Fully Convolutional Deep Network Architectures for Automatic Short Glass Fiber Semantic Segmentation from CT scans
2019-01-04Unverified0· sign in to hype
Tomasz Konopczyński, Danish Rathore, Jitendra Rathore, Thorben Kröger, Lei Zheng, Christoph S. Garbe, Simone Carmignato, Jürgen Hesser
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
We present the first attempt to perform short glass fiber semantic segmentation from X-ray computed tomography volumetric datasets at medium (3.9 m isotropic) and low (8.3 m isotropic) resolution using deep learning architectures. We performed experiments on both synthetic and real CT scans and evaluated deep fully convolutional architectures with both 2D and 3D kernels. Our artificial neural networks outperform existing methods at both medium and low resolution scans.