Table 1 Percentage accuracy given by different techniques for classification.
Research technique | Dataset | Accuracy (%) | References |
|---|---|---|---|
CNN and saliency techniques | IP102 | 92.43 | |
Local binary patterns with support vector machine | Self insect image datset | 20.01 | |
AlexNet | Self insect image datset | 94.7 | |
ResNet50 | Self insect image datset | 97.4 | |
ResNet34 | Self insect image datset | 97.7 | |
MMAL-Net | IP102 | 72.2 | |
RAN + FPN + MMAL-Net + ResNet50 | D0 | 99.96 | |
MMAL-Net | IP102 | 74.1 | |
RAN + FPN + MMAL-Net + ResNet50 | D0 | 99.8 | |
Yolov5-S for 10 pest classes | Insect 10 | 70.5 | |
Transfer learning with AlexNet, ResNet, VGGNet | National Bureau of Agricultural Insect Resources | 96.75 | |
AlexNet for IP102 dataset | IP102 | 89.6 | |
Faster R-CNN ResNet50 | IP102, BugWood, Rice knowledge bank | 94.00 | |
Ensembles of CNNs | Kylberg virus benchmark data set | 95.52 |
