• KIISE Transactions on Computing Practices
• /
• v.21 no.1
• /
• pp.64-69
• /
• 2015

Place awareness is an essential for location-based services that are widely provided to smartphone users. However, traditional GPS-based methods are only valid outdoors where the GPS signal is strong and also require symbolic place information of the physical location. In this paper, environmental sounds and images are used to recognize important aspects of each place. The proposed method extracts feature vectors from visual, auditory and location data recorded by a smartphone with built-in camera, microphone and GPS sensors modules. The heterogeneous feature vectors were then learned by an ensemble learning method that learns each group of feature vectors for each classifier respectively and votes to produce the highest weighted result. The proposed method is evaluated for place recognition using a data group of 3000 samples in six places and the experimental results show a remarkably improved recognition accuracy when using all kinds of sensory data comparing to results using data from a single sensor or audio-visual integrated data only.

• Journal of Broadcast Engineering
• /
• v.17 no.5
• /
• pp.827-837
• /
• 2012

As broadcasting environments change rapidly to digital, user requirements for next-generation broadcasting service which surpass current HDTV service become bigger and bigger. The next-generation broadcasting service progress from 2D to 3D, from HD to UHD and from 5.1ch audio to more than 10ch audio for high quality realistic broadcasting service. In this paper, we propose 10.2ch based multichannel audio reproduction system for UHDTV. The 10.2ch-based audio reproduction system add two side loudspeakers to enhance the surround sound localization effect and add two height and one ceiling loudspeakers to enhance the elevation localization effect. To evaluate the proposed system, we used APM(Auditory Process Model) for objective localization test and conducted subjective localization test. As a result of objective/subjective localization test, the proposed system shows the statistically same performance compare with 22.2ch audio system and shows the significantly better performance compared with 5.1ch audio system.

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.42 no.1
• /
• pp.13-20
• /
• 2005

In this paper, I recognized Korean isolated words using CVEM which is suggested by M. Bilginer et al. CVEM is an algorithm which is easy to extract the common properties from training voice signals and also doesn't need complex calculation. In addition CVEM shows high accuracy in recognition results. But, CVEM has couple of problems which are impossible to use for many training voices and no discriminant information among extracted common vectors. To get the optimal common vectors from certain voice classes, various voices should be used for training. But CVEM is impossible to get continuous high accuracy in recognition because CVEM has a limitation to use many training voices and the absence of discriminant information among common vectors can be the source of critical errors. To solve above problems and improve recognition rate, k-clustering subspace method and DCVEM suggested. And did various experiments using voice signal database made by ETRI to prove the validity of suggested methods. The result of experiments shows improvements in performance. And with proposed methods, all the CVEM problems can be solved with out calculation problem.

• Journal of Biomedical Engineering Research
• /
• v.25 no.6
• /
• pp.575-580
• /
• 2004

In this work, a method is presented to find an anatomical site of a current source crudely in a standard brain using grand average of MEG data. Minimum norm estimation algorithm and truncated singular value decomposition were applied to calculate the distributed sources that can reproduce the measured signals. Grand average over all subjects was obtained from the transformed signals, which would be detected in a standard sensor plane by the obtained distributed current sources. In the simulation study, it was shown that the localized dipole using the grand average is consistent with the mean location of localized dipoles of all subjects within several mm even with large inter-individual differences of sensor positions. This result suggests that the mean location of low level signal source can be estimated as a dipole source in grand average and it was confirmed in the localization of the current source of N100m. when the localized dipole is registered on a standard brain. This result also suggests that the activity region obtained from grand average can be crudely estimated on a standard brain using the source location of the N100m as a reference point.

• Journal of Broadcast Engineering
• /
• v.23 no.6
• /
• pp.855-865
• /
• 2018

Sound event detection is one of the research areas to model human auditory cognitive characteristics by recognizing events in an environment with multiple acoustic events and determining the onset and offset time for each event. DCASE, a research group on acoustic scene classification and sound event detection, is proceeding challenges to encourage participation of researchers and to activate sound event detection research. However, the size of the dataset provided by the DCASE Challenge is relatively small compared to ImageNet, which is a representative dataset for visual object recognition, and there are not many open sources for the acoustic dataset. In this study, the sound events that can occur in indoor and outdoor are collected on a larger scale and annotated for dataset construction. Furthermore, to improve the performance of the sound event detection task, we developed a dual CNN structured sound event detection system by adding a supplementary neural network to a convolutional neural network to determine the presence of sound events. Finally, we conducted a comparative experiment with both baseline systems of the DCASE 2016 and 2017.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.12 no.1
• /
• pp.25-50
• /
• 2001

Using Fourier transformation and wavelet analysis, we compared the auditory event-related potentials of the patients with attention deficit-hyperactivity disorders(abbr. ADHD, 13 boys) and normal control children(8 boys). Amplitudes of the event-related potentials which were calculated via Fourier transformation were compared between the groups and between conditions(non-target versus target) in each group. To the non-target stimuli, the patients with ADHD showed significantly greater amplitudes across almost all of the electrode sites and frequencies. To the target stimuli, the incidents which ADHD patients showed much higher amplitudes than normal controls significantly decreased, while those of the reverse results increased significantly. These results were consistent with the comparison results about negative difference wave(abbr. Nd wave) using Fourier transformation. In summary, it was proved that non-target stimulus which should be ignored elicited more robust electrical response from the patients with ADHD than normal children, but the target stimulus which reguired active processing did much less electrical activity in the patients. For the patients, they showed much inhibited electrical response to the target stimuli in some electrodes and frequency ranges. Normal children were more strongly stimulated by the target stimuli in almost all electrodes and frequency ranges than the patients, but less in prefrontal leads and frontal leads. Wavelet analysis results proved that early responses(0-300msec) to the nontarget stimuli of the patients were significantly greater than the normal controls in prefrontal, anterior frontal, some parts of temporal, and occipital lobes and that late response(300-370msec) were significantly lesser than normal children in parietal and central electrodes. Target stimuli elicited significantly higher electrical activity in both group than non-target stimuli did. Prefrontal and frontal lobes showed stronger responses in the patients than normal children irrespective of stimulus condition, but parietal and temporal lobes did higher activities in normal children than the patients only to the target stimuli. In conclusion, the patients with ADHD showed much greater responses to the stimuli which should be ignored, but failed to activated the necessary processes to the target stimuli. Also, we found that the frequency-dimension analysis and wavelet analysis were useful for the signal processing such as event related potentials.