ABSTRACT
This study presents an algorithmic concept that allows obtaining higher-quality tomographic images. The method solves the problem of imaging the interior of industrial tanks, reactors, or pipes. Research focuses on how to solve the inverse problem, which is converting measurements to images. Hybrid tomography combined electrical impedance tomography (EIT) and electrical capacitance tomography (ECT) measurements. The measurement vector was converted into images in two steps. In the first phase, the Long Short-Term Memory (LSTM) neural network was used, thanks to which raw reconstructions were obtained. A second network was then trained to convert the images obtained in the first step into enhanced images. The new method is effective and universal because its use is not limited to one type of tomography.
- Andy Adler and William R B Lionheart. 2006. Uses and abuses of EIDORS: an extensible software base for EIT. Physiological Measurement 27, 5: S25–S42.Google ScholarCross Ref
- Hela Garbaa, Lidia Jackowska-Strumillo, Krzysztof Grudzien, and Andrzej Romanowski. 2014. Neural network approach to ECT inverse problem solving for estimation of gravitational solids flow. 2014 Federated Conference on Computer Science and Information Systems, FedCSIS 2014.Google ScholarCross Ref
- Grzegorz Kłosowski, Tomasz Rymarczyk, Konrad Kania, Antoni Świć, and Tomasz Cieplak. 2020. Maintenance of industrial reactors supported by deep learning driven ultrasound tomography. Eksploatacja i Niezawodnosc 22, 1.Google Scholar
- Grzegorz Kłosowski, Tomasz Rymarczyk, Konrad Niderla, Magdalena Rzemieniak, Artur Dmowski, and Michał Maj. 2021. Comparison of machine learning methods for image reconstruction using the LSTM classifier in industrial electrical tomography. Energies 14, 21.Google ScholarCross Ref
- Michał Majchrowicz, Paweł Kapusta, Lidia Jackowska-Strumiłło, Robert Banasiak, and Dominik Sankowski. 2020. Multi-GPU, Multi-Node Algorithms for Acceleration of Image Reconstruction in 3D Electrical Capacitance Tomography in Heterogeneous Distributed System. Sensors 2020, Vol. 20, Page 391 20, 2: 391.Google Scholar
Recommendations
Use of the Two-Stage Neural System in Electrical Impedance Tomography for Imaging Moisture inside Walls
SenSys '22: Proceedings of the 20th ACM Conference on Embedded Networked Sensor SystemsDamp walls of buildings are a serious problem due to the social and economic consequences. Moisture causes accelerated wear of facades, paint coatings, weakening of the wall structure, and high maintenance and renovation costs. The growth of fungi and ...
A neural network approach for image reconstruction in electron magnetic resonance tomography
An object-oriented, artificial neural network (ANN) based, application system for reconstruction of two-dimensional spatial images in electron magnetic resonance (EMR) tomography is presented. The standard back propagation algorithm is utilized to train ...
Unsupervised Brain Segmentation System Using K-Means and Neural Network
Image Analysis and Processing. ICIAP 2022 WorkshopsAbstractVoxel-based morphometry is an analysis technique used to quantify matter volume in the human brain from magnetic resonance imaging studies. Atrophies and morphologic changes can be signs of neuronal depletion that can lead to various degenerative ...
Comments