http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
조아현,한찬훈,Jo, A-Hyeon,Haan, Chan-Hoon 한국음향학회 2022 韓國音響學會誌 Vol.41 No.3
The present study aims to investigate the difference of soundscape evaluation results from on-site field test and laboratory test which are commonly used for soundscape surveys. In order to do this, both field and lab tests were carried out at four different areas in Cheongju city. On-site questionnaire surveys were undertaken to 65 people at 13 points. Laboratory listening tests were carried out to 48 adults using recorded sounds and video. Laboratory tests were undertaken to two different groups who had experience of field survey or not. Also, two different sound reproduction tools, headphones and speakers, were used in laboratory tests. As a result, it was found that there is a very close correlation between sound loudness and annoyance in both field and laboratory tests. However, it was concluded that there must be a difference in recognizing the figure sounds between field and laboratory tests since it is hard to apprehend on-site situation only using visual and aural information provided in laboratory tests. In laboratory tests, it was shown that there is a some difference in perceived most loud figure sounds in two groups using headphones and speakers. Also, it was analyzed that there is a tendency that field experienced people recognize the figure sounds using their experienced memory while non-experienced people can not perceive the figure sounds.
인터뷰 형식의 오디오 데이터를 이용한 전이 학습모델 기반 우울증 진단
조아현(A-Hyeon Jo),곽근창(Keun-Chang Kwak) 대한전기학회 2021 전기학회논문지 P Vol.70 No.4
Depression can lead to serious mental and physical illness, so early detection is important. Currently, a system to help early detection of depression using AI technology is being developed in various ways. In particular, research on diagnosing depression through voices that can be easily encountered in daily life is being actively conducted. In this paper, we compare and analyze the depression diagnosis performance of transfer learning models using interview-type audio data. Data use the DAIC-WOZ Depression Database, which contains audio files in interview-type. As the transfer learning model, it uses VGGish and YAMNet built based on Convolutional Neural Network(CNN) among deep learning models that are widely being used for audio classification. The characteristics of speech data are extracted to black-and-white and color two-dimensional images using the Bark spectrogram, Mel spectrogram, and Log Mel-spectrogram methods. The performance of the depression diagnosis model is higher in YAMNet than in VGGish. In case that black-and-white images are input, YAMNet’s performance was the highest with 94.48% when mel spectrogram features were used.On the other hand, in case that color images are input, YAMNet’s performance was the highest at 97.34% when bark spectrogram features were used proving that it is most suitable for diagnosing depression.
음성감정 신호로부터 전이학습기반 딥러닝을 이용한 감정인식
조아현(A-Hyeon Jo),곽근창(Keun-Chang Kwak) 대한전기학회 2021 대한전기학회 학술대회 논문집 Vol.2021 No.11
최근 인공지능 기술의 발전으로 개인화 서비스가 증가함으로써 감정인식 기술이 중요하게 다뤄지고 있다. 감정인식 기술의 분야중 음성 감정인식은 사람의 목소리를 분석하여 감정 상태를 파악하는 기술이다. 본 논문에서는 한국어 음성데이터를 이용하여 특징 추출 방법에 따른 CNN 기반 전이 학습모델 중 하나인 ResNet18의 감정인식 성능을 비교한다. 데이터는 직접 구축한 한국어 음성 감정 데이터를 사용했고 감정 상태는 행복, 무감정, 화남, 슬픔 총 4가지로 분류된다. 특징추출 방법은 ERB 스펙트로그램과 로그 멜-스펙트로그램을 사용했고, 데이터를 4가지 경우로 나누어 성능을 비교한다. 이 중 모든 데이터를 사용했을 때, ERB 스펙트로그램을 사용한 경우 성능이 89.91%이고 로그 멜 스펙트로그램을 사용한 경우는 96.05%로 후자가 음성 감정인식에 더 적합하다는 것을 증명한다.
조아현(A-Hyeon Jo),곽근창(Keun-Chang Kwak) 한국정보기술학회 2022 한국정보기술학회논문지 Vol.20 No.1
Depression is a serious mental illness that affects an individuals overall life and causes changes in emotions, thoughts, physical condition, and behavior. Therefore, it is important to accurately diagnose depression in the early stages, and to this end, several systems are being developed using deep learning. In this paper, we propose a deep learning model for bi-directional LSTM-based diagnosis of depression using speech signals and compare and analyze performance. For training is used bi-directional LSTM model, one of the deep learning models. Features of speech signals were extracted using Mel-Frequency Cepstral Coefficient (MFCC) and Gammatone Cepstral Coefficient (GTCC), and data augmentation was performed to resolve imbalance of class. The performance of the Bi-LSTM model before data augmentation was about 56.58% and after data augmentation was about 98.49%, which improved the classification performance of the model by solving the imbalance of class through learning data augmentation.
