UAV system with trinocular vision for external obstacle detection of
transmission lines

Yunpeng Ma; Zhihong Yu; Yaqin Zhou; Qingwu Li; Qingwu Li; Qingwu Li; Yi Wu

doi:10.1364/AO.446141

Applied Optics
Vol. 61,
Issue 12,
pp. 3297-3311
(2022)
•https://doi.org/10.1364/AO.446141

UAV system with trinocular vision for external obstacle detection of transmission lines

Yunpeng Ma, Zhihong Yu, Yaqin Zhou, Qingwu Li, and Yi Wu

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Yunpeng Ma, Zhihong Yu, Yaqin Zhou, Qingwu Li, and Yi Wu, "UAV system with trinocular vision for external obstacle detection of transmission lines," Appl. Opt. 61, 3297-3311 (2022)

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

Check for updates

Abstract

External obstacle detection is a significant task in transmission line inspection and is related to the safe operation of the power transmission grid. In recent years, unmanned aerial vehicles (UAVs) equipped with different devices have been widely used for transmission line inspection. However, because of the complex environment of transmission lines and weak power line textures in the obtained images, most existing methods and systems cannot meet the requirements for real-time and high-accuracy external obstacle detection of transmission lines. In this paper, a novel, to the best of our knowledge, UAV system integrated trinocular vision technology with remote sensing is developed to achieve better external obstacle detection of transmission lines in real time, which is composed of a DJ-Innovations (DJI) UAV equipped with a global positioning system (GPS), angle sensors, trinocular vision including three visible cameras with the same parameters, and a small processor with a pre-implanted software algorithm. In this paper, a new method for external obstacle detection of transmission lines is proposed to satisfy the requirements for real-time and high-accuracy practical inspection applications. First, the original trinocular images need to be rectified. Then, the rectified trinocular images are adopted to achieve three-dimensional reconstruction of power lines. Finally, based on trinocular vision, bag of feature, and GPS, the clearance distance measurement, obstacle classification, and obstacle location are realized. Experimental tests on 220 kV transmission lines reveal that our proposed system can be applied in practical inspection environments and has good performance.

Full Article | PDF Article

More Like This

Trinocular vision and spatial prior based method for ground clearance measurement of transmission lines

Yaqin Zhou, Qingwu Li, Yi Wu, Yunpeng Ma, and Chunkuan Wang
Appl. Opt. 60(8) 2422-2433 (2021)

Dense range map reconstruction from a versatile robotic sensor system with an active trinocular vision and a passive binocular vision

Min Young Kim, Hyunkee Lee, and Hyungsuck Cho
Appl. Opt. 47(11) 1927-1939 (2008)

Efficient 3D scanning measurement system based on asymmetric trinocular vision and a multi-line laser

Maosen Wan, Renhao Zheng, Shuaidong Wang, He Huang, Huining Zhao, and Liandong Yu
Appl. Opt. 62(8) 2145-2153 (2023)

Previous Article Next Article

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

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 (12)

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 (5)

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

Input:	$①$ Power line point $P_{line} (x_{p}, y_{p})$ ; $②$ Its disparity value: $d_{p_{line}}$ ; $③$ Shooting angle: $θ_{x}$ and $θ_{z}$ ; $④$ Segmentation map by OTSU [38]: $I_{seg}$ .
1:	Compute the plumb line $L_{p}$ of $p_{line}$ : $L_{p} : ρ = x \cdot \cos ϕ + y \cdot \sin ϕ$ .
2:	Compute the initial point of the possible obstacle below $p_{line}$ using $I_{seg}$ : $P_{ok} (x_{0}, y_{0})$ .
3:	Compute the disparity $d_{k}$ of $P_{ok} (x_{0}, y_{0})$ .
4:	Determine the coordinate of the obstacle $P_{e}^{o} (x_{e}, y_{e})$ closest to point $p_{line}$ in space:
	Set $P_{ok}$ as the initial reference point.
	For iteration: $i = 1 : n$ .
	(a) Determine next point in $L_{p} : P_{oi} (x_{i}, y_{i})$ , $x_{i} = x_{i - 1} + 5 p i x e l, x_{i} < W, W$ denotes width of image.
	(b) Compute the differences between $p_{line}$ and $P_{oi}$ in $O_{W} - X_{W} Y_{W} Z_{W} : △ X_{i}^{W}, △ Z_{i}^{W}$ .
	(c) If ${\begin{cases} Δ X_{W} \leq \frac{b \cdot δ}{2} \\ Δ Z_{W} \leq \frac{b \cdot δ}{2} \end{cases}$ , compute distance ${D i s}_{p} = \sqrt{△ {(X_{i}^{W})}^{2} + △ {(Y_{i}^{W})}^{2} + △ {(Z_{i}^{W})}^{2}}$ . Set $P_{e}^{o} = k_{i} (x_{i}, y_{i})$ .
	End for.
Output:	$①$ Obstacle coordinate $P_{e}^{o}$ ; $②$ Clearance distance ${D i s}_{p}$ ;

