Fig. 1. Flow of pavement pothole detection algorithm

Fig. 2. Diagram of roughness

Fig. 3. Pavement point clouds. (a) Elevation rendering; (b) roughness rendering

Fig. 4. Diagram of pothole zone expansion

Fig. 5. Diagram of skewed distribution. (a) Positive skewed distribution; (b) negative skewed distribution

Fig. 6. Pothole and its neighboring pavement point clouds rendered by roughness (top view)

Fig. 7. Partially enlarged view. (a) Zone A; (b) zone B; (c) zone C

Fig. 8. Histograms of roughness. (a) Zone A; (b) zone B; (c) zone C

Fig. 9. Open source data point clouds rendered by elevation

Fig. 10. Vehicle-borne laser mobile measurement system

Fig. 11. Actually measured data point clouds rendered by elevation

Fig. 12. Roughness under different neighborhood range. (a) ra=0.1 m; (b) ra=0.2 m; (c) ra=0.3 m; (d) ra=0.4 m; (e) ra=0.5 m; (f) ra=0.6 m

Fig. 13. Roughness variation of pothole edge point, pothole internal point and pavement point

Fig. 14. Superimposed display of road point cloud and ground point cloud. (a) Open source data; (b) measured data

Fig. 15. Pothole detection results for open source data pavement. (a) Overall results; (b) detection results of region I; (c) detection results of region II

Fig. 16. Partially enlarged views of detection results in open source data pavement. (a) Result A; (b) result B; (c) result C; (d) result D; (e) result E; (f) result F

Fig. 17. Pothole detection results for measured data pavement

Fig. 18. Panoramic and partially enlarged views of detection results in measured data pavement. (a) Result A; (b) result B; (c) result C; (d) result D; (e) result E; (f) result F

Fig. 19. Comparison of geometric features