The effect of convolutional neural network architectures on phase segmentation of composite material X-ray micrographs
The effect of convolutional neural network architectures on phase segmentation of composite material X-ray micrographs
复制标题
DOI：
10.1177/00219983231168790
复制
发表时间：
2023-05
影响因子：
2.9
通讯作者：
Pedro Galvez-Hernandez;J. Kratz
中科院分区：
文献类型：
--
作者：
Pedro Galvez-Hernandez;J. Kratz
孔隙率严重降低了复合材料层合板的力学性能，自动分割孔隙相的方法也在不断发展。本研究探讨复合材料的孔隙率，采取层间空隙和干丝束地区的形式。深度学习用于X射线显微照片的分割，通过实施八个最先进的卷积神经网络（CNN）架构，这些架构使用包含25个，50个和100个图像的数据集进行训练。通过优化六个相关超参数，包括应用于输出概率图的截止概率，实现了为每个架构和训练集大小提供最高准确度的超参数组合。此外，CNN架构的属性（例如，层类型、连接、密度.）被发现不仅在分割结果中而且在相关的计算工作中起决定性作用。U-Net和FCDenseNet的性能优于FCN-8 s、FCN-16、SegNet、LinkNet、ResNet 18和Xception CNN架构。然而，CNN通常优于标准阈值方法，特别是在包含低孔隙率（1.07%）的子体积中，其中对强度的影响在高性能复合材料中非常敏感。在低孔隙度样本中，U-Net和FCDenseNet始终将空隙分割到85%以上的准确度，而阈值处理的准确度只有一半，约为40%。结果提供了一个强大的动机，以取代阈值分割方法的复合X射线显微照片。在效率方面，与FCDenseNet相比，U-Net网络的复杂性降低，平均减少了训练时间（-36%）和预测时间（-17%）。
Porosity severely reduces the mechanical performance of composite laminates and methods for automatic segmentation of void phases are growing. This study investigates porosity in composite materials that take the form of interlaminar voids and dry tow areas. Deep Learning was used for the segmentation of X-ray micrographs via the implementation of eight state-of-the-art Convolutional Neural Network (CNN) architectures trained with data sets containing twenty-five, fifty, and one-hundred images. The combination of hyperparameters providing the highest accuracy for each architecture and training set size was achieved through the optimisation of six relevant hyperparameters, including the cut-off probability applied to output probability maps. Additionally, the properties of the CNN architectures (e.g., layer typology, connections, density…) were found to play a determining role, not only in the segmentation results but also in the associated computing effort. U-Net and FCDenseNet outperformed the FCN-8s, FCN-16, SegNet, LinkNet, ResNet18 and Xception CNN architectures. However, the CNNs generally outperformed the standard thresholding approaches, especially in sub-volumes containing low porosity (1.07%) where the influence on strength is very sensitive in high-performance composites. In low porosity samples, U-Net and FCDenseNet consistently segmented voids to 85% + accuracy, whereas thresholding was only half as accurate, at around 40%. The results provide a strong motivation to replace thresholding as a segmentation method for composite X-ray micrographs. In terms of efficiency, the reduced complexity of the U-Net network allowed for an average reduction of the training time (−36%) and prediction time (−17%) when compared to FCDenseNet.