Name
Abstract | ||
|---|---|---|
In recent years, machine learning has become a hot research area and researchers in the field of electrical capacitance tomography (ECT) have also extended machine learning theory to the solution of ECT image reconstruction problem. In this paper, a deep convolutional neural network is constructed for ECT image reconstruction, which can not only solve the forward problem, but also the inverse problem of ECT. The convolutional network consists of two sub-networks. The sub-network for estimating capacitance from permittivity distribution image is mainly composed of convolutional layers and pooling layers, which is called encoder. The sub-network for reconstructing permittivity distribution image from capacitance is composed of full-connected layers, which is called decoder. Testing results show that the proposed CNN has high capacitance estimation accuracy and high image reconstruction quality, along with good generalization ability. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/IST48021.2019.9010096 | 2019 IEEE International Conference on Imaging Systems and Techniques (IST) |
Keywords | DocType | ISSN |
electrical capacitance tomography,machine learning,deep learning,convolutional neural network,image reconstruction | Conference | 1558-2809 |
ISBN | Citations | PageRank |
978-1-7281-3869-5 | 1 | 0.35 |
References | Authors | |
4 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jin Zheng | 1 | 1 | 0.35 |
| Haocheng Ma | 2 | 1 | 0.35 |
| Lihui Peng | 3 | 12 | 5.36 |