复杂背景下的结构光条纹中心提取算法研究

doi:10.3976/j.issn.1002-4026.20230133

Abstract

Abstract:

The most critical step in a line-structured light three-dimensional scan modeling system is to extract the centerline of the light stripe, but the interference of various environmental factors makes this extraction difficult. Several problems exists in a line-structured light streak image issues such as light spot interference, uneven distribution of light intensity, large differences in the width of the light bars, and complex background. This paper proposed a solution to overcome these problems. First, the structured light image is binarized using the Otsu method. Then, the improved density-based spatial clustering of applications with nose (DBSCAN) algorithm is used to retain the core points and remove the boundary and noise points. Finally, the core points are used as inputs to construct the graph data structure, and the shortest path search algorithm that fits the line-structured light streak image is used to obtain the center-line of the light streak. The experimental results show that the algorithm of this paper runs within 150 ms and the error is within 0.2 pixels. Moreover, this algorithm is applicable to various complex environments, meeting the requirements of real-time calculations, accuracy, and stability.

Key words: complex background, line-structured light, centerline extraction, DBSCAN algorithm, shortest path

CLC Number:

TN249

GAO Qiuling, CHENG Wei, LI Wenlong, GE Hailong, HOU Xingqiang, SONG Ruhui, WEI Jiajie, JIA Tianshuo, CAI Xinyan. Centerline extraction algorithm of structured light streak in a complex background[J].Shandong Science, 2024, 37(2): 65-73.

Figures/Tables 11

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Table 2

Table 3

References 16

[1]	HE K J, SUI C Y, LYU C Y, et al. 3D reconstruction of objects with occlusion and surface reflection using a dual monocular structured light system[J]. Applied Optics, 2020, 59(29): 9259-9271. doi: 10.1364/AO.402146
[2]	闫光绪, 贺赛先. 基于单目便携式激光扫描的小工件测量[J]. 应用激光, 2020, 40(2): 315-322. DOI: .
[3]	冀振燕, 宋晓军, 付文杰, 等. 激光光条中心线提取研究综述[J]. 测控技术, 2021, 40(6): 1-8. DOI: 10.19708/j.ckjs.2021.06.001.
[4]	李莹莹, 张志毅, 袁林. 线结构光光条中心提取综述[J]. 激光与光电子学进展, 2013, 50(10): 13-22. 10.3788/lop50.100002.
[5]	张衡, 苗红霞, 郭章旺, 等. 动态环境下的激光条纹中心线提取方法[J]. 计算机测量与控制, 2021, 29(12): 226-233. DOI: 10.16526/j.cnki.11-4762/tp.2021.12.041.
[6]	郭宏阳, 周建平, 薛瑞雷, 等. 强光干扰下的焊缝图像激光条纹提取算法研究[J]. 组合机床与自动化加工技术, 2020(6): 57-60. DOI: 10.13462/j.cnki.mmtamt.2020.06.014.
[7]	周渊, 孟祥群, 江登表, 等. 复杂干扰情况下的结构光条纹中心提取方法[J]. 中国激光, 2020, 47(12): 172-180. DOI: 10.3788/CJL202047.1204004.
[8]	章秀华, 洪汉玉, 徐洋洋, 等. 复杂光照条件下矿石三维视觉实时筛选方法[J]. 红外与激光工程, 2021, 50(11): 394-404.
[9]	JIAN X, CHEN X, HE W P, et al. Outdoor 3D reconstruction method based on multi-line laser and binocular vision[J]. IFAC-PapersOnLine, 2020, 53(2): 9554-9559. DOI: 10.1016/j.ifacol.2020.12.2436.
[10]	方夏章, 李垚, 方毅. 室内环境下线结构光光条提取方法[J]. 传感器与微系统, 2018, 37(10): 32-34. DOI: 10.13873/j.1000-9787(2018)10-0032-03.
[11]	SHI X Q, SUN Y Z, LIU H T, et al. Research on laser stripe characteristics and center extraction algorithm for desktop laser scanner[J]. SN Applied Sciences, 2021, 3(3): 1-12. DOI: 10.1007/s42452-021-04309-w.
[12]	牛晗, 伍希志, 任桂芹, 等. 基于Otsu与CANNY算法的竹片缺陷图像检测[J]. 森林工程, 2022, 38(6): 75-81. DOI: 10.3969/j.issn.1006-8023.2022.06.010.
[13]	赫德军, 武欣嵘, 俞璐. 密度聚类方法研究[J]. 通信技术, 2022, 55(2): 135-142. DOI: 10.3969/j.issn.1002-0802.2022.02.001.
[14]	张朋, 李小林, 王李妍. 基于DBSCAN的动态邻域密度聚类算法[J]. 计算机科学, 2023, 50(S1): 609-615.
[15]	CHEN J S, SU G D, XIANG S B. Robust welding seam tracking using image seam extraction[J]. Science and Technology of Welding and Joining, 2012, 17(2): 155-161. DOI: 10.1179/1362171811y.0000000091.
[16]	刘宇豪, 郭京波, 刘文栋, 等. 多重干扰下管片端面条纹中心提取优化算法[J]. 应用激光, 2021, 41(4): 869-875. DOI: 10.14128/j.cnki.al.20214104.869.

