Table 3 Accuracies(%) obtained by applying Inception-ResNet-V2, VGG19, ResNet152, DenseNet201, Xception, MobileNetV2 models for both raw and edge-mapped images.
Model | SARS-COV-2 Ct-Scan Dataset | COVID-CT dataset | covid-chestxray-dataset + Chest X-Ray Images (Pneumonia) dataset | CMSC-678-ML-Project GitHub (3-class) | CMSC-678-ML-Project GitHub (4-class) | |||||
|---|---|---|---|---|---|---|---|---|---|---|
Raw image | Edge image | Raw image | Edge image | Raw image | Edge image | Raw image | Edge image | Raw image | Edge image | |
Inception-ResNet-V2 | 77.85 | 80.08 | 74.35 | 78.95 | 98.22 | 98.05 | 82.61 | 91.3 | 77.56 | 86.45 |
VGG19 | 78.27 | 82.55 | 79.60 | 84.27 | 98.45 | 96.50 | 86.96 | 97.83 | 79.65 | 92.2 |
ResNet152 | 77.87 | 84.58 | 86.65 | 87.97 | 98.68 | 97.82 | 91.31 | 91.40 | 86.13 | 85.88 |
DenseNet201 | 75.86 | 85.69 | 89.11 | 90.21 | 99.07 | 97.35 | 95.65 | 96.13 | 88.65 | 90.44 |
Xception | 83.30 | 81.79 | 82.01 | 87.58 | 96.74 | 99.22 | 82.61 | 86.96 | 82.15 | 83.97 |
MobileNetV2 | 77.46 | 80.48 | 78.18 | 76.97 | 98.76 | 98.52 | 93.48 | 84.74 | 81.45 | 82.25 |
