Narong Borijindargoon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Narong Borijindargoon has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
Boon Poh Ng
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore
4 shared publications- Boon Poh Ng
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore
2 shared publications - Bihan Wen
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, Singapore
1 shared publication - Dhimas Arief Dharmawan
Research Center for Artificial Intelligence and Cyber Security, National Research and Innovation Agency, Bandung, Indonesia
1 shared publication Kaiyi Yang
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore
1 shared publication- Saiprasad Ravishankar
Department of Computational Mathematics, Science and Engineering, Michigan State University, East Lansing, MI, USA
1 shared publication
Latest Publications
- 6 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1405-1409
Learning Sparsifying Transforms for Image Reconstruction in Electrical Impedance Tomography
Electrical Impedance Tomography (EIT) is a fast and non-invasive imaging technology that reconstructs the internal electrical properties of a subject. However, its functionality is limited by low spatial resolution arising from an ill-posed and ill-conditioned inverse problem. Several sparsity-promoting regularization methods have been applied to improve the quality of EIT image reconstruction, including various ℓ 0 and ℓ 1 -based analytical models (TV, TwIST, etc.), and a patch-based sparse representation via a learned dictionary (using the K-SVD algorithm), dubbed CS-EIT. To further exploit the potential of compressed sensing in Electrical Impedance Tomography, this paper incorporates the recent novel method of transform learning for EIT image reconstruction. We propose a blind compressed sensing algorithm, dubbed TL-EIT, which simultaneously optimizes the sparsifying transform and updates the reconstructed image. We demonstrate using both synthetic and in vivo data that the proposed TL-EIT is more effective than other sparsity-based algorithms for reconstructing high-quality EIT images. In addition, TL-EIT also accelerates the reconstruction process in comparison to other learning-based algorithms like CS-EIT.
- 6 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1405-1409
Learning Sparsifying Transforms for Image Reconstruction in Electrical Impedance Tomography
Electrical Impedance Tomography (EIT) is a fast and non-invasive imaging technology that reconstructs the internal electrical properties of a subject. However, its functionality is limited by low spatial resolution arising from an ill-posed and ill-conditioned inverse problem. Several sparsity-promoting regularization methods have been applied to improve the quality of EIT image reconstruction, including various ℓ 0 and ℓ 1 -based analytical models (TV, TwIST, etc.), and a patch-based sparse representation via a learned dictionary (using the K-SVD algorithm), dubbed CS-EIT. To further exploit the potential of compressed sensing in Electrical Impedance Tomography, this paper incorporates the recent novel method of transform learning for EIT image reconstruction. We propose a blind compressed sensing algorithm, dubbed TL-EIT, which simultaneously optimizes the sparsifying transform and updates the reconstructed image. We demonstrate using both synthetic and in vivo data that the proposed TL-EIT is more effective than other sparsity-based algorithms for reconstructing high-quality EIT images. In addition, TL-EIT also accelerates the reconstruction process in comparison to other learning-based algorithms like CS-EIT.
- Open Access
- 1 October 2019
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 13 (8) , 1092-1101
Hybrid sensitivity‐correlation regularisation matrix for electrical impedance tomography
In electrical impedance tomography, the primary task of image reconstruction process is to solve a discrete ill‐posed inverse problem. The estimated solution is commonly obtained under a regularisation framework so that the noise amplified solution, which occurs during the matrix inversion process, can be avoided. The regularisation framework aims at balancing the model‐data fitness while simultaneously constraining the solution space with additional prior (commonly prescribed through the regularisation matrix and solution‐norm). In this study, a relationship between two robust regularisation matrices namely Newton one‐step error reconstruction and fidelity‐embedded regularisation is explicitly highlighted in both spatial and singular value decomposition domains. A hybrid regularisation matrix which encompasses the two prior knowledge, non‐uniform sensitivity distribution and array response correlation, is then proposed as an alternative prior. Experimental results along with several evaluated performance parameters highlight the ability of the proposed prior to achieve a well‐balanced and robust performance.
- 1 October 2019
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 13 (8) , 1092-1101
Open AccessHybrid sensitivity‐correlation regularisation matrix for electrical impedance tomography
In electrical impedance tomography, the primary task of image reconstruction process is to solve a discrete ill‐posed inverse problem. The estimated solution is commonly obtained under a regularisation framework so that the noise amplified solution, which occurs during the matrix inversion process, can be avoided. The regularisation framework aims at balancing the model‐data fitness while simultaneously constraining the solution space with additional prior (commonly prescribed through the regularisation matrix and solution‐norm). In this study, a relationship between two robust regularisation matrices namely Newton one‐step error reconstruction and fidelity‐embedded regularisation is explicitly highlighted in both spatial and singular value decomposition domains. A hybrid regularisation matrix which encompasses the two prior knowledge, non‐uniform sensitivity distribution and array response correlation, is then proposed as an alternative prior. Experimental results along with several evaluated performance parameters highlight the ability of the proposed prior to achieve a well‐balanced and robust performance.
