-
Notifications
You must be signed in to change notification settings - Fork 0
/
README
31 lines (16 loc) · 1.83 KB
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
The first and most obvious relation between the different qualities of an image is of storage of pixels and image quality.
All images on a screen are nothing but a collection of pixels. The image only becomes recognizable when vast amounts of
pixels are used to convey it; when there's so few pixels you can count them on your fingers, you're essentially just looking
at a small collection of colored cubes. Since each pixel requires storage, that means the more storage is used to store
pixels, the higher the image quality. However, performance must always be taken into account when it comes to computing
systems. Speed is important, and constantly lugging around so much data in high resolution images, speed degrades. This
is what makes compression, the grouping of pixels so that they collectively take up less space, so important.
The lower the estimated compressed storage becomes relative to the estimated original storage, the more unrecognizable
the original image becomes. The closer it comes to the original storage, the more recognizable the image becomes. This
is the relation between compression size and image quality. In the quadTree program, this can be better visualized by
the ratio of compressed to uncompressed data: as the ratio increases, the image begins looking more like it's original
form, while slowly degrading into a small pile of pixels as the ratio decreases. This also leads to the relation between
the amount of quad tree nodes. Each leaf node stores a single pixel value, and data relating to what portion of the image
that pixel value colors. The less nodes there are, the less seperate pixel values make up the image as a whole, but also
the less storage the image takes up. Compresssion is excellent for performance, with higher intensity leading to less
storage being used but image quality being decreased.