Skip to main content

Advances, Systems and Applications

Table 3 Experimental results of different models on the mangrove dataset

From: MSFANet: multi-scale fusion attention network for mangrove remote sensing lmage segmentation using pattern recognition

Model

Miou(%)

PA(%)

F1_score(%)

Mangrove(CPA) (%)

River & Ocean(CPA) (%)

MSFANet ResNet-50

0.86015

0.96419

0.92029

0.98981

0.96425

MSFANet ResNet-101

0.84928

0.96261

0.91238

0.98951

0.94620

UNET

0.68159

0.91982

0.77046

0.98508

0.91875

GCN ResNet-50

0.78231

0.95376

0.86321

0.98847

0.94395

GCN ResNet-101

0.78443

0.95078

0.86935

0.98013

0.93630

DeepLabv3 + 

0.73638

0.92917

0.83042

0.98569

0.93651

ExFuse ResNet-50

0.72884

0.94070

0.80712

0.98487

0.93891

ExFuseResNet-101

0.78559

0.94418

0.86893

0.98364

0.94242