Page 127 - Digital Analysis of Remotely Sensed Imagery
P. 127

98    Cha pte r  T h ree

               codes can be coded using variable-length encoding to further reduce
               data volume.
                   Run-length encoding can be extended to two-dimensional (2D)
               binary images using various coding procedures, one of which is
               known as relative-address coding. In this coding the binary transitions
               that begin and end each pixel are tracked. The simplest way of
               achieving this is to note the beginning and end positions of a nonvoid
               pixel in a row by row fashion. The binary image shown in Fig. 3.8 is
               coded as below:
                   Row 4        4, 6
                   Row 5        2, 6
                   Row 6        1, 8
                   Row 7        1, 10
                   Row 8        3, 12
                   Row 9        3, 12
                   Row 10       5, 8
                   Row 11       5, 6
                   Row 12       5, 6

                   As shown in the above diagram, a large compression ratio is
               achieved if the binary image extends far in a row. Instead of recoding
               all the pixels, only a pair of coordinates is needed for the presentation
               of each row of image. This compression ratio may be further improved
               by combining different coding methods (e.g., Huffman and run-length)
               in one compression.

                             1  2  3  4  5  6  7  8  9 101112
                           1
                           2
                           3
                           4
                           5
                           6
                           7
                           8
                           9
                          10
                          11
                          12

               FIGURE 3.8  Two-dimensional run-length coding is a useful way of reducing
               remote sensing data volume, even though its most common application is in
               compressing fax fi les.
   122   123   124   125   126   127   128   129   130   131   132