No-reference image quality metric based on image classification

EURASIP Journal on Advances in Signal Processing

Table 5 Pearson correlation coefficient of the NR image quality metric based on image classification as a function of the threshold value (TID2008 database)

Threshold	-0.8	-0.7	-0.6	-0.5	-0.4	-0.3	-0.2	-0.1	0	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8
NR metric using the proposed blocking and blur metrics	0.301	0.341	0.421	0.498	0.673	0.742	0.793	0.840	0.855	0.852	0.853	0.851	0.854	0.855	0.846	0.838	0.826
NR metric using Jeong's blocking and blur metrics [15]	0.301	0.336	0.406	0.470	0.624	0.681	0.723	0.769	0.801	0.801	0.818	0.834	0.843	0.846	0.839	0.833	0.824
NR metric using [6] and [10]	0.364	0.410	0.485	0.549	0.668	0.708	0.740	0.767	0.763	0.744	0.744	0.742	0.744	0.745	0.739	0.735	0.730
NR metric using [6] and [11]	0.351	0.412	0.487	0.554	0.708	0.752	0.790	0.827	0.829	0.824	0.802	0.776	0.778	0.779	0.772	0.767	0.757
NR metric using [3] and [10]	0.517	0.548	0.600	0.646	0.737	0.760	0.766	0.758	0.758	0.741	0.738	0.739	0.742	0.743	0.737	0.735	0.729
NR metric using [3] and [11]	0.508	0.550	0.602	0.651	0.774	0.801	0.815	0.818	0.825	0.821	0.797	0.773	0.776	0.778	0.771	0.767	0.756