Remote Sensing Technology and Application

Fusion of Multiscale Low-rank Representation and Two Way Recursive Filtering for Hyperspectral Image Classification

Mei LU, Jiatian LI, Wen LI, Mihong HU, Jiaxin YANG

Remote Sensing Technology and Application, 2024, 39(2): 393-404. DOI: 10.11873/j.issn.1004-0323.2024.2.0393

Class	Train	Test	SVM	PCA	IFRF	HiFi	CCJSR	R-VCANet	SSRN	MSLRR	MSLRR_ TWRF
Alfalfa	10	36	38.81	23.61	77.45	98.61	67.24	100	90	98.33	100
Corn_n	10	1 418	50.69	52.56	75.68	67.48	63.88	49.53	74.76	71.09	70.92
Corn_m	10	820	42.22	39.41	62.20	84.87	64.30	73.49	73.47	75.51	79.52
Corn	10	227	27.06	25.06	54.89	92.73	41.34	94.63	86.64	96.21	96.78
Grass_m	10	473	75.37	60.97	88.19	78.48	95.78	88.92	95.86	85.67	86.96
Grass_t	10	720	84.91	79.21	91.15	96.60	95.82	94.47	93.33	99.64	98.03
Grass_P	10	18	39.69	26.64	50.84	98.89	37.46	100	79.82	97.78	97.78
Hay_w	10	468	96.20	97.33	100	96.88	99.57	95.73	91.75	100	100
Oats	10	10	14.18	21.27	30.77	100	13.77	100	100	100	100
Soybean_n	10	962	52.26	39.96	69.67	85.30	62.36	76.17	77.54	84.83	86.75
Soybean_m	10	2 445	66.89	63.46	87.61	70.49	78.69	65.38	83.93	89.91	91.90
Soybean_c	10	583	33.15	33.63	77.36	84.15	68.03	71.90	79.80	87.32	91.56
Wheat	10	195	79.56	78.75	76.86	99.33	90.40	99.18	98.08	99.49	99.49
Woods	10	1 255	91.32	87.97	97.97	93.64	96.56	91.68	93.45	90.91	97.10
Buildings	10	376	38.23	35.28	78.70	89.65	73.62	81.49	83.29	91.65	92.69
stone	10	83	83.48	84.72	96.75	99.28	90.75	99.52	97.02	95.78	97.95
OA	/	/	59.44±0.03	54.60±0.03	79.14±0.03	81.97±0.02	74.93±0.02	75.49±0.01	80.63±0.04	87.04±0.03	89.05±0.02
AA	/	/	57.13±0.03	53.11±0.02	76.01±0.05	89.77±0.01	71.22±0.2	86.38±0.01	87.42±0.02	91.51±0.09	92.96±0.08
Kappa	/	/	54.47±0.03	49.27±0.04	76.51±0.04	79.69±0.03	71.76±0.02	72.44±0.03	77.97±0.05	85.20±0.04	87.48±0.03

Table 2 Classification accuracy of different methods for Indian Pines

Other figure/table from this article