Search

In this study, we proposed a sound localization method by using the binaural model for a real environment, in which the reflected sound and background noise is present. In order to estimate of horizontal angle, the reduction method of the initial reflected sound that combining the modified Thomposon-τ method and the modified sub-band peak hold processing, was proposed and its effectiveness has been performed. In addition, to create the ESR database of vertical angle, the modified Thompson-τ method is used for elimination of the abnormal value. In the estimation of the vertical angle, a new method that is called wins calculation method is proposed and it is strong in abrupt noise, without using the traditional cross-correlation method. The experiments were carried by using /a/ sound in real environment with 45[dB] background noise and 300[ms] reverberation time. As the results obtained that the estimate correct rate of the horizontal angle is 94.6%, the correct rate of the vertical angle is next 98.6%, and the validity of the proposed method was confirmed.

In this paper, we propose the novel directional Wavelet Transform based on the Complex Discrete Wavelet Transform(CDWT). The CDWT is one of the Wavelet Transform and widely used as space-frequency analysis method for the signal and the image. The CDWT decomposes the image into directional components from the frequency analysis. And then, we can get the border, discontinuities and edges in the image. This property is called Directional Selection and this property is expected to give us shape information and geometric features. But, previous Directional Selection only offers the six directional components and features. Thus we firstly design the directional filter that can extract many directional feature rather than previous one. Secondly we propose the novel directional Wavelet Transform combining with previous CDWT and designed directional filters. Finally, we apply the proposed method into medical image processing.

In this study, in order to verify the effectiveness of the variable-band-filters discrete wavelet transform (VBF-DWT) and construction method of the variable band filter (VBF), a fetal ECG extraction has been carried out and the main results obtained are shown as follows. The approach to configuration variable band filter (VBF) by selecting the frequency band only fetal ECG component presented is effective to configure the optimal base sensible signal. The extraction of the fetal electrocardiogram is successful by applying soft-thresholding to VBF-DWT, which uses the constructed VBF. The information extropy is selected as an evaluation index, and two kinds of ECG signal are used to evaluate the relative performance of the wavelet packet transform (WPT) and the proposed VBF-DWT. The first kind of ECG signal is synthesized using a maternal ECG, fetal ECG, and white noise; the second kind is real measured mixed noisy maternal-fetal ECG obtained using ICA. The results show that the VBF-DWT basis (designed using VBF) outperforms the WPT basis that was elected by the basis algorithm (BBA).