Multiscale Dense Cross-Attention Mechanism with Covariance Pooling for Hyperspectral Image Scene Classification

Table 6

Classification results of different methods on the PU data (%).

Methods	1%			5%			10%
	OA	AA	Kappa	OA	AA	Kappa	OA	AA	Kappa
AlexNet [23]	87.15	84.24	82.66	92.81	91.77	90.43	92.85	93.46	91.32
ResNet [33]	84.06	82.38	77.59	90.81	91.32	87.76	94.15	94.78	93.19
DenseNet [34]	82.32	80.61	75.69	90.25	89.49	87.08	92.27	93.46	91.21
PRAN [35]	89.27	88.61	85.63	93.45	92.32	91.48	94.71	93.55	93.02
FSSFNet [36]	86.63	84.23	82.11	93.72	92.35	91.65	94.37	92.65	92.52
SAGP [37]	84.32	83.84	78.83	91.09	89.75	88.14	93.73	92.97	91.65
MDCA-CP	89.40	87.97	86.81	93.55	91.91	92.21	94.99	93.02	93.11