Research Article
Night-Time Vehicle Detection Algorithm Based on Visual Saliency and Deep Learning
Table 1
Vehicle detection rates with different DBN structures and two shallow models.
| | Classifier types | Correct detection | Correct rate |
| | 2D-DBN (two hidden layers) | 5057/5733 | 88.21% | | 2D-DBN (three hidden layers) | 5250/5733 | 91.57% | | 2D-DBN (four hidden layers) | 5337/5733 | 93.09% | | 2D-DBN (five hidden layers) | 5291/5733 | 92.29% | | HOG + SVM | 4976/5733 | 86.80% | | Haar + Adaboost | 4792/5733 | 83.59% |
|
|
