Region Splitting:

First, the input image is split into four regions R1, R2, R3, and R4. This is illustrated in fig.b where only two homogeneous regions are identified which are R1 and R4. Hence, the region R1 and R4 are not distributed. The region to be splitted further are R2 and R3. The quadtree partitioning of the regions R2 and R3 is shown in fig. c. After this step, the homogeneous regions are merged together to get the result which is shown in fig. d. The quadtree representation of the procedure is shown in fig. e.

The pixel representation of given image can be give as,