Underwater pipeline and cable inspection using autonomous underwater vehicles

Jacobi, Marco; Karimanzira, Divas

doi:10.1109/oceans-bergen.2013.6608089

Cited by 30 publications

(14 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The automation of Subsea Pipeline Inspection relies on data collected from the instrumentation mounted on Remote Operating Vehicles. Jacobi et al [10], [11] proposed a pipeline tracking method for Autonomous Underwater Vehicle (AUV) guidance by fusing optical, magnetic, and acoustic sensor data collected in a simulation setting. Bharti et al [12] used multibeam echo-sounder data for pipeline detection and estimation of orientation.…”

Section: Related Workmentioning

confidence: 99%

A Comparison of the Performance of 2D and 3D Convolutional Neural Networks for Subsea Survey Video Classification

Stamoulakatos¹,

Cardona²,

Michie³

et al. 2021

OCEANS 2021: San Diego – Porto

View full text Add to dashboard Cite

Utilising deep learning image classification to automatically annotate subsea pipeline video surveys can facilitate the tedious and labour-intensive process, resulting in significant time and cost savings. However, the classification of events on subsea survey videos (frame sequences) by models trained on individual frames have been proven to vary, leading to inaccuracies. The paper extends previous work on the automatic annotation of individual subsea survey frames by comparing the performance of 2D and 3D Convolutional Neural Networks (CNNs) in classifying frame sequences. The study explores the classification of burial, exposure, free span, field joint, and anode events. Sampling and regularization techniques are designed to address the challenges of an underwater inspection video dataset owing to the environment. Results show that a 2D CNN with rolling average can outperform a 3D CNN, achieving an Exact Match Ratio of 85% and F1-Score of 90%, whilst being more computationally efficient.

show abstract

Section: Related Workmentioning

confidence: 99%

A Comparison of the Performance of 2D and 3D Convolutional Neural Networks for Subsea Survey Video Classification

Stamoulakatos¹,

Cardona²,

Michie³

et al. 2021

OCEANS 2021: San Diego – Porto

View full text Add to dashboard Cite

show abstract

“…La inclusión de sistemas robotizados en el entorno subacuático es muy compleja, principalmente debido a un comportamiento dinámico variable e impredecible del entorno que tiene un impacto significativo en las diferentes actividades que pueden llevarse a cabo. Una de estas actividades complejas, es la inspección de cables submarinos [7,2], debido a la dificultad de operar en un ambiente hóstil. Esto constituye una línea de investigación de interés creciente.…”

Section: Introductionunclassified

Arquitectura de control híbrida para la navegación de vehículos submarinos no tripulados

Lastra¹,

Trujillo²,

Velasco³

et al. 2020

XXXVIII Jornadas De Automática: Gijón, 6, 7, Y 8 De Septiembre De 2017

View full text Add to dashboard Cite

Todos los derechos reservados. De conformidad con lo dispuesto en la legislación vigente, podrán ser cas�gados con penas de multa y privación de libertad quienes reproduzcan o plagien, en todo o en parte, una obra literaria, ar�s�ca o cien�fica, fijada en cualquier �po y soporte, sin la precep�va autorización. JA2017 Prefacio Prefacio Las Jornadas de Automática se celebran desde hace 40 años en una universidad nacional facilitando el encuentro entre expertos en estaárea en un foro que permite la puesta en común de las nuevas ideas y proyectos en desarrollo. Al mismo tiempo, propician la siempre necesaria colaboración entre investigadores delámbito de la Ingeniería de Control y Automática, así como de campos afines, a la hora de abordar complejos proyectos de investigación multidisciplinares. En esta ocasión, las Jornadas estarán organizadas por la Universidad de Oviedo y se han celebrado del 6 al 8 de septiembre de

show abstract

“…The sensor data are analysed either by an on-board operator 1 or automatically within the underwater vehicle. 2 Although post-action inspections are necessary to identify potential risks, predictive measures should also be taken to enhance the safety of the pipelines, that is, monitoring the pipeline laying process.…”

Section: Introductionmentioning

confidence: 99%

“…Though each single feature point is sensitive to strong speckle noise in underwater sonar images, 12 the sonar's motion can be robustly perceived by watching the movement trends of a set of feature points, that is, it is possible to estimate the motion parameters by analysing their spatial distribution. From another perspective, modelling the spatial distribution provides an intuitive expression for the basic mapping function, that is, equation (2). Consequently, a closed-form solution or a gradient descent procedure could be derived.…”

Section: Introductionmentioning

confidence: 99%

Two-dimensional forward-looking sonar image registration by maximization of peripheral mutual information

Song

Herrmann

et al. 2017

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Monitoring the field of operation of an underwater vehicle is crucial during missions near the sea floor. The forwardlooking sonar is often the only available sensor for the observation of the ambient turbid water environment. Sonar image registration is not only a first step towards a panoramic mosaic but it also provides an initial motion parameter estimation for the vehicle self-localization and navigation. In this article, a peripheral mutual information (PMI) maximization method is proposed for the sonar image registration. Peripheral mutual information is inspired by regional mutual information (RMI) which makes use of the closed-form solution for the Shannon entropy by the assumption that the data vectors made of neighbouring pixels are normally distributed, an assumption that ignores correlations between the pixels in sonar images. To accommodate the fact that the neighbouring pixels show dependencies due to acoustic reverberation and dispersion, only the peripheral information in the neighbourhood of a pixel is used in peripheral mutual information for the calculation of the mutual information. Experiments show that the peripheral mutual information registration function is much smoother than that of regional mutual information. Further experiments on the two-dimensional forward-looking sonar image registration demonstrate the efficiency of peripheral mutual information.

show abstract

Underwater pipeline and cable inspection using autonomous underwater vehicles

Cited by 30 publications

References 5 publications

A Comparison of the Performance of 2D and 3D Convolutional Neural Networks for Subsea Survey Video Classification

A Comparison of the Performance of 2D and 3D Convolutional Neural Networks for Subsea Survey Video Classification

Arquitectura de control híbrida para la navegación de vehículos submarinos no tripulados

Two-dimensional forward-looking sonar image registration by maximization of peripheral mutual information

Contact Info

Product

Resources

About