期刊
POWDER TECHNOLOGY
卷 392, 期 -, 页码 296-305出版社
ELSEVIER
DOI: 10.1016/j.powtec.2021.07.005
关键词
Granular particle; Particle detection; Mask R-CNN; Instance segmentation; Shape evaluation
资金
- National Natural Science Foundation of China [51878416]
- RGC HONG KONG [:15209119, R5037-18F]
- Changsha Municipal Bureau of Science and Technology Innovation Fund [kq2008001, kq2008002]
- Key Program of Department of Transportation of Jiangxi Province [:2019C0011]
This study developed a systematic tool based on deep learning and computational geometry to evaluate and identify the shape of granular materials. By establishing image datasets with labeled masks and employing the Mask R-CNN model, successful identification and shape analysis of particles were achieved.
Particle identification and shape evaluation of granular materials from their realistic packing images are challenging and of great interest to many engineers and researchers. In this study, a systematic tool is developed based on computing techniques, including deep learning and computational geometry. First, image datasets of the target granular particles with well-labeled masks are established. The Mask Region Convolutional Neural Network (Mask R-CNN) is employed to implement the end-to-end instance segmentation and contour extraction of particles on different realistic images. Since Mask R-CNN models have several different feature extraction backbones, the optimal model is selected and then trained on the established datasets using transfer learning technique. After the particles are successfully identified from images of cobble and ballast, the elongation, angularity, and roughness are evaluated and the statistical shape analysis is conducted. The proposed method has strong generalization ability, especially for densely-packed particles. (c) 2021 Published by Elsevier B.V.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据