PolSAR Image Classification on Flevoland NASA/JPL AIRSAR 1989 99% (test) benchmark leaderboard · SOTA2 Research

PolSAR Image Classification on Flevoland NASA/JPL AIRSAR 1989 99% (test)

99.84Water AccuracyHybridCVNet

Evaluation Results

Method	Links
HybridCVNet2026.05		99.84	98.66	97.46	91.47	95.15	92.57	97.44	99.39	98.67	96.18	98.36	97.51	99.58	97.84	86.16	97.39	96.41	0.9715
3D-CNN2026.05		99.46	94.56	96.15	85.97	94.14	81.95	92.69	96.28	92.4	82.6	92.45	95.29	98.66	95.55	98.62	93.75	93.12	0.9317
HybridRVNet2026.05		98.99	98.72	97.04	88.95	97.12	84.91	91.53	98.57	95.05	91.91	97.94	92.07	99.24	95.63	69.72	95.39	93.15	0.9496
WaveletCNN2026.05		98.59	86.87	98.5	89.09	95.48	95.74	71.16	70.8	89.39	95.2	57.96	99.76	98.16	96.66	74.57	88.12	87.86	0.8701
swin2026.05		98.23	93.51	93.42	79.52	96.58	88.45	89.79	96.93	93.84	87.6	93.42	92.01	98.01	99.13	65.57	93.58	91.07	0.9298
PolSARFormer2026.05		98.23	93.51	93.42	79.52	96.58	88.45	89.79	96.93	93.84	87.6	93.42	92.01	98.01	99.15	65.57	95.74	93.61	0.9535
ViT2026.05		96.93	98.9	97.61	84.19	88.8	85.81	89.46	95.75	97.51	95.17	93.56	87	99.38	98.77	9	93.28	87.85	0.9265