An Adaptive Edge Detection Method Based on Local Edge Feature Descriptor

Gao Jiayue; Xu Hongli; Shao Kailiang; Yin Hui

doi:10.3788/CJL202047.0604003

Journals >Chinese Journal of Lasers >Volume 47 >Issue 6 >Page 604003 > Article

Chinese Journal of Lasers
Vol. 47, Issue 6, 604003 (2020)

An Adaptive Edge Detection Method Based on Local Edge Feature Descriptor

Gao Jiayue¹, Xu Hongli^1、2、*, Shao Kailiang¹, and Yin Hui^1、3

Author Affiliations

¹School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China

²Key Laboratory of Beijing for Railway Engineering, Beijing 100044, China

³Beijing Key Lab of Traffic Data Analysis and Mining, Beijing 100044, China

show less

DOI: 10.3788/CJL202047.0604003 Cite this Article Set citation alerts

Gao Jiayue, Xu Hongli, Shao Kailiang, Yin Hui. An Adaptive Edge Detection Method Based on Local Edge Feature Descriptor[J]. Chinese Journal of Lasers, 2020, 47(6): 604003 Copy Citation Text

show less

Fig. 1. Normal vector of point cloud

Download full size

Fig. 2. Normal vector of feature region and non-feature region. (a) Feature area; (b) non-feature area

Download full size

Fig. 3. Feature point and non-feature point

Download full size

Fig. 4. Algorithm for adaptive k-neighbor selection

Download full size

Fig. 5. star model and local feature parameter distribution of the model. (a) star model; (b) LEFD distribution

Download full size

Fig. 6. fandisk model and local feature parameter distribution map of the model. (a) fandisk model; (b) LEFD distribution

Download full size

Fig. 7. Algorithm for edge detection

Download full size

Fig. 8. Model point cloud visualization

Download full size

Fig. 9. Comparison of results of complex models. (a)-(c) Original point cloud; (d)-(f) results of BE method;(g)-(i) results of the method in literature[10]; (j)-(l) results of the method in literature[10]; (m)-(o) results of our method

Download full size

Fig. 10. Comparison of results of simple models. (a)-(c) Original point cloud; (d)-(f) results of BE method;(g)-(i) results of the method in literature[10]; (j)-(l) results of the method in literature[10]; (m)-(o) results of our method

Download full size

Fig. 11. Dihedral angle with different angles. (a) 22°; (b) 45°; (c) 90°

Download full size

Fig. 12. Results of noise immunity. (a) fandisk model, adding noise with standard deviation of 0.5%; (b) fandisk model, adding noise with standard deviation of 2%; (c) bunny model, adding noise with standard deviation of 0.5%; (d) bunny model, adding noise with standard deviation of 2%

Download full size

Fig. 13. Data acquired by Kinect. (a) chair 1; (b) chair 2

Download full size

Fig. 14. Detection of Kinect data with our method. (a) chair 1; (b) chair 2

Download full size

Point	$\begin{matrix} n_{p}^{k} \end{matrix}$
Point	k=10	k=14	k=18	k=22	k=26	k=30
a	1.47×10^-10	1.80×10^-10	3.54×10^-10	2.79×10^-10	1.74×10^-10	2.50×10^-10
b	1.73×10^-10	2.58×10^-10	0.149	0.207	0.248	0.194
c	0.527	0.573	0.604	0.592	0.662	0.634

Table 1. Calculated characteristic value of mean angle of normal vector

Point	σ_p₁	σ_p₂	σ_p₃	σ_p₄	σ_p₅
a	1.224	1.966	0.788	0.623	1.436
b	1.491	5×10⁸	1.389	1.198	0.782
c	1.087	1.054	0.980	1.118	0.957

Table 2. Rate of change of mean angle of normal vector

Method	P			R
Method	22°	45°	90°	22°	45°	90°
BE	0.713	0.790	0.892	0.554	0.691	0.684
Method in literature[10]	0.795	0.907	0.902	0.962	0.873	0.909
Method in literature[12]	0.773	0.854	0.892	0.926	0.915	0.922
Ours	0.834	0.920	0.924	0.941	0.932	0.916

Table 3. Edge detection of several dihedral angles

Gao Jiayue, Xu Hongli, Shao Kailiang, Yin Hui. An Adaptive Edge Detection Method Based on Local Edge Feature Descriptor[J]. Chinese Journal of Lasers, 2020, 47(6): 604003

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information