A COMPARATIVE STUDY OF DEEP ARCHITECTURES FOR VOXEL SEGMENTATION IN VOLUME IMAGES

Publikation: Beitrag in Buch/Konferenzbericht/Sammelband/GutachtenBeitrag in KonferenzbandBeigetragen

Beitragende

Abstract

This study investigates the performance of eight different deep learning architectures for voxel segmentation in volume images. The motivation is to segment carbon in carbon reinforced concrete (CRC) in micro-tomography (μ-CT) data. Although there are many 3D convolutional neural networks (CNNs) available, it is not yet clear which one works best for these specific tasks. In this study, the authors compare the following networks: DenseVoxNet, HighResNet, Med3D, Residual 3D U-Net, 3D SkipDenseSeg, 3D U-Net, V-Net, and LV-Net. To provide a more general recommendation for selecting a neural network, three medical datasets were added in addition to the three CRC datasets to facilitate the selection of an appropriate network based on the dataset characteristics. The experiments emphasize the importance of the initial random state, used for example to initialize the network weights. On average, the 3D U-Net is the best generalizing CNN, followed by the Residual 3D U-Net and the 3D SkipDenseSeg. While the 3D U-Net is a good architecture to start with, the experiments show that it does not perform best on all domains. To achieve optimal results, the authors propose recommendations for selecting a 3D neural network based on the dataset attributes.

Details

OriginalspracheEnglisch
TitelISPRS Geospatial Week 2023
Redakteure/-innenN. El-Sheimy, A. A. Abdelbary, N. El-Bendary, Y. Mohasseb
Seiten1667-1676
Seitenumfang10
BandXLVIII-1/W2-2023
AuflageXLVIII-1/W2-2023
PublikationsstatusVeröffentlicht - 14 Dez. 2023
Peer-Review-StatusNein

Publikationsreihe

Reihe The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
ISSN1682-1750

Workshop

TitelISPRS Geospatial Week 2023
KurztitelGSW’2023
Dauer2 - 7 September 2023
Webseite
BekanntheitsgradInternationale Veranstaltung
OrtSemiramis Cairo Hotel
StadtCairo
LandÄgypten

Externe IDs

Mendeley 33f5e4b2-d65b-3537-b441-28beba1796ab
Scopus 85175057151

Schlagworte

Schlagwörter

  • 3D CNN, 3D segmentation, CNN, Comparison, Deep Learning, Tomography, magnetic resonance imaging (MRI)