represented by 3 colors RGB (Red-Green-Blue) in our devices. This really means that 3 different images are overlapped in order to create all visible colours of the picture displayed.
This means the following:
- Each pixel is defined by 3 Bytes: 1 Red + 1 Green + 1 Blue.
- Every one of them has 256 Tones. Remember, from 0 to 255, both included.
- The final image will have 256 x 256 x 256 colors. As a result you will get the well-known 16,7 millions of colors.
- Remember that 1 Bye= 8 Bits. Then, the image will have a 24 bits color depth (8 bits by color channel RGB).
In general, the software dedicated to image edition will allow us to work in 3 different color depths:
- 8 bits x 3 channels = 24 bits in each pixel. 16,7 Millions of colors.
- 16 bits x 3 channels= 48 bits in each pixel. 281 Billions of colors.
- 32 bits x 3 channels= 96 bits in each pixel. This is the HDRI format. It means High Dynamic Range Images.
Depending on the purpose of the images we will need more or less range of colors. We should make the most suitable choice for each case since the file size and its true advantages are not always well-balanced. This means that is up to you to say when it is worth it or not to work with that much color depth.
To sum up we can say that, in general, we use to work with 8 bits most of the time (JPEG format) since they have a good quality and their size is smaller.
On the other hand, the 16 bits files that we tend to work with are usually created by a reflex camera. They produce RAW format images with 12-14-16 bits that will be modified and adjusted in photo-editing processes.
HDR is a 32 bits image created by the digital superposition of a series of images with different levels of exposition (amount of light). However, this is a very simplified explanation about the whole process since I don’t want to make it too complex at this point.
I hope you liked this first part about color. Maybe it has been too technical but it’s a good base for the upcoming content.
Thanks for reading! Talk soon! 🙂