An Approach to Pipe Image Interpretation Based Condition Assessment for Automatic Pipe Inspection

Table 2

Classifier test results.

Feature set	No. of training cases	No of testing cases	% accuracy of testing	% accuracy on good_1.bmp	% accuracy on good_2.bmp
RGB	3000	17867	96.88	99.15	99.93
HSB	1000	19867	97.30	99.39	99.98
Gabor	5000	15867	97.11	99.25	99.92
Window	3999	10100	99.83	99.98	100.00
HS	155000	19867	67.81	64.11	92.61