Page 194 - Introduction to Autonomous Mobile Robots

Perception

Figure 4.51 179
Two panoramic images and their associated fingerprint sequences [95].

Figure 4.52
Three actual string sequences. The top two are strings extracted by the robot at the same position [95].

in section 4.3.2.1, an adaptive threshold is used to reduce the number of edges. Suppose the
µ
Hough table’s tallies for each candidate vertical line have a mean and a standard devia-
σ
tion . The chosen threshold is simply µ + σ .
Vertical color bands are identified in largely the same way, identifying statistics over the
occurrence of each color, then filtering out all candidate color patches except those with
tallies greater than µ + σ . Figure 4.51 shows two sample panoramic images and their asso-
ciated fingerprints. Note that each fingerprint is converted to an ASCII string representa-
tion.
Just as with histogram distance metrics in the case of image histogramming, we need a
quantifiable measure of the distance between two fingerprint strings. String-matching algo-
rithms are yet another large field of study, with particularly interesting applications today
in the areas of genetics [34]. Note that we may have strings that differ not just in a single
element value, but even in their overall length. For example, figure 4.52 depicts three actual