3D pose estimation of large and complicated workpieces based on binocular stereo vision

Zhifeng Luo; Ke Zhang; Zhigang Wang; Jian Zheng; Yixin Chen

doi:10.1364/AO.56.006822

Applied Optics
Vol. 56,
Issue 24,
pp. 6822-6836
(2017)
•https://doi.org/10.1364/AO.56.006822

3D pose estimation of large and complicated workpieces based on binocular stereo vision

Zhifeng Luo, Ke Zhang, Zhigang Wang, Jian Zheng, and Yixin Chen

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Zhifeng Luo, Ke Zhang, Zhigang Wang, Jian Zheng, and Yixin Chen, "3D pose estimation of large and complicated workpieces based on binocular stereo vision," Appl. Opt. 56, 6822-6836 (2017)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

A binocular stereo vision method is proposed for automatically locating the position and posture of workpieces, which is especially important when processing large, complicated structures, for example, the laser hardening and laser cladding of automotive die. First, a binocular stereo vision positioning system was designed and modeled, from which a method of background subtraction was proposed to extract the edge line of the foreground area. Furthermore, the intersection point of the workpiece contour line was taken as the characteristic point of the workpiece, and an algorithm that combines epipolar constraint with gray value similarity was proposed to quickly and accurately realize the feature points matching. Finally, experiments show that the workpiece can be positioned accurately, and that the precision of position recognition could be controlled within $\pm 0.5 mm$ when the camera was 1 m away from the workpiece, meeting the requirement of robot processing.

Full Article | PDF Article

More Like This

Three-dimensional reconstruction of mobile binocular stereo vision based on push-broom line structured light for a workpiece surface

Yue Wang, Xueyou Han, Jing Rui, Hailan Zhang, Lei Yin, Xuefeng Zhang, and Xiangjun Wang
J. Opt. Soc. Am. A 40(11) 2096-2107 (2023)

High precision measurement for the chamfered hole radius and spacing of a large-size workpiece based on binocular vision combined with plane dynamic adjustment

Junhui Huang, Miaowei Qi, Zhao Wang, Chao Xing, and Jianmin Gao
Appl. Opt. 60(29) 9232-9240 (2021)

Optimization-based non-cooperative spacecraft pose estimation using stereo cameras during proximity operations

Limin Zhang, Feng Zhu, Yingming Hao, and Wang Pan
Appl. Opt. 56(15) 4522-4531 (2017)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (27)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (10)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (19)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Items	Parameters	Value
Left and right camera	Camera Resolution	PointGray BFLY-50A2M $2592 \times 1944$ pixel
	Frame Rate	13 fps
	Sensor Type	CMOS
	Sensor Size $(2 T_{x} \times 2 T_{y})$	$5.8 mm \times 4.3 mm$
	Pixel Size	$2.2 um \times 2.2 um$
	Focal Length	6 mm
Structural parameters	Distance $2 L$	300 mm
Structural parameters	Angle $2 α$	16°

Intrinsic Parameters	Left Camera	Right Camera
$f_{c}$	[3780.39161; 3786.66161]	[3762.52812; 3770.94509]
$f_{c_} error$	[4.95396; 4.03090]	[4.45253; 4.26426]
cc	[1335.89287; 838.41230]	[1260.04705; 795.17519]
$c c_error$	[3.85301; 6.56333]	[3.49159; 5.5482]
$k_{c}$	[−0.12175; 0.18301; 0; −0.00695; 0.00201]	[−0.08015; −0.04991;0; −0.00830; −0.0004]
$k_{c_} error$	[0.01369; 0.13495; 0;0.00107; 0.00090]	[0.01123; 0.07463; 0; 0.00100; 0.00102]

Structural Parameters
$O m / rad$	$[\begin{matrix} - 0.00632 & - 0.27143 & 0.00621 \end{matrix}]$
$o m_error / rad$	$[\begin{matrix} 0.00148 & 0.00217 & 0.00163 \end{matrix}]$
$T / mm$	$[\begin{matrix} 298.13384 & 0.74195 & 29.46139 \end{matrix}]$
$T_error / mm$	$[\begin{matrix} 0.49562 & 0.46237 & 0.31758 \end{matrix}]$

