Fault-Level Grading of Photovoltaic Cells Employing Lightweight Deep Learning Models

Table 5

Binary classification results of deep ANN with 0.5 as threshold using original data: (a) GLCM features and (b) LBP features.

		Actual class
		Healthy	Full faulty
(a) GLCM features
Predicted class	Healthy	223	33
	Full faulty	48	90
Classwise accuracy		82.3%	73.2%
Overall accuracy		79.4%
(b) LBP features
Predicted class	Healthy	230	29
	Full faulty	41	94
Classwise accuracy		84.9%	76.4%
Overall accuracy		82.2%