Research Article
A Novel Approach for Detection of Pavement Crack and Sealed Crack Using Image Processing and Salp Swarm Algorithm Optimized Machine Learning
Table 2
Prediction performance comparison
| Class label | Indices | IP-SSA-SVM | Adam-DCNN | RFC | Adam-BPANN | Mean | Std. | Mean | Std. | Mean | Std. | Mean | Std. |
| Noncrack | CAR (%) | 90.50 | 5.33 | 89.50 | 7.44 | 86.50 | 5.97 | 79.33 | 8.28 | Precision | 0.85 | 0.11 | 0.86 | 0.14 | 0.63 | 0.18 | 0.60 | 0.18 | Recall | 0.87 | 0.09 | 0.86 | 0.13 | 0.96 | 0.08 | 0.74 | 0.16 | NPV | 0.93 | 0.05 | 0.92 | 0.11 | 0.99 | 0.02 | 0.89 | 0.08 | F1 Score | 0.85 | 0.08 | 0.85 | 0.10 | 0.74 | 0.13 | 0.65 | 0.16 |
| Sealed crack | CAR (%) | 92.83 | 4.75 | 88.33 | 3.67 | 75.17 | 6.35 | 82.00 | 7.98 | Precision | 0.92 | 0.11 | 0.85 | 0.08 | 0.97 | 0.05 | 0.76 | 0.16 | Recall | 0.88 | 0.07 | 0.82 | 0.08 | 0.58 | 0.06 | 0.72 | 0.12 | NPV | 0.94 | 0.04 | 0.90 | 0.05 | 0.64 | 0.09 | 0.85 | 0.07 | F1 Score | 0.89 | 0.08 | 0.83 | 0.05 | 0.73 | 0.05 | 0.73 | 0.13 |
| Crack | CAR (%) | 91.33 | 4.24 | 86.17 | 6.60 | 88.67 | 3.81 | 79.33 | 8.89 | Precision | 0.86 | 0.10 | 0.76 | 0.17 | 0.66 | 0.11 | 0.76 | 0.13 | Recall | 0.89 | 0.09 | 0.81 | 0.12 | 1.00 | 0.00 | 0.68 | 0.13 | NPV | 0.94 | 0.05 | 0.92 | 0.05 | 1.00 | 0.00 | 0.81 | 0.10 | F1 Score | 0.87 | 0.07 | 0.78 | 0.14 | 0.79 | 0.08 | 0.71 | 0.11 |
|
|