Automated Classification System for Tick-Bite Defect on Leather

Table 1

Comparison of results when employing ANN (A) and XBoost ANN (X) as the classifiers.

Methods	TP		FP		FN		TN		Accuracy		Macro-avg		Weighted-avg
											F1-score		F1-score
	A	X	A	X	A	X	A	X	A	X	A	X	A	X
HoG	331	266	0	65	151	114	0	37	0.69	0.63	0.41	0.52	0.56	0.61
(5,5)
HoG	331	274	0	57	151	48	0	103	0.69	0.78	0.41	0.75	0.56	0.78
(7,7)
HoG	331	281	0	50	151	35	0	116	0.69	0.82	0.41	0.8	0.56	0.83
(9,9)
With HoG	314	313	17	18	50	49	101	102	0.86	0.86	0.83	0.83	0.86	0.86
(5,5)
With HoG	325	325	6	6	21	25	130	126	0.94	0.94	0.93	0.92	0.94	0.93
(7,7)
With HoG	323	323	8	8	41	41	110	110	0.90	0.90	0.87	0.87	0.89	0.89
(9,9)

The bold numbers represent the highest values within the experimental results presented.