- 30 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 26 (10) , 1511-1515
Design of Optimal Adaptive Filters for Two-Dimensional Filamentary Structures Segmentation
The problems of filamentary structures segmentation encompass retinal vessel detection in fundus images, reconstruction and tracing of neurons in microscopic images and segmentation of human vasculatures in two-dimensional digital angiography. In this letter, we focus on retinal vessel segmentation which is an important and challenging problem among others. Many retinal vessel segmentation algorithms have been developed, where most of them rely on the filtering technique. However, the available filters suffer from undesirable responses at some vessel and non-vessel structures. Thus, we propose a new framework based on optimal adaptive filters for retinal vessel segmentation. The filter coefficients are obtained by solving multiple inverse problems under a regularization framework. The performance of the proposed framework is evaluated on fundus images from the DRIVE and STARE datasets. In the experimental section, we show that the proposed framework outperforms all of the compared methods in four indicators namely sensitivity, F1-score, G-mean and Mathews correlation coefficient.
- 30 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 26 (10) , 1511-1515
Design of Optimal Adaptive Filters for Two-Dimensional Filamentary Structures Segmentation
The problems of filamentary structures segmentation encompass retinal vessel detection in fundus images, reconstruction and tracing of neurons in microscopic images and segmentation of human vasculatures in two-dimensional digital angiography. In this letter, we focus on retinal vessel segmentation which is an important and challenging problem among others. Many retinal vessel segmentation algorithms have been developed, where most of them rely on the filtering technique. However, the available filters suffer from undesirable responses at some vessel and non-vessel structures. Thus, we propose a new framework based on optimal adaptive filters for retinal vessel segmentation. The filter coefficients are obtained by solving multiple inverse problems under a regularization framework. The performance of the proposed framework is evaluated on fundus images from the DRIVE and STARE datasets. In the experimental section, we show that the proposed framework outperforms all of the compared methods in four indicators namely sensitivity, F1-score, G-mean and Mathews correlation coefficient.
- Open Access
- 1 January 2019
- Published by The Electromagnetics Academy in Progress In Electromagnetics Research Letters
- Vol. 87, 29-37
DIRECTIONAL ADAPTIVE MUSIC-LIKE ALGORITHM UNDER SYMMETRIC α-STABLE DISTRIBUTED NOISE
An algorithm named MUSIC-like algorithm was previously proposed as an alternative method to the MUltiple SIgnal Classification (MUSIC) algorithm for direction-of-arrival (DOA) estimation. Without requiring explicit model order estimation, it was shown to have robust performance particularly in low signal-to-noise ratio (SNR) scenarios. In this letter, the working principle of a relaxation parameter β, a parameter which was introduced into the formulation of the MUSIC-like algorithm, is provided based on geometrical interpretation. To illustrate its robustness, the algorithm will be examined under symmetric α-stable distributed noise environment. An adaptive framework is then developed and proposed in this letter to further optimize the algorithm. The proposed adaptive framework is compared with the original MUSIC-like, MUSIC, FLOM-MUSIC, and SSCM-MUSIC algorithms. A notable improvement in terms of targets resolvability of the proposed method is observed under different impulse noise scenarios as well as different SNR levels.
- 1 January 2019
- Published by The Electromagnetics Academy in Progress In Electromagnetics Research Letters
- Vol. 87, 29-37
Open AccessDIRECTIONAL ADAPTIVE MUSIC-LIKE ALGORITHM UNDER SYMMETRIC α-STABLE DISTRIBUTED NOISE
An algorithm named MUSIC-like algorithm was previously proposed as an alternative method to the MUltiple SIgnal Classification (MUSIC) algorithm for direction-of-arrival (DOA) estimation. Without requiring explicit model order estimation, it was shown to have robust performance particularly in low signal-to-noise ratio (SNR) scenarios. In this letter, the working principle of a relaxation parameter β, a parameter which was introduced into the formulation of the MUSIC-like algorithm, is provided based on geometrical interpretation. To illustrate its robustness, the algorithm will be examined under symmetric α-stable distributed noise environment. An adaptive framework is then developed and proposed in this letter to further optimize the algorithm. The proposed adaptive framework is compared with the original MUSIC-like, MUSIC, FLOM-MUSIC, and SSCM-MUSIC algorithms. A notable improvement in terms of targets resolvability of the proposed method is observed under different impulse noise scenarios as well as different SNR levels.
