Table 5 Performance of the proposed method using the ridgelet based EWT.

Train/test (%)	Technique	Classifier	TPR	TNR	PPV	F-score	AUC	Accuracy
70/30	Gaussian Noise	SVM	100	7.14	84.33	91.49	53.47	84.52
	Gaussian Noise	LPBoost	97.14	21.43	86.07	91.27	59.28	84.52
	Impulse Noise	SVM	95.71	50	90.54	93.05	72.85	88.09
	Impulse Noise	LPBoost	95.71	64.28	93.05	94.36	79.99	90.48
	Shear	SVM	100	7.14	84.33	91.49	53.47	84.52
	Shear	LPBoost	97.14	21.43	86.07	91.27	59.28	84.52
	Translation	SVM	100	14.28	85.36	92.1	57.14	85.71
	Translation	LPBoost	98.57	50	90.78	94.51	74.28	90.47
	Rotation	SVM	100	7.14	84.33	91.49	53.47	84.52
	Rotation	LPBoost	94.28	42.85	89.18	91.66	68.56	85.71
	TSSR based Enhancement	SVM	100	28.57	87.5	93.33	64.28	88.09
	TSSR based Enhancement	LPBoost	100	42.86	89.74	94.6	71.43	90.47
80/20	Gaussian Noise	SVM	100	11.11	85.45	92.15	55.55	85.71
	Gaussian Noise	LPBoost	100	22.22	87.04	93.07	61.11	87.5
	Impulse Noise	SVM	95.74	55.55	91.84	93.75	75.64	89.28
	Impulse Noise	LPBoost	95.74	66.67	93.75	94.73	81.2	91.07
	Shear	SVM	100	11.11	85.45	92.15	55.55	85.71
	Shear	LPBoost	100	33.33	88.68	94	66.66	89.28
	Translation	SVM	100	11.11	85.45	92.15	55.55	85.71
	Translation	LPBoost	95.74	66.67	93.75	94.73	81.2	91.07
	Rotation	SVM	100	11.11	85.45	92.15	55.55	85.71
	Rotation	LPBoost	95.74	44.44	90	92.78	70.09	87.5
	TSSR based Enhancement	SVM	100	55.56	92.15	95.91	77.8	92.86
	TSSR based Enhancement	LPBoost	97.87	55.56	92	94.84	76.71	91.07

Quick links

Search