Microwave tomography sensing for potential agarwood trees imaging

• Published in: Computers and electronics in agriculture Vol. 164; p. 104901
• Main Authors: Rahiman, M.H.F., Thomas, T.W.K., Soh, P.J., Rahim, R.A., Jamaludin, J., Ramli, M.F., Zakaria, Z.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01-09-2019; Elsevier BV
• Subjects: Agarwood; Algorithms; Antennas; Aquilaria; Computer simulation; Electric fields; Environmental impact; Finite element method; Harvesting; Image reconstruction; Inspection; Microwave tomography; Regularization; Timber (structural); Tomography; Transverse electric modes; Aquilaria; Microwave tomography; Agarwood; Image reconstruction
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2019.104901

•A simulation of microwave tomography using finite element modelling for agarwood investigation.•A 16-antenna of 1 GHz microwave tomography sensor was simulated to evaluate agarwood trees and obtain the tomogram of agarwood using four different image reconstruction algorithms.•Several phantoms were evaluated and the results were discussed qualitatively and quantitatively.•It is interesting to note that the results showed microwave tomography has the potential in agarwood imaging.
Agarwood or its scientific name, Aquilaria is an expensive and precious non-timber forest product. The problems faced by the agarwood industry are the indiscriminate harvesting of agarwood and ineffective identification and inspection process of Aquilaria tree by agarwood prospectors and researchers. Hence, microwave tomography (MWT) is proposed to evaluate agarwood in order to encounter these problems. This technique operates in a safe, non-invasive and insensitive to environmental impacts. A simulation-based study had been performed in which 16 antennas microwave tomography system was modelled using the finite element analysis. In this study, six profiles were being tested and simulated at 1 GHz. An electric field was simulated based on transverse electric mode to be emitted into the modelled wood with agarwood. Besides, electric field at the receiving antennas was also simulated when the simulated incident field was emitted into the modelled wood. The readings of the electric field for both incidents and receiving antennas were recorded to find the difference of electric field which will be used for image reconstruction. The images for the six profiles were reconstructed using Linear Back Projection (LBP), Filtered Back Projection (FBP), Newton’s One-Step Error Reconstruction (NOSER) and Tikhonov Regularization (TR) image reconstruction algorithms. The reconstructed images were then analysed using the Mean Structural Similarity (MSSIM) index. The phantom and reconstructed images were analysed qualitatively and quantitatively. The findings showed promising results as the proposed MWT was able to distinguish and identify agarwood in Aquilaria tree.