Improvement in the Accuracy of the Postclassification of Land Use and Land Cover Using Landsat 8 Data Based on the Majority of Segment-Based Filtering Approach

Table 3

Result of the accuracy assessment for the LULC classification based on the ML classifier with the MaSegFil approach.

LULC class	C_0	C_1	C_2	C_3	C_4	C_5	C_6	C_7	C_8	C_9	C_10	Total	User accuracy	Kappa
C_0	109	1	0	0	0	1	0	0	0	0	0	111	0.982	0.000
C_1	30	261	28	0	0	9	0	0	0	0	0	328	0.796	0.000
C_2	23	4	79	0	0	2	23	21	4	0	0	156	0.506	0.000
C_3	14	0	1	4	0	0	21	1	20	0	0	61	0.066	0.000
C_4	0	2	3	0	3	0	0	4	0	2	0	14	0.214	0.000
C_5	28	26	13	0	0	38	0	1	4	0	0	110	0.345	0.000
C_6	10	0	4	1	0	0	589	10	0	0	0	614	0.959	0.000
C_7	4	0	0	0	0	0	18	365	1	3	0	391	0.934	0.000
C_8	10	0	5	0	0	0	20	5	184	0	0	224	0.821	0.000
C_9	0	0	0	0	0	0	1	2	0	60	4	67	0.896	0.000
C_10	0	0	0	0	0	0	0	0	0	5	46	51	0.902	0.000
Total	228	294	133	5	3	50	672	409	213	70	50	2127	0.000	0.000
Procedure accuracy	0.478	0.888	0.594	0.800	1.000	0.760	0.876	0.892	0.864	0.857	0.920	0.000	0.817	0.000
Kappa	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.779

C_0: primary dryland forest; C_1: secondary dryland forest; C_2: fields; C_3: open field; C_4: wetland forest; C_5: plantation; C_6: settlement; C_7: rice fields; C_8: shrubs; C_9: fishpond; C_10: water body.