聚类方法分类	代表方法	簇类形状	时间复杂度	准确性	对噪声和离群点是否敏感
基于划分	k-means、k-modes、 CLARA、PAM等	球形	较低	依赖于初始聚类中心的选择, 易陷入局部最优解	敏感
基于层次	BIRCH、ROCK、CURE等	任意形状	较高,不适合处理大规模数据	过程具有不可逆性,聚类终止条件具有不精确性	敏感
基于密度	DBSCAN、OPTICS、DPC等	任意形状	较低	对密度分布不均的数据聚类效果差	不敏感
基于网格	STING、Wave Cluster、 CLIQUE等	任意形状	低	取决于网格的大小,难以检测到数据的斜边界	敏感
基于模型	GMM、SOM等	与选取的模型有关

实验编号	极值法		细化法		灰度重心法		Steger算法		本文算法
实验编号	图像1	图像2	图像1	图像2	图像1	图像2	图像1	图像2	图像1	图像2
1	194	169	95	106	262	267	608	618	128	124
2	177	157	93	107	270	279	604	620	130	126
3	172	170	96	102	247	254	606	619	129	123
4	165	174	101	105	263	264	609	624	132	128
5	175	171	97	99	277	263	610	622	128	125
6	176	173	92	105	255	250	602	622	126	122
7	170	167	95	109	262	268	604	621	125	121
8	181	183	91	97	258	260	601	619	124	124
9	166	176	97	104	259	265	602	617	126	125
10	168	178	94	107	262	277	605	615	127	123
平均	174	172	95	104	262	265	605	620	129	124

实验编号	极值法		细化法		灰度重心法		Steger算法		本文算法
实验编号	图像1	图像2	图像1	图像2	图像1	图像2	图像1	图像2	图像1	图像2
1	0.257	0.605	0.543	1.186	0.215	0.333	0.098	0.130	0.146	0.173
2	0.297	0.604	0.545	1.187	0.218	0.332	0.095	0.128	0.145	0.165
3	0.295	0.610	0.556	1.182	0.222	0.335	0.097	0.128	0.145	0.178
4	0.291	0.595	0.541	1.178	0.214	0.334	0.099	0.126	0.156	0.173
5	0.296	0.596	0.548	1.166	0.215	0.340	0.102	0.127	0.147	0.170
6	0.296	0.607	0.558	1.179	0.213	0.343	0.096	0.133	0.146	0.175
7	0.295	0.605	0.546	1.188	0.225	0.329	0.097	0.129	0.147	0.171
8	0.295	0.608	0.549	1.195	0.228	0.331	0.102	0.126	0.150	0.178
9	0.297	0.598	0.533	1.184	0.223	0.326	0.096	0.130	0.141	0.187
10	0.315	0.601	0.543	1.175	0.216	0.329	0.098	0.128	0.140	0.177
平均	0.293	0.603	0.546	1.182	0.219	0.333	0.098	0.129	0.146	0.175

Centerline extraction algorithm of structured light streak in a complex background

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 16

Related Articles 3

Metrics

Comments

Recommended 0

[1]	WANG Bo, BAO Feng, WU Xin-ran. Optimization of random assignment model of urban rail transit passenger flow based on fully observable and controllable experimental method [J]. Shandong Science, 2022, 35(4): 90-97.
[2]	SUN Yue-peng, GUO Ren-yong, YU Tao. Comparison among the collection region distributions of multidirectional pedestrian flows based on density clustering [J]. Shandong Science, 2021, 34(5): 64-74.
[3]	SUN Yue, SONG Rui, QIU Guo. A site location method of customized business bus based on passenger demand data [J]. Shandong Science, 2019, 32(1): 102-112.