Stable Learning for Neural Network Tomography by Using Back Projected Image
This paper presents a stable learning method of the neural network tomography, in case of asymmetrical few view projection. The neural network collocation method (NNCM) is one of effective reconstruction tools for symmetrical few view tomography. But in case of asymmetrical few view, the NNCM tends to unstable and fails to reconstruct appropriate tomographic images. We solve the unstable problem by introducing the back projected image in the early learning stage of NNCM. The numerical simulation with an assumed tomographic image show the effectiveness of the proposed method.
KeywordsFew view tomography neural network back propagation collocation method inverse problem ill-posed problem model fitting
Unable to display preview. Download preview PDF.
- 3.Takeda, T., Ma, X.F.: Ionospheric Tomography by Neural Network Collocation Method. Plasma and Fusion Research 2, S015-1–6 (2007)Google Scholar
- 5.Rumelhart, D.E., McClelland, J.L.: PDP Research Group: Parallel Distributed Processing. MIT Press, Cambridge (1986)Google Scholar
- 6.Liu, Y., Peterson, B.J., Iwama, N., Konoshima, S.: LHD Experiment Group, JT-60 team: Application of Tomographic Imaging to Multi-pixel Bolometric Measurements. Plasma and Fusion Research 2, S1124-1–4 (2007)Google Scholar