Research Article
Enhancing Photovoltaic Module Fault Diagnosis with Unmanned Aerial Vehicles and Deep Learning-Based Image Analysis
Table 13
Performance comparison of proposed method with various state-of-the-art techniques.
| | Reference | Methodology used | Classification accuracy (%) |
| | [38] | DeepSolarEye (ResNet-based) | 97.80 | | [39] | Yolo V3 | 96.30 | | [13] | Custom CNN | 79.06 | | [40] | Custom CNN | 94.30 | | [16] | Custom CNN | 97.90 | | [41] | Custom CNN | 95.07 | | [42] | Pretrained CNN+random forest | 98.25 | | [43] | Pretrained CNN+ k-nearest neighbor | 98.95 | | [44] | Ensemble model | 99.04 | | Proposed | DenseNet-201+kNN | 100.00 |
|
|
