Huffman coding
An example of the table of Huffman code
| Symbol |
Codeword |
| 0 |
00 |
| 1 |
010 |
| 2 |
011 |
| 3 |
100 |
| 4 |
101 |
| 5 |
110 |
| 6 |
1110 |
| 7 |
11110 |
| 8 |
111110 |
| 9 |
1111110 |
| 10 |
11111110 |
| 11 |
111111110 |
Huffman coding (as well other entropy coding techniques) is not effective for signals with flat probability distribution function. This function is just the image histogram. Since histograms of good-quality images are relatively flat, entropy coders will not perform very well on raw image data. See next interactive example.
To overcome this problem, technique called intraframe prediction is used prior to entropy coding. Prediction will be explained on next pages.
| 