Fig. 1. Structure of coal flow scanning system. (a) Scanning scene; (b) light stripe image

Fig. 2. Light stripe images collected by the experimental platform

Fig. 3. Light stripe image obtained by acquisition. (a) Imaging of point E on the surface of static object; (b) imaging between points D and E on the surface of moving objects; (c) imaging between points C and E on the surface of moving objects

Fig. 4. Light stripe imaging model. (a) There is no height difference in the movement direction; (b) there is a height difference in the movement direction

Fig. 5. Gray distribution of light stripe cross section pixel. (a) Flat-top Gaussian distribution; (b) multi-beam Gaussian superposition distribution

Fig. 6. Flow chart of the fast extraction algorithm for the center of the light strip

Fig. 7. Flow chart of the fast positioning algorithm in the light strip area

Fig. 8. Strategies for locating light strip area . (a) Full-map localization strategy; (b) regional growth strategy

Fig. 9. Segmentation results of different threshold methods. (a) Extreme value method; (b) Otsu method; (c) P-tile threshold method

Fig. 10. Extraction results of different algorithms. (a) Extraction results of Fig. 3 (a) by the gray-scale centroid algorithm; (b) extraction results of Fig. 3 (c) by the gray-scale centroid algorithm; (c) extraction results of Fig.3 (c) by the Steger algorithm; (d) extraction result of Fig. 3 (c) by our algorithm

Fig. 11. Simulation diagrams of different distribution models. (a) Gaussian-like distribution; (b) flat-topped Gaussian distribution; (c) multi-beam Gaussian superposition distribution

Fig. 12. Extraction results of linear structured light center on coal flow surface

Fig. 13. Partial enlarged view. (a) Light strip center extraction on the belt; (b) light strip center extraction on the coal stream

Table 1. Extract results of different algorithms

Table 2. Extract results of different algorithms