Feature Points	$p_{1}$	$p_{2}$	$p_{3}$	$p_{4}$	$p_{5}$	$p_{6}$	$p_{7}$	$p_{8}$	$p_{9}$	$p_{10}$	$p_{11}$
$q_{1}$	58.463	147.261	103.319	6.068	103.977	51.587	56.109	103.613	42.150	145.363	189.523
$q_{2}$	116.220	205.333	46.140	63.638	46.801	109.318	1.239	46.435	15.248	88.335	132.652
$q_{3}$	165.323	254.706	2.473	112.582	1.810	158.401	49.995	2.177	64.046	39.850	84.302
$q_{4}$	8.389	96.913	152.893	43.844	153.549	1.534	105.829	153.186	91.913	194.806	238.829
$q_{5}$	25.909	62.427	186.848	78.031	187.503	32.750	139.884	187.140	125.998	228.673	272.601
$q_{6}$	164.285	253.662	1.445	111.548	0.782	157.363	48.964	1.149	63.015	40.875	85.324
$q_{7}$	149.204	237.713	12.056	96.978	12.712	142.349	35.002	12.348	48.916	53.964	97.982
$q_{8}$	206.176	295.007	44.329	153.761	43.671	199.297	91.559	44.035	105.524	2.267	41.911
$q_{9}$	251.130	340.215	88.819	198.565	88.159	244.231	136.184	88.524	150.189	46.637	2.331
$q_{10}$	102.041	190.284	58.732	49.973	59.386	95.207	11.816	59.024	2.056	100.514	144.398
$q_{11}$	102.655	190.902	58.123	50.585	58.778	95.822	11.206	58.415	2.667	99.907	143.793
$q_{12}$	13.162	102.599	149.785	39.610	150.448	6.236	102.235	150.081	88.175	192.132	236.609
$q_{13}$	167.314	256.689	4.476	114.577	3.813	160.392	51.994	4.180	66.045	37.844	82.292

Feature Points	$p_{1}$	$p_{2}$	$p_{3}$	$p_{4}$	$p_{5}$	$p_{6}$	$p_{7}$	$p_{8}$	$p_{9}$	$p_{10}$	$p_{11}$
$q_{1}$	0.014	0.527	0.287	0.384	0.287	0.593	0.169	0.403	0.368	0.434	−0.829
$q_{2}$	−0.404	0.699	0.160	−0.316	0.160	0.764	0.964	0.140	−0.560	−0.502	−0.362
$q_{3}$	−0.677	0.632	0.867	−0.421	0.867	0.282	0.366	0.874	−0.158	−0.062	−0.609
$q_{4}$	−0.043	0.686	−0.063	0.205	−0.063	0.980	0.751	−0.007	−0.032	0.014	−0.689
$q_{5}$	−0.831	0.176	0.372	−0.965	0.372	−0.134	0.368	0.237	−0.684	−0.636	0.246
$q_{6}$	−0.784	0.678	0.815	−0.549	0.815	0.352	0.460	0.766	−0.228	−0.112	−0.615
$q_{7}$	0.114	−0.574	−0.373	−0.198	−0.373	−0.680	−0.338	−0.425	−0.375	−0.421	0.957
$q_{8}$	0.534	−0.389	−0.272	0.599	−0.272	−0.243	−0.637	−0.210	0.894	0.944	−0.194
$q_{9}$	0.163	−0.631	−0.440	−0.137	−0.440	−0.695	−0.398	−0.475	−0.333	−0.386	0.978
$q_{10}$	0.562	−0.393	−0.211	0.631	−0.211	−0.205	−0.626	−0.146	0.955	0.956	−0.271
$q_{11}$	0.593	−0.524	−0.257	0.665	−0.257	−0.383	−0.810	−0.194	0.871	0.843	−0.038
$q_{12}$	0.811	−0.171	−0.463	0.954	−0.463	0.181	−0.368	−0.320	0.670	0.619	−0.240
$q_{13}$	−0.690	0.297	0.762	−0.795	0.762	−0.095	0.380	0.621	−0.450	−0.420	−0.093

