Fig. 1. Laser stripe that breaks due to surface contamination of the object being measured (simulation)

Fig. 2. Laser stripe disturbed by light spot (simulation)

Fig. 3. Laser stripe disturbed by sparks (simulation)

Fig. 4. Demonstration of density clustering algorithm. (a) Pixelated laser stripe image; (b) analysis of density clustering for local laser stripe image

Fig. 5. Pseudo-code of density clustering algorithm for binary image

Fig. 6. Demonstration of density clustering algorithm to repair broken stripe, ε=3, P=4

Fig. 7. Graph data structure based on core points

Fig. 8. Demonstration of broken stripe processing. (a) Enlarged view of simulated broken stripe interference; (b) experimental results on broken laser stripe

Fig. 9. Experimental results of light spot interference

Fig. 10. Experimental results of spark interference

Fig. 11. Enlarged view of the dotted framed area of Fig. 10

Fig. 12. Errors of different disturbed image extraction results relative to the reference centerline

Fig. 13. Extraction results of the proposed algorithm when applied to the actual stripe images. (a)--(d) Original images; (e)--(h) enlarged views