Similarity Measure Learning in Closed-Form Solution for Image Classification

Table 2

Classification performance comparison on the Yale database.

Method	2 Train		3 Train		4 Train		5 Train		6 Train		7 Train
	Error (%)	d	Error (%)	d	Error (%)	d	Error (%)	d	Error (%)	d	Error (%)	d
PCA	56.6±6.3	29	50.6±8.3	44	47.4±7.2	58	43.8±6.5	74	40.8±7.4	32	39.5±5.1	30
RS-2DPCA	44.2±6.0	17	32.5±5.4	17	27.6±5.9	17	22.4±5.2	17	17.5±6.3	17	16.1±5.2	17
LDA	52.8±7.5	10	35.1±5.9	14	27.1±5.3	14	21.2±5.7	14	18.7±4.9	14	17.6±4.6	14
LPP	42.6±6.8	14	31.2±7.0	14	27.3±6.2	19	21.1±4.8	23	17.8±5.8	24	16.3±5.4	21
MFA	41.7±7.4	18	33.6±6.5	23	28.4±5.9	27	21.5±5.2	20	16.1±5.3	19	15.2±4.0	25
CDA	43.2±5.9	19	32.9±5.8	22	26.8±6.7	23	20.3±5.4	18	16.9±6.8	26	16.0±6.2	19
CEA	42.0±6.1	21	30.7±4.3	25	25.2±4.9	18	19.2±5.1	19	15.3±5.4	20	14.1±4.8	18
ISM-GE	43.1±6.5	19	29.2±5.9	20	23.6±6.8	19	17.5±6.3	20	14.7±5.6	22	12.5±5.1	19
MSE	42.4±7.2	23	32.3±6.7	24	28.2±5.2	22	19.6±5.9	21	16.2±5.1	24	15.3±5.7	22
CSML	40.3 ± 6.8	14	29.4 ± 5.6	19	22.7 ± 6.1	13	17.8 ± 4.7	20	13.4 ± 4.2	19	11.7 ± 5.2	15
KCSML	37.4 ± 7.1	18	28.7 ± 6.3	24	23.1 ± 7.5	21	15.9 ± 6.0	26	10.2 ± 5.4	21	9.6 ± 6.1	27