- Open Access
Directional Adaptive MUSIC-like Algorithm under α-Stable Distributed Noise
An algorithm called MUSIC-like algorithm was originally proposed as an alternative method to the MUltiple SIgnal Classification (MUSIC) algorithm for direction-of-arrival (DOA) estimation. Without requiring explicit model order estimation, it was shown to have robust performance particularly in low signal-to-noise ratio (SNR) scenarios. In this letter, the working principle of a relaxation parameter {\beta}, a parameter which was introduced into the formulation of the MUSIC-like algorithm, is provided based on geometrical interpretation. To illustrate its robustness, the algorithm will be examined under symmetric {\alpha}-stable distributed noise environment. An adaptive framework is then developed and proposed in this letter to further optimize the algorithm. The proposed adaptive framework is compared with the original MUSIC-like, MUSIC, FLOM-MUSIC, and SSCM-MUSIC algorithms. A notable improvement in terms of targets resolvability of the proposed method is observed under different impulse noise scenarios as well as different SNR levels.
Open AccessDirectional Adaptive MUSIC-like Algorithm under α-Stable Distributed Noise
An algorithm called MUSIC-like algorithm was originally proposed as an alternative method to the MUltiple SIgnal Classification (MUSIC) algorithm for direction-of-arrival (DOA) estimation. Without requiring explicit model order estimation, it was shown to have robust performance particularly in low signal-to-noise ratio (SNR) scenarios. In this letter, the working principle of a relaxation parameter {\beta}, a parameter which was introduced into the formulation of the MUSIC-like algorithm, is provided based on geometrical interpretation. To illustrate its robustness, the algorithm will be examined under symmetric {\alpha}-stable distributed noise environment. An adaptive framework is then developed and proposed in this letter to further optimize the algorithm. The proposed adaptive framework is compared with the original MUSIC-like, MUSIC, FLOM-MUSIC, and SSCM-MUSIC algorithms. A notable improvement in terms of targets resolvability of the proposed method is observed under different impulse noise scenarios as well as different SNR levels.
- 20 August 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 66 (6) , 4661-4671
MUSIC-Like Algorithm for Source Localization in Electrical Impedance Tomography
In electrical impedance tomography (EIT), the noise amplified solution caused during matrix inversion can be avoided with non-parametric spectral based estimation when the conductivity variation is bounded and spatially sparse. Among many spectral based algorithms used in direction-of-arrival (DOA) estimation, an algorithm called multiple signal classification (MUSIC) is one of the most well-known that has super resolution performance. However, its dependency on the model order estimation can lead to performance degradation especially for quasi-static environment like EIT application and this is due to source location changes and conductivity variation. In this paper, the relationship between source position, conductivity variation, ill-conditioned array manifold and eigenvalues of the covariance matrix are explored. An algorithm called MUSIClike which has high resolution performance comparable to MUSIC is then proposed for EIT application. It is formulated under the beamforming framework, and therefore does not require an estimation of model order from the covariance matrix. Simulation results show that the proposed method is capable of obtaining high resolution performance under various noise levels. An 8-electrode EIT system prototype was built using the proposed method, and experimental results confirm the high resolution performance capability of the proposed method.
- 20 August 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 66 (6) , 4661-4671
MUSIC-Like Algorithm for Source Localization in Electrical Impedance Tomography
In electrical impedance tomography (EIT), the noise amplified solution caused during matrix inversion can be avoided with non-parametric spectral based estimation when the conductivity variation is bounded and spatially sparse. Among many spectral based algorithms used in direction-of-arrival (DOA) estimation, an algorithm called multiple signal classification (MUSIC) is one of the most well-known that has super resolution performance. However, its dependency on the model order estimation can lead to performance degradation especially for quasi-static environment like EIT application and this is due to source location changes and conductivity variation. In this paper, the relationship between source position, conductivity variation, ill-conditioned array manifold and eigenvalues of the covariance matrix are explored. An algorithm called MUSIClike which has high resolution performance comparable to MUSIC is then proposed for EIT application. It is formulated under the beamforming framework, and therefore does not require an estimation of model order from the covariance matrix. Simulation results show that the proposed method is capable of obtaining high resolution performance under various noise levels. An 8-electrode EIT system prototype was built using the proposed method, and experimental results confirm the high resolution performance capability of the proposed method.