Research Article
Identification Method of Citrus Aurantium Diseases and Pests Based on Deep Convolutional Neural Network
Table 2
Results of recognition rate.
| Serial number | ā | Number of correctly identified samples (piece) | Recognition rate (%) |
| 1 | Method of reference [4] | 175 | 87.5 | 2 | 173 | 86.5 | 3 | 171 | 85.5 | 4 | 174 | 87 | 5 | 177 | 88.5 | Average | 174.0 | 87.0 |
| 1 | Method of reference [5] | 182 | 91 | 2 | 184 | 92 | 3 | 185 | 92.5 | 4 | 183 | 91.5 | 5 | 181 | 90.5 | Average | 183.0 | 91.5 |
| 1 | Method of this paper | 196 | 98 | 2 | 199 | 99.5 | 3 | 198 | 99 | 4 | 199 | 99.5 | 5 | 197 | 98.5 | Average | 197.8 | 98.9 |
|
|