Table 2 Performance comparison of the developed approach with state-of-the-art approaches on ImageCLEF-2012 dataset.
Approach | Description | Accuracy (%) |
|---|---|---|
IBM multimedia analytics27 | Data augmentation, visual features and SVM classification | 69.60 |
Dimitrovski et al.28 | Several visual and textual features and SVM classification | 78.60 |
medGIFT31 | Mixed type of hand crafted features | 66.20 |
Present study | ||
Hybrid feature set | Mixed hand-crafted features of SIFT, BOW, LBP, LTP, HOG, ECD, ECDWT and LDA classification | 71.4 |
TLRN-LDA | ResNet50 transfer learning Deep features and LDA classification | 87.91 |
