• Acta Optica Sinica
  • Vol. 38, Issue 8, 0815016 (2018)
Yan Diao*, Chenke Wu, Hua Luo, and Bijiao Wu
Author Affiliations
  • School of Manufacturing Science and Engineering, Sichuan University, Chengdu, Sichuan 610065, China
  • show less
    DOI: 10.3788/AOS201838.0815016 Cite this Article Set citation alerts
    Yan Diao, Chenke Wu, Hua Luo, Bijiao Wu. Line Detection Optimization Algorithm Based on Improved Probabilistic Hough Transform[J]. Acta Optica Sinica, 2018, 38(8): 0815016 Copy Citation Text show less
    Leaves produce a mesh edge that affects the line results. (a) Original image; (b) Canny edge; (c) line detected result
    Fig. 1. Leaves produce a mesh edge that affects the line results. (a) Original image; (b) Canny edge; (c) line detected result
    Steps of line detection in sorted edge points
    Fig. 2. Steps of line detection in sorted edge points
    Steps of line detection in unsorted edge points
    Fig. 3. Steps of line detection in unsorted edge points
    Algorithm operation flow chart
    Fig. 4. Algorithm operation flow chart
    Efficiency comparison of two kinds of algorithms. (a) Time consumption comparison of each group of experiments; (b) time ratio of each group of experiments
    Fig. 5. Efficiency comparison of two kinds of algorithms. (a) Time consumption comparison of each group of experiments; (b) time ratio of each group of experiments
    Straight line detection on a sorted edge image. (a) Analog image; (b) PPHT results; (c) our algorithm results
    Fig. 6. Straight line detection on a sorted edge image. (a) Analog image; (b) PPHT results; (c) our algorithm results
    Resistance of two algorithms to the boundary polymerization. (a) Original image; (b) Canny edge; (c) PPHT results; (d) our algorithm results
    Fig. 7. Resistance of two algorithms to the boundary polymerization. (a) Original image; (b) Canny edge; (c) PPHT results; (d) our algorithm results
    Comparison of a building image detection results. (a) Original image; (b) Canny edge; (c) PPHT results; (d) our algorithm results
    Fig. 8. Comparison of a building image detection results. (a) Original image; (b) Canny edge; (c) PPHT results; (d) our algorithm results
    Comparison of algorithm effect on a large picture. (a) Canny edge; (b) PPHT results; (c) our algorithm results
    Fig. 9. Comparison of algorithm effect on a large picture. (a) Canny edge; (b) PPHT results; (c) our algorithm results
    AlgorithmImage size /pixel×pixelLmin /pixelGap /pixelΔρ /pixelΔθ /radLinesTime /ms
    PPHTOurs4500×27288051π /180020197362311
    Table 1. Performance comparison of PPHT and our algorithm in sorted edge points
    NO.AlgorithmImage size /pixel×pixelLmin /pixelGap /pixelΔρ /pixelΔθ /radDetectedlinesTime /ms
    1PPHTOurs540×96012051π /18006923810209
    2PPHTOurs960×54012051π /180015963715227
    3PPHTOurs4608×2592200101π /180004302083206717674
    Table 2. Performance comparison of PPHT and our algorithm in unsorted edge points mode
    Yan Diao, Chenke Wu, Hua Luo, Bijiao Wu. Line Detection Optimization Algorithm Based on Improved Probabilistic Hough Transform[J]. Acta Optica Sinica, 2018, 38(8): 0815016
    Download Citation