문화재 보존 및 복원에 사용되는 천연 광물성안료의 물성평가 - 석청(石靑)중심으로
박주현,정혜영,고인희,정서린,조아현,Park, Ju-Hyun,Jeong, Hye-Yeong,Go, In-Hee,Jeong, Sir-Lin,Jo, A-Hyeon 국립문화재연구소 2015 保存科學硏究 Vol.36 No.-
문화재 수리, 복원재료 및 회화용으로 사용되는 천연광물성 무기안료인 석청의 물리적 특성을 연구하였다. 본 연구를 위해 시중에서 유통되고 있는 청색계열의 안료(KA-한국, JA, JB, JC-일본, CA, CB, CC-중국)를 선정하였다. 연구에 앞서 XRD, XRF분석을 통하여 안료의 결정구조 및 조성을 확인하였으며, 본 연구에 사용된 안료는 Azurite가 주원료임이 확인되었다. 조성물을 확인한 안료는 입도, 색도, 비중, 흡유량 등 물리적 특성을 평가하였으며, 입도에 따른 물리적 특성의 변화를 확인하였다. 안료는 입도가 작아짐에 따라 명도는 커지고, 청색도는 커지다가 다시 작아지는 특성이 있으며, KA가 안료번호에 따라 가장 잘 분급되어 있음을 알 수 있었다. 비중은 입자크기가 커질수록 커지고, 흡유량은 입자크기가 작아질수록 커졌다. 이러한 결과를 토대로 안료의 기초적인 물리적 특성을 규명하고 나아가 문화재 수리, 복원에 선택적으로 사용하는데 도움이 될 것으로 기대한다. The purpose of this study is to analyse the properties of natural mineral pigments used in restoration and conservation of cultural assets. For this study blue-based pigments that are sold in market were selected. The component analysis using by XRF and XRD shows that blue pigment consist of Azurite. And each specimens were evaluated particle size, chromaticity, specific gravity and oil-absorption according to rating system of pigments particle size. Results show that the value of L* increase with the decrease in particle size. The results suggest that the physical properties which is specific gravity, oil-absorption and chromaticity depend on the particle sizes of pigments. When it comes to particle size of pigments decreased by increasing the number of rating system. In addition, the chromaticity related to particle size. The result from this study expects to be used as useful referencing data for conservation and restoration of cultural heritage and understanding phenomena of the properties according to various particle size of Seokcheong pigment.
수온별 아질산 급성 노출에 따른 넙치, Paralichthys olivaceus의 혈액학적 성상 및 혈장성분의 영향
홍수민 ( Su-min Hong ),조아현 ( A-hyun Jo ),김다은 ( Da-eun Kim ),박연숙 ( Yeon-sook Park ),이혜성 ( Hye-sung Lee ),전유현 ( Yu-hyeon Jeon ),김석렬 ( Seok-ryel Kim ),김대희 ( Dae-hee Kim ),강예재 ( Yue Jai Kang ),김준환 ( Jun-hwan 한국어병학회 2021 한국어병학회지 Vol.34 No.2
Olive flounder (Paralichthys olivaceus) (Weight 110.9±17.1 g, length 22.3±1.2 cm) were exposed to waterborne nitrite at 0, 30, 60, 120, 240, 480 and 960 mg NO<sub>2</sub> <sup>-</sup>/L according to water temperature at 20℃ and 25℃ for 96 hours. The lethal concentration 50 (LC<sub>50</sub>) of olive flounder, P. olivaceus exposed to waterborne nitrite was 513.87 mg NO<sub>2</sub> <sup>-</sup>/L at 20℃ and 208.35 mg NO<sub>2</sub> <sup>-</sup>/L at 25℃, which means a significant difference in LC<sub>50</sub> by the water temperature. Hemoglobin and hematocrit were significantly decreased by waterborne nitrite exposure. The inorganic component, plasma calcium, was significantly decreased, and the organic components such as plasma glucose and cholesterol were significantly decreased showing a similar tendency with calcium. In enzymatic components, the AST and ALP were also significantly decreased by nitrite exposure. The results of this study indicate that exposure to nitrite can affect the survival and hematological physiology of P. olivaceus, and the effect of exposure to nitrite had a significant effect on nitrite toxicity depending on the water temperature.