A fast source-oriented image clustering method for digital forensics

EURASIP Journal on Image and Video Processing

Table 3 Comparison of clustering algorithms on \(\mathcal {D}_{3}\)

Algorithms	1024×1024					512×512					512×256
	\(\mathcal {P}\)	\(\mathcal {R}\)	\(\mathcal {F}\)	\(\mathcal {A}\)	\(\mathcal {N}\)	\(\mathcal {P}\)	\(\mathcal {R}\)	\(\mathcal {F}\)	\(\mathcal {A}\)	\(\mathcal {N}\)	\(\mathcal {P}\)	\(\mathcal {R}\)	\(\mathcal {F}\)	\(\mathcal {A}\)	\(\mathcal {N}\)
SC [31]	0.10	1.00	0.18	0.06	0.10	0.17	0.94	0.29	0.09	0.18	0.41	0.72	0.52	0.20	0.56
HC [34]	0.71	0.50	0.58	0.44	1.44	0.81	0.21	0.33	0.52	3.92	0.76	0.13	0.22	0.31	5.94
SNNC [43]	0.45	0.26	0.32	0.23	1.74	0.35	0.15	0.21	0.11	2.34	0.30	0.12	0.17	0.07	2.50
NCUT [45]	0.92	0.13	0.22	0.52	7.30	0.72	0.14	0.24	0.37	4.96	0.53	0.15	0.23	0.25	3.58
MRF [49]	0.51	0.47	0.49	0.21	1.08	0.44	0.45	0.44	0.15	0.96	0.32	0.47	0.38	0.11	0.68
Proposed	0.87	0.75	0.80	0.74	1.16	0.68	0.64	0.66	0.48	1.06	0.52	0.47	0.49	0.16	1.10