Method	Obstacle Type	Maximum Absolute Error (m)	Minimum Absolute Error (m)	Mean Absolute Error (m)
Ours	Tree	0.4335	0.1082	0.3107
	Building	0.4286	0.0980	0.3084
	Ground	0.4109	0.0940	0.2968
	All	0.4335	0.0940	0.3053
LiDAR	Tree	0.3959	0.0783	0.2904
	Building	0.3804	0.0654	0.2835
	Ground	0.3877	0.0667	0.2802
	All	0.3880	0.0701	0.2847

Method	Obstacle Type	Total Number	Accurate Number	Accuracy Rate (%)
Ours	Tree	50	47	94.00%
	Building	30	28	93.33%
	Ground	20	19	95.00%
	All	100	94	94.00%
LiDAR	Tree	50	49	98.00%
	Building	30	29	96.66%
	Ground	20	17	85.00%
	All	100	95	95.00%

Method	Error Type	Mean Error	Standard Deviation
Ours	Longitude error	$- 9 .5586 \times 10^{- 6}$	$3 .6736 \times 10^{- 5}$
Ours	Latitude error	$6 .8594 \times 10^{- 6}$	$5 .0684 \times 10^{- 5}$
LiDAR	Longitude error	$- 1 .1085 \times 10^{- 7}$	$2 .2869 \times 10^{- 5}$
LiDAR	Latitude error	$6 .0134 \times 10^{- 6}$	$3 .8492 \times 10^{- 5}$

Input:	$①$ Power line point $P_{line} (x_{p}, y_{p})$ ; $②$ Its disparity value: $d_{p_{line}}$ ; $③$ Shooting angle: $θ_{x}$ and $θ_{z}$ ; $④$ Segmentation map by OTSU [38]: $I_{seg}$ .
1:	Compute the plumb line $L_{p}$ of $p_{line}$ : $L_{p} : ρ = x \cdot \cos ϕ + y \cdot \sin ϕ$ .
2:	Compute the initial point of the possible obstacle below $p_{line}$ using $I_{seg}$ : $P_{ok} (x_{0}, y_{0})$ .
3:	Compute the disparity $d_{k}$ of $P_{ok} (x_{0}, y_{0})$ .
4:	Determine the coordinate of the obstacle $P_{e}^{o} (x_{e}, y_{e})$ closest to point $p_{line}$ in space:
	Set $P_{ok}$ as the initial reference point.
	For iteration: $i = 1 : n$ .
	(a) Determine next point in $L_{p} : P_{oi} (x_{i}, y_{i})$ , $x_{i} = x_{i - 1} + 5 p i x e l, x_{i} < W, W$ denotes width of image.
	(b) Compute the differences between $p_{line}$ and $P_{oi}$ in $O_{W} - X_{W} Y_{W} Z_{W} : △ X_{i}^{W}, △ Z_{i}^{W}$ .
	(c) If ${\begin{cases} Δ X_{W} \leq \frac{b \cdot δ}{2} \\ Δ Z_{W} \leq \frac{b \cdot δ}{2} \end{cases}$ , compute distance ${D i s}_{p} = \sqrt{△ {(X_{i}^{W})}^{2} + △ {(Y_{i}^{W})}^{2} + △ {(Z_{i}^{W})}^{2}}$ . Set $P_{e}^{o} = k_{i} (x_{i}, y_{i})$ .
	End for.
Output:	$①$ Obstacle coordinate $P_{e}^{o}$ ; $②$ Clearance distance ${D i s}_{p}$ ;

Abstract

Data availability

Cited By

Figures (12)

Tables (5)

Equations (19)

Applied Optics