TY - JOUR

T1 - Adaption of deeplab V3+ for damage detection on port infrastructure imagery

AU - Scherff, Marvin

AU - Hake, Frederic

AU - Alkhatib, Hamza

N1 - Funding Information: Funding This research was funded by the Deutsche Forschungs-gemeinschaft (DFG, German Research Foundation) - NE 1453/5-1 and as part of the Research Training Group i.c.sens [RTG 2159].

PY - 2023/4/21

Y1 - 2023/4/21

N2 - Regular inspection and maintenance of infrastructure facilities are crucial to ensure their functionality and safety for users. However, current inspection methods are labor-intensive and can vary depending on the inspector. To improve this process, modern sensor systems and machine learning algorithms can be deployed to detect defects based on rapidly acquired data, resulting in lower downtime. A quality-controlled processing chain allows to provide hence informed uncertainty assessments to inspection operators. In this study, we present several Deeplab V3+ models optimized to predict corroded segments of the quay wall at JadeWeserPort, Germany, which is a dataset from the 3D HydroMapper research project. Our models achieve generally high accuracy in detecting this damage type. Therefore, we examine the use of a Region Growing-based weakly supervised approach to efficiently extend our model to other common types in the future. This approach achieves about 90 % of the results compared to corresponding fully supervised networks, of which a ResNet-50 variant peaks at 55.6 % Intersection-over-Union regarding the test set's corrosion class.

AB - Regular inspection and maintenance of infrastructure facilities are crucial to ensure their functionality and safety for users. However, current inspection methods are labor-intensive and can vary depending on the inspector. To improve this process, modern sensor systems and machine learning algorithms can be deployed to detect defects based on rapidly acquired data, resulting in lower downtime. A quality-controlled processing chain allows to provide hence informed uncertainty assessments to inspection operators. In this study, we present several Deeplab V3+ models optimized to predict corroded segments of the quay wall at JadeWeserPort, Germany, which is a dataset from the 3D HydroMapper research project. Our models achieve generally high accuracy in detecting this damage type. Therefore, we examine the use of a Region Growing-based weakly supervised approach to efficiently extend our model to other common types in the future. This approach achieves about 90 % of the results compared to corresponding fully supervised networks, of which a ResNet-50 variant peaks at 55.6 % Intersection-over-Union regarding the test set's corrosion class.

KW - Image segmentation, Damage Detection, Deep Learning, Optimization, Supervised, Weakly Supervised

KW - Image segmentation

KW - Damage Detection

KW - Weakly Supervised

KW - Supervised

KW - Deep Learning

KW - Optimization

UR - http://www.scopus.com/inward/record.url?scp=85156203394&partnerID=8YFLogxK

U2 - 10.5194/isprs-archives-XLVIII-M-1-2023-301-2023

DO - 10.5194/isprs-archives-XLVIII-M-1-2023-301-2023

M3 - Conference article

VL - 48

SP - 301

EP - 308

JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

SN - 1682-1750

IS - 1

T2 - 39th International Symposium of Remote Sensing on Enviroment

Y2 - 24 April 2023 through 27 April 2023

ER -