3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion

Research output: Contribution to conferencePaper

Abstract

During planned, periodic outages, a selection of fuel channels within the UK fleet of Advanced Gas-cooled Reactor (AGR) cores are inspected using specialist tools which record video footage and other sensory data for each channel which undergoes inspection. Current visualization techniques comprise of manually produced montages by inspection engineers of points of interest (i.e. structural defects) and 2-D panoramic images of the fuel channels automatically produced using bespoke image stitching software. Both techniques however provide limited structural information due to the loss of depth data as a result of the image formation process. By recovering the depth information from the footage, a 3-D model could be constructed and subsequently, allow for more accurate profiling of specific defects observed during inspection in addition to obtaining the fuel channels structure using existing footage. This work explores the preliminary application of a 3-D visualization technique known as Structure-from-Motion (SfM) which aims to obtain 3-D information by exploiting image correspondences across multiple viewpoints of the same scene in the RVI footage. This paper investigates the difficulties of applying state-of-the-art SfM to RVI footage and we present new techniques to improve feature correspondence searching in repetitive, non-descript environments.

Conference

Conference10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017
Abbreviated titleNPIC and HMIT 2017
CountryUnited States
CitySan Francisco
Period11/06/1715/06/17
Internet address

Fingerprint

Gas cooled reactors
Structure from Motion
Nuclear Reactor
3D
Inspection
Visualization
Defects
Correspondence
Reactor cores
Data Depth
Outages
Stitching
3D Visualization
Image processing
Profiling
3D Model
Reactor
Engineers
Software
Gas

Keywords

  • visual inspection
  • Structure-from-Motion
  • Image processing
  • feature detection

Cite this

Law, K., West, G., Murray, P., & Lynch, C. (2017). 3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion. Paper presented at 10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, San Francisco, United States.
Law, Kristofer ; West, Graeme ; Murray, Paul ; Lynch, Chris. / 3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion. Paper presented at 10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, San Francisco, United States.10 p.
@conference{6b3584510ac14259b9c5bb9fabf90ba5,
title = "3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion",
abstract = "During planned, periodic outages, a selection of fuel channels within the UK fleet of Advanced Gas-cooled Reactor (AGR) cores are inspected using specialist tools which record video footage and other sensory data for each channel which undergoes inspection. Current visualization techniques comprise of manually produced montages by inspection engineers of points of interest (i.e. structural defects) and 2-D panoramic images of the fuel channels automatically produced using bespoke image stitching software. Both techniques however provide limited structural information due to the loss of depth data as a result of the image formation process. By recovering the depth information from the footage, a 3-D model could be constructed and subsequently, allow for more accurate profiling of specific defects observed during inspection in addition to obtaining the fuel channels structure using existing footage. This work explores the preliminary application of a 3-D visualization technique known as Structure-from-Motion (SfM) which aims to obtain 3-D information by exploiting image correspondences across multiple viewpoints of the same scene in the RVI footage. This paper investigates the difficulties of applying state-of-the-art SfM to RVI footage and we present new techniques to improve feature correspondence searching in repetitive, non-descript environments.",
keywords = "visual inspection, Structure-from-Motion, Image processing, feature detection",
author = "Kristofer Law and Graeme West and Paul Murray and Chris Lynch",
year = "2017",
month = "6",
day = "11",
language = "English",
note = "10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, NPIC and HMIT 2017 ; Conference date: 11-06-2017 Through 15-06-2017",
url = "http://npic-hmit2017.org/",

}

Law, K, West, G, Murray, P & Lynch, C 2017, '3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion' Paper presented at 10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, San Francisco, United States, 11/06/17 - 15/06/17, .

3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion. / Law, Kristofer; West, Graeme; Murray, Paul; Lynch, Chris.

2017. Paper presented at 10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, San Francisco, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - 3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion

AU - Law, Kristofer

AU - West, Graeme

AU - Murray, Paul

AU - Lynch, Chris

PY - 2017/6/11

Y1 - 2017/6/11

N2 - During planned, periodic outages, a selection of fuel channels within the UK fleet of Advanced Gas-cooled Reactor (AGR) cores are inspected using specialist tools which record video footage and other sensory data for each channel which undergoes inspection. Current visualization techniques comprise of manually produced montages by inspection engineers of points of interest (i.e. structural defects) and 2-D panoramic images of the fuel channels automatically produced using bespoke image stitching software. Both techniques however provide limited structural information due to the loss of depth data as a result of the image formation process. By recovering the depth information from the footage, a 3-D model could be constructed and subsequently, allow for more accurate profiling of specific defects observed during inspection in addition to obtaining the fuel channels structure using existing footage. This work explores the preliminary application of a 3-D visualization technique known as Structure-from-Motion (SfM) which aims to obtain 3-D information by exploiting image correspondences across multiple viewpoints of the same scene in the RVI footage. This paper investigates the difficulties of applying state-of-the-art SfM to RVI footage and we present new techniques to improve feature correspondence searching in repetitive, non-descript environments.

AB - During planned, periodic outages, a selection of fuel channels within the UK fleet of Advanced Gas-cooled Reactor (AGR) cores are inspected using specialist tools which record video footage and other sensory data for each channel which undergoes inspection. Current visualization techniques comprise of manually produced montages by inspection engineers of points of interest (i.e. structural defects) and 2-D panoramic images of the fuel channels automatically produced using bespoke image stitching software. Both techniques however provide limited structural information due to the loss of depth data as a result of the image formation process. By recovering the depth information from the footage, a 3-D model could be constructed and subsequently, allow for more accurate profiling of specific defects observed during inspection in addition to obtaining the fuel channels structure using existing footage. This work explores the preliminary application of a 3-D visualization technique known as Structure-from-Motion (SfM) which aims to obtain 3-D information by exploiting image correspondences across multiple viewpoints of the same scene in the RVI footage. This paper investigates the difficulties of applying state-of-the-art SfM to RVI footage and we present new techniques to improve feature correspondence searching in repetitive, non-descript environments.

KW - visual inspection

KW - Structure-from-Motion

KW - Image processing

KW - feature detection

UR - http://www.ans.org/pubs/proceedings/

UR - http://npic-hmit2017.org/

M3 - Paper

ER -

Law K, West G, Murray P, Lynch C. 3-D advanced gas-cooled nuclear reactor fuel channel reconstruction using structure-from-motion. 2017. Paper presented at 10th International Topical Meeting on Nuclear Plant Instrumentation, Control, and Human-Machine Interface Technologies, NPIC and HMIT 2017, San Francisco, United States.