Automatic Pylon Extraction Using Color-Aided Classification From UAV LiDAR Point Cloud Data

Huang, Junjun; Shen, Yueqian; Wang, Jinguo; Ferreira, Vagner G.

doi:10.1109/tim.2023.3293543

Cited by 5 publications

(2 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To solve this problem, the vertical voxel feature is used to roughly segment a tower and the shape prior of a tower section is added to refine the result [6]. To classify the tower bottom and the ground, some domain colour features are added to generate local features for better classification results [7].Some methods first extract the powerlines and then locate a tower through density features or detection of the intersection points for power lines [8][9][10]. This kind of methods need large calculations and are sensitive to interference such as trees or signal poles.…”

Section: Tower Localizationmentioning

confidence: 99%

“…These all bring great difficulties to manual acceptance check. With the rapid development of lidar technology, there has been initial progress in tower localization [3][4][5][6][7][8][9][10], the detection of tilt angle for towers [11][12][13][14][15][16][17][18], and the sag detection for power lines [19][20][21]. However, these works are mainly used for patrol inspection not for acceptance check that requires strictly high accuracy for data processing.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent acceptance check for towers of overhead transmission line based on point clouds

Ma,

Wang,

et al. 2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

The acceptance check is a key step to decide whether an overhead transmission line can be put into use. However, for the acceptance check of towers, the manual means is still adopted mostly nowadays, which is dangerous and inefficient. The main challenges for intelligent check technique are the automatic tower segmentation and intelligent requirements of some detection items. Here, based on point clouds from lidars, some intelligent methods are proposed for tower related acceptance check items. The geometric structural changing rules for towers are found and analyzed to identify and segment towers from the scene point cloud automatically. A model‐constrained method is proposed for fine filtering of towers. With point cloud registration strategies, the tilt angle of a tower is calculated intelligently and the defect detection of a tower is conducted successfully. Experiments show that the results of the proposed methods are accurate, efficient and automatic, which have potential for real applications.

show abstract

Section: Tower Localizationmentioning

confidence: 99%