Automatic Recognition of Asphalt Pavement Cracks Based on Image Processing and Machine Learning Approaches: A Comparative Study on Classifier Performance

Table 3

Prediction results of LSSVM with different value of feature number.

Statistics	CAR (%)	Number of features
		40	80	100	200	400
Average	AC	95.33	96.72	96.94	95.22	96.33
	DC	94.00	94.89	92.67	90.83	90.56
	LC	87.67	86.56	84.39	86.17	85.56
	NC	91.17	90.17	90.39	89.83	89.33
	TC	94.94	94.28	92.83	93.83	92.61
	Overall	92.62	92.52	91.44	91.18	90.88
Standard Deviation (Std.)	AC	3.46	2.29	2.15	2.43	1.88
	DC	3.02	3.58	3.38	3.78	3.64
	LC	4.48	4.79	5.00	4.80	4.88
	NC	3.56	3.29	3.57	3.37	4.21
	TC	2.08	3.78	3.79	2.44	3.71
	Overall	1.46	1.71	1.46	1.46	1.75