Quality control of geological voxel models using experts' gaze

article




Maanen, P.P. van

Busschers, F.S.

Brouwer, A.M.

Meulen, M.J. van der

Erp, J.B.F. van
Due to an expected increase in geological voxel model data-flow and user demands, the development of improved quality control for such models is crucial. This study explores the potential of a new type of quality control that improves the detection of errors by just using gaze behavior of 12 geological experts. Gaze is used as input for an attention model that results in 'attended areas' on sliced representations of part of a geological voxel model. We compared attended areas to errors as manually marked by the experts. We found a clear match between manually marked errors and attended areas as determined using gaze. We also found that a large proportion of this match is reached within a small amount of total viewing time.
Topics



















Attention modelGazeGeological voxel modelQuality controlData flow analysisErrorsGeologyQuality assuranceAttention modelGazeGaze behaviorUser demandsVoxel modelsQuality controldetection methoderror analysisgeologyimage analysismodel testquality control
TNO Identifier
522501
Repository link
https://resolver.tno.nl/uuid:2bfb6e12-d173-49f0-a14e-84d1c3e5e94f
ISSN
00983004
Source
Computers and Geosciences, 76, pp. 50-58.
Publisher
Elsevier Ltd
Pages
50-58
Files
To receive the publication files, please send an e-mail request to TNO Repository.