Visual mapping of internal pipe walls using sparse features for application on board Autonomous Underwater Vehicles

Bodenmann, Adrian; Thornton, Blair; Ura, Tamaki; Painumgal, Unnikrishnan V.

doi:10.1109/oceanse.2009.5278231

Cited by 5 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Piciarelli et al [5] directly obtained undistorted flattened images of a pipeline internal surface using the mapping relationship between the right angle and cylindrical and spherical coordinate systems after a pipeline robot captured panoramic images of the pipeline internal surface. Bodenmann et al [6] first mapped the captured panoramic images of the pipeline internal surface to a pipeline 3D model and thereafter drew flattened images of the pipeline based on this 3D model. But, these methods are not suitable for flattening the external surfaces of pipelines captured by an inspection UAV.…”

Section: Geometric Distortion Correction Techniquementioning

confidence: 99%

“…However, the method relies on parallelism and equal line spacing information of printed text on the page and is therefore not suitable for pipeline surface image flattening tasks that lack text information. Currently, there are relatively few studies on geometric distortion correction (GDC) algorithms for pipeline surface images; most studies have focused on GDC tasks for internal pipeline surface images [4][5][6][7]. These methods first require a crawling robot to obtain a panoramic image of the internal pipeline surface; therefore, they are not applicable to GDC tasks for pipeline external surface images.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Correction for geometric distortion in the flattened representation of pipeline external surface

Cheng,

Zhong,

Tan

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

Conventional cameras often produce external pipeline surface images with significant perspective-projection distortions, making determining the type and size of defects on the pipeline surface difficult. To address this issue, we present a novel approach for correcting the geometric distortion of a constant diameter pipeline (CDP) external surface image based on apparent contour pairs (ACPs). First, the camera is calibrated and the real projection position of ACPs accurately extracted. Second, a single-view perspective projection model of the CDP is adopted to realize a three-dimensional reconstruction of the pipeline. Subsequently, the flattened image of the CDP's outer surface is obtained by performing equidistant sampling of cross sections, determining the visibility of discrete points, and mapping projection points onto a flat surface. Simulations and real experiments demonstrated the high accurate method in the correction of CDP images. Experiments conducted under different outdoor scenarios confirmed the robustness of the proposed method.

show abstract

Section: Geometric Distortion Correction Techniquementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Correction for geometric distortion in the flattened representation of pipeline external surface

Cheng,

Zhong,

Tan

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

show abstract

“…6) Underwater vehicles: Submersible robots have been developed to inspect underwater structures. Several prototypes have been designed to inspect oil storage tanks (Abdulla, et al 2010) and pipelines (Conte, et al 1996, Bodenmann, et al 2009.…”

Section: Robotic Inspection For Civil Infrastructurementioning

confidence: 99%

Survey on robotics and automation technologies for civil infrastructure

Myung¹,

Wang²,

Kang³

et al. 2014

Smart Structures and Systems

View full text Add to dashboard Cite

Over the past several decades, substantial amounts of sensors and sensing systems have been developed for civil infrastructure systems. This special issue focuses on state-of-the-art robotics and automation technologies, including construction automation, robotics, instrumentation, monitoring, inspection, control, and rehabilitation for civil infrastructure. The issue also covers construction informatics supporting sensing, analysis and design activities needed to operate smart and sustainable civil infrastructure. Examples include robotic systems applied to civil infrastructure and equipped with various sensing technologies, such as optical sensors, laser sensors, wireless sensors, multi-sensor fusion, etc. This special issue is published in an effort to disseminate current advances of various robotics and automation technologies for civil infrastructure and built environment.

show abstract

“…2. The fish eye lens of the camera has a field of view of 180 degrees, which permits the camera to view the pipe wall interior which can be used for visual mapping and inspection of its interior [10]. This also maximizes the operating range of the sensor as it allows the camera to see the full laser circle without clipping even during displaced, pitched and yawed conditions.…”

Section: Sensor Descriptionmentioning

confidence: 99%

A conical laser light-sectioning method for navigation of Autonomous Underwater Vehicles for internal inspection of pipelines

et al. 2009

Self Cite

View full text Add to dashboard Cite

This paper presents a novel sensing method for navigation of Autonomous Underwater Vehicles, AUVs, through pipelines to conduct autonomous internal inspection. Unlike remotely operated pipe inspection robots, AUVs do not have an umbilical cable and so they can easily maneuver through bent sections of pipelines and the distance they can cover is not restricted by the length of the cable. Presently pipe inspection robots come in contact with the walls of the pipe during their operation. However old and aged pipelines may have loose corroded materials or biological growth which may get detached and pollute the fluid or further damage pipe interior when the pipe inspections robots touch the walls. AUVs can operate without coming in contact with the pipe wall and so this technique is a non-contact measurement and inspection technique. The proposed navigation sensor makes use of computer vision techniques to estimate the relative position and orientation of the vehicle inside the pipe with 4 degrees of freedom, which will enable the AUV to swim through the center of the pipe. A conical laser is projected on the pipe wall and the image of the laser is acquired by a camera. The features of the image are extracted and matched with a feature database prepared apriori. The position and orientation of the matching feature record in the database gives the estimated position and orientation of the vehicle. Experiments were conducted in dry experimental pipelines to verify the performance of the proposed sensor and results are presented. Control Simulations were performed to verify the ability of the sensor to navigate an AUV. The results of these controls simulated are also presented in this paper.Index terms-Image features, feature extraction, shape signatures, image feature database, navigation sensor, pipe inspection, autonomous underwater vehicle.I.

show abstract

Visual mapping of internal pipe walls using sparse features for application on board Autonomous Underwater Vehicles

Cited by 5 publications

References 1 publication

Correction for geometric distortion in the flattened representation of pipeline external surface

Correction for geometric distortion in the flattened representation of pipeline external surface

Survey on robotics and automation technologies for civil infrastructure

A conical laser light-sectioning method for navigation of Autonomous Underwater Vehicles for internal inspection of pipelines

Contact Info

Product

Resources

About