Feature Points	PCS1/Pixel		PCS2/Pixel		CCS1/mm
Feature Points	$u_{1}$	$v_{1}$	$u_{2}$	$v_{2}$	$x_{c}$	$y_{c}$	$z_{c}$
3	1459	945.7	890.4	926.1	32.971	28.571	1011.747
5	1106	826.3	562.2	801.8	–58.752	–3.409	964.583
6	1188	915.2	659.6	895.5	–36.843	19.101	940.316
9	1285	734.7	732.7	711.8	–13.248	–27.277	983.939
10	1371	820	831.5	800.1	8.911	–4.826	959.825
11	1451	778.4	905.6	757.6	29.470	–15.738	968.530
12	1362	693	804.7	673.7	6.847	–38.124	993.102

Distance Point Num.	Binocular Vision Measurement (mm)	Designed Dimension (mm)	Error (mm)
$d_{5 - 6}$	39.6931	40.00	−0.3069
$d_{5 - 12}$	80.0454	80.00	0.0454
$d_{5 - 9}$	55.1954	55.00	0.1954
$d_{6 - 10}$	54.9088	55.00	−0.0912
$d_{6 - 11}$	79.5089	80.00	−0.4911
$d_{9 - 10}$	39.7049	40.00	−0.2951
$d_{10 - 11}$	24.8504	25.00	−0.1496
$d_{9 - 12}$	24.6050	25.00	−0.3950
$d_{11 - 12}$	40.2083	40.00	0.2083

Num.	Position 1 CCS1 (mm)	Position 2 CCS1 (mm)	Position 3 CCS1 (mm)
1	—	—	—
2	—	(−39.65, 247.99, 825.71)	(182.32, 124.57, 990.48)
3	(32.97, 28.57, 1011.75)	(34.43, 250.76, 856.17)	—
4	—	—	—
5	(−58.75, −3.41, 964.58)	(−39.93, 176.14, 832.85)	(161.13, 57.47, 974.43)
6	(−36.84, 19.10, 940.31)	(−27.10, 194.92, 800.41)	(178.23, 82.47, 947.60)
7	—	(−3.44, 195.73, 809.71)	—
8	—	(−16.65, 177.11, 842.42)	—
9	(−13.25, −27.28, 983.94)	(10.88, 178.30, 854.12)	(210.54, 38.29, 988.45)
10	(8.91, −4.83, 959.83)	(24.07, 196.90, 821.00)	(227.36, 63.18, 962.41)
11	(29.47, −15.74, 968.53)	(46.94, 197.90, 830.84)	(249.93, 54.31, 969.03)
12	(6.85, −38.12, 993.10)	(34.28, 179.26, 863.76)	(232.97, 29.48, 994.71)

	Position Under Camera Coordinate System CCS1 (mm)	Pose Under Camera Coordinate System CCS1 (rad)
Position 1	(−55.61, 38.83, 1006.76)	(2.1489, 1.1138, −0.5335)
Position 2	(−51.91, 229.20, 858.75)	(1.5900, 1.3001, −0.5713)
Position 3	(165.41, 99.91, 1017.05)	(2.0898, 1.2731, −0.6076)

Position	Position Under Camera Coordinate System CCS1 (mm)	Pose Under Camera Coordinate System CCS1 (rad)
1	(214.16, 52.50, 1166.33)	(−0.1270, 0.1323, −0.4248)
2	(76.47, 114.16, 1243.56)	(−0.7560, −0.2725, −2.8310)
3	(259.71, 329.41, 1025.58)	(1.1323, 0.9386, 2.2061)
4	(195.57, 260.69, 1141.82)	(−0.2014, −0.9096, −1.5856)

Abstract

Cited By

Figures (27)

Tables (10)

Equations (19)

Applied Optics