Research Article
Fast Traffic Sign Detection Approach Based on Lightweight Network and Multilayer Proposal Network
Table 5
Detection results of the proposed method and other methods on the Tsinghua-Tencent 100K dataset.
| | Methods | Small | Medium | Large | Inference time (s) | | (%) | (%) | (%) | (%) | (%) | (%) | (%) | (%) | (%) | |
| | Faster R-CNN [17] | 24.1 | 49.8 | 32.5 | 65.6 | 83.7 | 73.6 | 80.8 | 91.2 | 85.7 | 2.15 | | SSD [26] | 25.3 | 43.4 | 32.0 | 67.8 | 77.5 | 72.3 | 81.5 | 86.9 | 84.1 | 0.12 | | Multiclass Network [9] | 81.7 | 87.4 | 84.5 | 90.8 | 93.6 | 92.2 | 90.6 | 87.7 | 89.1 | 5.62 | | MR-CNN [8] | 82.9 | 89.3 | 86.0 | 92.6 | 94.4 | 93.5 | 92.0 | 88.2 | 90.1 | — | | Proposed method | 80.1 | 89.0 | 84.3 | 91.3 | 94.6 | 92.9 | 92.5 | 91.8 | 92.1 | 0.28 |
|
|
