Table 1.

Performance analysis of test data using traditional machine learning technique with and without image pre-processing (Technique 1).

ML models	Without image pre-processing						With image pre-processing
	Accu.	Prec	Rec.	Spec	F1-s	Time	Accu.	Prec.	Rec.	Spec	F1-s	Time
Logistic regression	0.739	0.739	0.74	0.73	0.74	5.85	0.756	0.754	0.76	0.74	0.76	4.3
SVM	0.762	0.762	0.76	0.76	0.74	370.7	0.807	0.808	0.81	0.79	0.80	268.2
Decision tree	0.714	0.715	0.72	0.71	0.71	24.8	0.737	0.773	0.73	0.74	0.75	10.4
KNN	0.718	0.719	0.72	0.69	0.72	13.9	0.849	0.870	0.85	0.84	0.85	9.1
Naive Bayes	0.714	0.716	0.71	0.70	0.72	6.3	0.760	0.761	0.77	0.76	0.76	4.2
Random forest	0.778	0.779	0.77	0.74	0.77	8.2	0.883	0.883	0.88	0.87	0.88	5.8
Gradient boosting	0.791	0.792	0.80	0.79	0.82	44.3	0.855	0.856	0.86	0.86	0.85	31.8
Adaptive boosting	0.718	0.716	0.71	0.72	0.71	272.2	0.815	0.813	0.82	0.81	0.81	218.9
XG boosting	0.806	0.810	0.80	0.79	0.81	172.2	0.856	0.855	0.85	0.86	0.86	150.1
CATBoost	0.712	0.710	0.72	0.71	0.70	72.2	0.790	0.792	0.79	0.78	0.79	74.1