| Title | A Study on Adaptive Notch Filter Using Fourier Sine Series Expansion |
| Author | *Yasutomo Kinugasa (Matsue College of Technology, Japan), Masaki Kobayashi, Chi Wang (Chubu University, Japan), Yoshio Itoh (Tottori University, Japan) |
| Page | pp. 373 - 376 |
| Keyword | Narrowband Signal, Fourier Sine Series, Gradient Algorithm, All-pass Filter |
| Abstract | In this paper, new adaptive notch filter that realized the phase shift of the all-pass circuit by the Fourier sine series expansion and the adaptive algorithm was proposed. The proposed adaptive notch filter was always stable since the all-pass filter was composed by an exponential filter. The degree of the all-pass filter only depended on the required steepness of the elimination characteristic and did not depend on the number of narrowband signals that should be eliminated. |
| Title | The modified Newton-Raphson algorithm using Region of Interest in EIT |
| Author | *Jeong-Min Kang, Min Seong Kim, Jae-Hyoung Kim, Ji-Hoon Kim, Chan Yong Kim, Bong-Yeol Choi (Kyungpook National University, Republic of Korea), Sin Kim, Jeong Seong Lee, Kyung Youn Kim (Cheju National University, Republic of Korea) |
| Page | pp. 377 - 380 |
| Keyword | EIT, modified Newton-Rapshon, Region of Interest, FEM |
| Abstract | Electrical impedance tomography (EIT) reconstructs internal resistivity distribution using the various reconstruction methods. In reconstruction methods for EIT, the modified Newton-Raphson (mNR) method is known as nonlinear least-square method which has relatively good performance. But mNR does not obtain a satisfactory image due to ill-conditioning of Hessian matrix. To solve this problem, many methods have been developed and researched. This paper proposes a new mNR, which improves a reconstruction image in point of shape and resistivity, using assumed value and region of interest (ROI). |
| Title | A Probabilistic Short-length Linear Predictability Approach to Blind Source Separation |
| Author | *Mahdi Khosravy, Mohammad Reza Alsharif (University of the Ryukyus, Japan), Katsumi Yamashita (Osaka Prefecture University, Japan) |
| Page | pp. 381 - 384 |
| Keyword | blind signal separation, predictability, gradient ascent, eigen value |
| Abstract | A merit function based on short length linear predictability of signal in an objective probabilistic algorithm is defined and used for blind source separation (BSS) of linear mixtures of signals.
Its performance has been driven with many tests performed with mixtures of different kinds. Accurate numerical manners approve near-perfect performance of the method. Since short length of prediction window as well as extension to eigenvalue routine, signals can be extracted simultaneously and fast. |
| Title | Phase Correlated Bilateral Motion Estimation for Frame Rate Up-Conversion |
| Author | *DongGon Yoo, SukJu Kang, SungKyu Lee, Young Hwan Kim (Pohang University of Science and Technology, Republic of Korea) |
| Page | pp. 385 - 388 |
| Keyword | frame rate up-conversion, motion blur, liquid crystal display, phase correlation method, bilateral motion estimation |
| Abstract | In this paper, we propose a new motion estimation method. The proposed motion estimation method enhances the image quality by using the temporal symmetry in the phase correlation method. Experimental results show that the average PSNR of the proposed method is up to 1.36 dB higher than that of the benchmark methods. Also, the computation time of the proposed method is very low compared to the benchmark methods. |