• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.449-450
• /
• 2022

반려견의 행동인식기술은 다양한 센서들에서 입력되는 반려견의 동작과 관련된 정보를 분석하고 해석하여 반려견이 어떤 행동을 취하고 있는지를 인식하는 기술이다. 음성인식기술은 컴퓨터가 청각 자료를 수집, 분석하여 훈련된 데이터와 비교를 통해 소리를 분류하는 기술이다. 본 논문에서는 딥러닝을 기반으로 행동인식기술과 음성인식기술을 적용하여 반려견의 감정을 판단하는 기법을 제안한다. 이러한 기법은 반려견의 감정을 쉽게 파악하여 반려견 보호자가 반려견의 행동과 감정에 대한 이해를 쉽고 빠르게 할 수 있으므로, 보호자에게 즐거운 반려 생활이 가능하도록 도움을 줄 수 있다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.1
• /
• pp.71-78
• /
• 2009

Sound source localization systems in service robot applications estimate the direction of a human voice. Time delay information obtained from a few separate microphones is widely used for the estimation of the sound direction. Correlation is computed in order to calculate the time delay between two signals. In addition, PHAT weighting function can be applied to significantly improve the accuracy of the estimation. However, FFT and IFFT operations in the PHAT weighting function occupy more than half of the area of the sound source localization system. Thus efficient FFT and IFFT designs are essential for the IP implementation of sound source localization system. In this paper, we propose an efficient FFT/IFFT design method based on the characteristics of human voice.

• Korean Journal of Child Studies
• /
• v.24 no.5
• /
• pp.91-104
• /
• 2003

This study analized the strategies in beginning readers by age and reading ability. Sixty 4-and 5-year old subjects took a reading test based on Bsatjes & Brown(1997) and Park, et a1.(1989). They read contextual and non-contextual storybooks. Errors in oral reading were recorded as mispronunciations, substitutions, omissions, insertions, teacher-assistance and self-corrections. Mispronunciations and substitutions were Specifically evaluated for graphic and contextual reading strategies. Data were analyzed by percentage and mean. Results revealed that children made more mispronunciation errors in reading the non-contextual story book. They used graphic information more than contextual information. Fine-year olds and high-level readers developed the use of graphic and contextual cues simultaneously.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.15 no.5 s.98
• /
• pp.558-563
• /
• 2005

This research aims to derive the general emotion and preference of the 25 sounds by using the questionnaire method to provide the contents for the soundscape design fit to the characteristics of the given situation. The results can be summarized as follows : Nature sounds except dog barking and locust crying showed highest preference, and the traffic sounds, baby crying and footstep of upper floor showed the lowest. Hereby, the sound recognition can be different each other according to quality of tone and situation even they are the same sounds. The most impressive, wanted to be conserved and retrospective sound was the nature sound, and the most wanted to be removed sound was the machinery sound. And the social sound was shown as the most korean and fit to the art theme street.

• Annual Conference of KIPS
• /
• 2017.11a
• /
• pp.1076-1078
• /
• 2017

본 연구는 이상음원(비명, 유리창 파손음, 경적소리 등) 발생 시, 2개의 마이크로폰에 입력되는 사운드에 대하여 음원 방향추적 장치와 연결된 카메라에 음원의 방향 정보를 전송함으로써, 카메라의 View Point를 음원 발생방향으로 이동시켜 사고현장을 더욱 신속하게 대처할 수 있는 시스템에 대한 연구이다. 일반적인 음성을 이용한 감시카메라는 단순히 소리 발생 여부만 감지하지만, 본 시스템은 이상음원 발생 지점으로 카메라의 방향 제어를 가능하게 한다. 이상음원의 검출은 기존에 수집한 DB를 기반으로 비교, 분석 과정을 통하여 이상음원을 분류한다. 음원 발생 방향은 음원 발생 시, 마이크로폰에 도달하는 음원의 시간차에 따른 음파의 위상차를 계산하여 음원 발생 방향을 판단하게 된다.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.11a
• /
• pp.806-809
• /
• 2004

This research aims to derive the general emotion and preference of the 25 sounds by using the questionnaire method to provide the contents for the soundscape design fit to the characteristics of the given situation. The results can be summarized as follows; Nature sounds except dog barking and locust crying showed highest preference, and the traffic sounds, baby crying and footstep of upper floor showed the lowest. Hereby, the sound recognition can be different each other according to quality of tone and situation even they are the same sounds. The most impressive, wanted to be conserved and retrospective sound was the nature sound, and the most wanted to be removed sound was the machinery sound. And the social sound was shown as the most korean and fit to the art theme street.

