Table 3.

Performance of the proposed algorithm under different values of $T$.

Dataset	Metric	2	3	4	5	6
Tawsifur	Noise number	63	29	41	38	30
	RE (%)	67.53	85.05	78.87	80.41	84.54
	OA (%)	93.43	99.36	97.86	98.07	99.07
Skytells	Noise number	72	28	26	36	35
	RE (%)	66.51	86.98	87.91	83.26	83.72
	OA (%)	90.95	97.77	97.84	97.18	97.25
CXRs	Noise number	43	73	37	60	48
	RE (%)	75.00	57.56	78.49	65.12	72.09
	OA (%)	84.18	79.10	85.14	81.55	83.92

Note: bold numbers represent the best results. Experimental results of NLA are as follows: Noise number = 194, OA = 71.36 % (Tawsifur dataset); Noise number = 215, OA = 68.66 % (Skytells dataset); Noise number = 172, OA = 60.82 % (CXRs dataset).