Quaternary tree

Implementation

• The most important part is the correct reading of the binary file and the correct transposition of the pixels in a matrix dynamically allocated.
• As a way of working, we need a main function processing that calculates the similarity score
  and compares it with a certain threshold.
• Then, calls are made according to the function (recursively) as many times as needed to divide the initial matrix in 4 quarters and the information is stored each time in a tree.
• A function that transposes the elements from one tree to another is also necessary
  vector and vice versa, the implementation consists of recursive calls.
• I did too 2 functions for specific writing of output files, and a function
  which calculates the average value of a color channel.