An Open Access Journal
From: Vision-based vehicle detection and counting system using deep learning in highway scenes
Vehicle | Remote area | Proximal area | Average correct rate | ||||
---|---|---|---|---|---|---|---|
category | Our method | Full-image | Our method | Full-image | Our method | Full-image | |
detection method | detection method | detection method | |||||
Actual number of vehicles | Car | 91.96% | 8.58% | 98.79% | 83.54% | 95.375% | 46.06% |
Bus | 89.94% | 18.08% | 96.05% | 83.86% | 92.995% | 50.97% | |
Truck | 94.51% | 18.21% | 98.61% | 80.99% | 96.56% | 49.6% | |
Overall correct rate | 92.14% | 14.96% | 97.82% | 82.80% | 94.976% | 48.86% |