Hyperspectral Image Classification on QUH-Tangdaowan (test) benchmark leaderboard · SOTA2 Research

Hyperspectral Image Classification on QUH-Tangdaowan (test)

99.83Rubber Track AccuracyIP-SWIN

Evaluation Results

Method	Links
IP-SWIN2025.11		99.83	98.34	92.66	98.61	83.46	88.16	77.39	99.48	98.34	98.93	88.93	18.83	99.88	86.91	83.79	91.7	85.59	99.64	94.32	88.36	93.54
SS-MixNet2025.11		99.78	98.45	95.08	99.18	86.31	91.4	77.94	99.8	98.54	96.52	81.35	80.64	99.64	93.91	83.22	95.8	86.3	99.98	95.68	92.44	95.08
SimPool Former2025.11		99.61	97.16	92.84	92.73	87.54	83.53	72.45	98.4	97.81	76.28	43.64	73.3	41.28	72.23	68.98	92.13	76.76	99.6	90.64	81.46	89.29
3D-CNN2025.11		98.98	98.85	92.45	98.35	89.8	90.07	77.38	99.86	97.54	95.96	74.88	68.89	99.94	82.28	85.66	94.18	76.97	99.93	94.5	90.11	93.73
2D-CNN2025.11		98.74	90.25	87.11	94.61	35.18	64.78	63.35	96.38	90.11	55.36	26.8	67.16	47.86	69.19	76.78	92.97	62.08	97.63	86.75	73.13	84.8
Hybrid KAN2025.11		98.38	92.91	78.55	93.68	71	87.07	62.21	98.57	93.85	80.15	34.86	50.87	86.36	42.44	64.38	88.35	64.09	99.38	87.54	77.06	85.75