Research Article
Macroscopic Rock Texture Image Classification Using a Hierarchical Neuro-Fuzzy Class Method
Table 4
Neural networks performance in the classification of the following rock subclassess: gneiss, basalt, diabase, and rhyolite.
| | Rock | Database test | Correct percentage of training set | Correct percentage of validation set |
| | Gneiss | Gneiss1 | 96% | 92% | | Gneiss2 | 100% | 96% | | Gneiss3 | 100% | 96% | | Gneiss4 | 100% | 96% |
| | Basalt | Basalt1 | 81% | 55% | | Basalt2 | 100% | 86% | | Basalt3 | 100% | 85% | | Basalt4 | 96% | 67% |
| | Diabase | Diabase1 | 73.6% | 51.2% | | Diabase2 | 93.6% | 69.6% | | Diabase3 | 97.6% | 68% | | Diabase4 | 92% | 54.4% |
| | Rhyolite | Rhyolite1 | 76% | 49.6% | | Rhyolite2 | 97.6% | 75.2% | | Rhyolite3 | 98.4% | 64% | | Rhyolite4 | 88% | 59.2% |
|
|