• Annual Conference of KIPS
• /
• 2011.11a
• /
• pp.231-234
• /
• 2011

본 논문에서는 긴급통화만 가능하던 기존의 위기대처시스템들과 차별화를 두기 위해, 안드로이드에서 제공하는 시스템 중 하나인 미디어 부분을 이용하여 사용자의 음성을 입력받아 소리세기를 출력시켰다. 또한 GoogleAPI를 활용하여 현재위치를 찾아 긴급 메시지로 전송 가능한 시스템을 제공한다. 본 시스템은 사용자의 편리성과 효율성을 높이기 위하여, 단 1회 실행만으로도 연속적인 위기대처시스템의 인식과 주기적인 실시간 위치를 찾아 메시지 전송이 가능하다. 부가적으로 현재위치를 위도 경도로만 출력 되어 전송 되어지는 것이 아니라, 위도 경도의 값을 주소로 변환하여 출력 하므로서 보다 정확하고 편리하게 서비스를 이용할 수 있다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.257-258
• /
• 2024

현재 시각장애인들이 현금을 사용하게 될 시 지폐가 얼마인지 확인할 방법이 없어 불편을 겪거나 금전적 사기를 당할 위험이 잦다. 한국은행에서는 이러한 사고를 막기 위해 점자 지폐를 만들어 발부하고 있지만 시각장애인 91%가 식별하지 못해 많은 불편을 겪고 있다. 본 논문에서는 딥러닝을 활용하여 화폐를 인식하고 TTS 기술을 사용하여 지폐의 값이 얼마인지 소리로 알려주는 시스템을 개발하였다. 지폐 인식을 위해 데이터를 직접 수집하여 YOLOv5 알고리즘을 활용하여 학습시킨 Weights 파일을 사용하였다. 이를 활용하여 시각장애인들은 더 안전하게 현금을 사용하고, 금전적인 문제를 예방할 수 있다.

• Archives of design research
• /
• v.16 no.3
• /
• pp.101-110
• /
• 2003

People are surrounded with numerous sounds. The sound is generated from nature and people. For example, the sound enables people or animal responsive to instinctive action. Music or fine arts are presented differently by its distinctive medium. On the other hand, those art forms are similar in a way that people uses right side of brain and intuitions for creative effects. Conjunctions between sounds and visual arts have been progressed to data. From art forms in subjective views to art forms using high technology such as the computer, experiments for sounds visualization are practiced constantly. For that reason, intrinsic attributes of sounds in design area and distinctive qualities are discussed in this study. With respect to existing category of studies and consideration of the tendency in recent researches, the object is to propose direction for the study in regards to methodology of design, which is reconstructing visualized expression.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.406-413
• /
• 2020

In this paper, to improve the classification accuracy of bird and amphibian acoustic sound, we utilize GLU (Gated Linear Unit) and Self-attention that encourages the network to extract important features from data and discriminate relevant important frames from all the input sequences for further performance improvement. To utilize acoustic data, we convert 1-D acoustic data to a log-Mel spectrogram. Subsequently, undesirable component such as background noise in the log-Mel spectrogram is reduced by GLU. Then, we employ the proposed temporal self-attention to improve classification accuracy. The data consist of 6-species of birds, 8-species of amphibians including endangered species in the natural environment. As a result, our proposed method is shown to achieve an accuracy of 91 % with bird data and 93 % with amphibian data. Overall, an improvement of about 6 % ~ 7 % accuracy in performance is achieved compared to the existing algorithms.