A Novel Approach for Automatic Detection of Concrete Surface Voids Using Image Texture Analysis and History-Based Adaptive Differential Evolution Optimized Support Vector Machine

Table 1

The training and testing datasets.

Datasets	Sample index	Features							Class label
		F1	F2	F3	…	F106	F107	F108
Training	1	108.115	33.580	−15.236	…	6866.946	11.141	658.625	−1
	2	109.645	33.763	−18.768	…	5026.965	11.007	381.245	−1
	…	…	…	…	…	…	…	…	…
	899	106.063	32.555	−14.066	…	32167.854	10.600	640.701	1
	900	108.305	33.135	−13.449	…	31379.524	10.674	1039.198	1
Testing	1	108.435	32.954	−16.630	…	9425.744	11.802	2016.670	−1
	2	105.565	30.844	−16.665	…	8383.504	11.505	1440.862	−1
	…	…	…	…	…	…	…	…	…
	99	107.228	32.628	−12.087	…	30558.259	10.551	631.624	1
	100	107.223	32.418	−13.006	…	30292.417	10.575	514.372	1