Table 1.
Portrayal information of the dataset.
Fig 1.
Sample images of each class.
Fig 2.
Instance distribution for each class.
Fig 3.
Schematic representation of methodology.
Fig 4.
Images of the augmented samples.
Fig 5.
ORNS architecture.
Fig 6.
Feature extraction after activation of each layer (one sample).
Fig 7.
weighted ensemble in layer - L.
Fig 8.
Organization of multi-layer CWE.
Table 2.
Trainable parameters for each architecture.
Table 3.
Performance evaluation of RNX002 and RNY002 in CWE-Layer 1.
Table 4.
Performance evaluation of RNX004 and RNY004 in CWE-Layer 1.
Table 5.
Performance evaluation of RNX006 and RNY006 in CWE-Layer 1.
Table 6.
Performance evaluation of RNX008 and RNY008 in CWE-Layer 1.
Table 7.
Performance evaluation of RNX016 and RNY016 in CWE-Layer 1.
Table 8.
Performance evaluation of RNX032 and RNY032 in CWE-Layer 1.
Table 9.
Performance evaluation of RNX040 and RNY040 in CWE-Layer 1.
Table 10.
Performance evaluation of RNX064 and RNY064 in CWE-Layer 1.
Table 11.
Performance evaluation of RNX080 and RNY080 in CWE-Layer 1.
Table 12.
Performance evaluation of RNX120 and RNY120 in CWE-Layer 1.
Table 13.
Performance evaluation of RNX160 and RNY160 in CWE-Layer 1.
Table 14.
Performance evaluation of RNX320 and RNY320 in CWE-Layer 1.
Table 15.
Performance evaluation of RNX and RNY in CWE-Layer 2.
Table 16.
Performance evaluation of common variants ensemble of RNX and RNY in CWE-Layer 2.
Table 17.
Performance evaluation of RNXY in CWE-Layer 3.
Table 18.
Performance evaluation of RN_XY in CWE-Layer 3.
Table 19.
Performance evaluation of RN in CWE-Layer 4.
Fig 9.
Confusion matrix obtained by RNXY architecture in CWE-Layer 3.
Fig 10.
ROC-AUC curve obtained by RNXY architecture in CWE-Layer 3.
Fig 11.
Confusion matrix obtained by RN_XY architecture in CWE-Layer 3.
Fig 12.
ROC-AUC curve obtained by RN_XY architecture in CWE-Layer 3.
Fig 13.
Confusion matrix obtained by RN architecture in CWE-Layer 4.
Fig 14.
ROC-AUC curve obtained by RN architecture in CWE-Layer 4.
Fig 15.
Step by step implementation of gradient class activation map.
Fig 16.
GradCAM visualization for each class.
Fig 17.
GradCAM visualization for AT explainability (Example by RNX002).
Table 20.
Performance evaluation of CWE without AT.
Table 21.
Performance evaluation using SA instead of CWE.
Table 22.
Performance evaluation using MV instead of CWE.
Table 23.
Performance evaluation using WA instead of CWE.
Table 24.
Performance evaluation of single-layer CWE.
Table 25.
Comparison of our proposed architecture with existing others.
Table 26.
Comparison of our proposed architecture with state-of-the-art methods.