http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
거대 점오염원이 주변 대기질에 미치는 영향에 관한 연구
김유근,이화운,전병일,장은숙,홍정혜,문윤섭,원경미,송정희 부산대학교 환경문제연구소 1996 環境硏究報 Vol.14 No.1
In order to show the effect of a vast point pollutant source on air quality of Pusan Thermoeletric Power Plant and its surrounding area, air quality around Pusan Thermoeletric Power Plant was simulated by ISCLT-2 which was supplied by EPA. For this purpose the emission amount of SO_2, NO_2 and TSP was calculated and atmospheric stability was classified for a recent decade(1985~1994) in Pusan. A result of the emission amount showed that much amount of NO_2, NO_2 and TSP are emitted from industrial area. It was clear that NO_2 is much emitted from line source and industrial area. And as a result of classification of atmospheric stability, neutral, stable and unstable state were 58%, 24.1% and 17.9%, respectivly. The result of ai quality simulation by ISCLT-2 showed that Pusan Thermoeletric Power Plant is affecting on the increse of 2.0ppb, 3.0ppb and 5.0㎍/㎥, SO_2, NO_2, and TSP respectively at its surrounding area, site A-3 which was located westward 2.2㎞ distance from Plant
Bilinear Model-Based Maximum Likelihood Linear Regression Speaker Adaptation Framework
Hwa Jeon Song,Hyung Soon Kim IEEE 2009 IEEE signal processing letters Vol.16 No.12
<P>This letter proposes a novel framework for speaker adaptation, using bilinear model-based maximum likelihood linear regression (MLLR) method. First, a set of speaker models is decomposed into the style factor identified as each speaker's characteristics and the common content factor across the speakers, by the bilinear model. Then, using some adaptation data from a new speaker, the speaker-specific model is generated by properly adjusting the dimensionality of the content factor and estimating a new style factor simultaneously. Experimental results show that the proposed framework outperforms MLLR with fewer number of parameters to be estimated.</P>
효과적인 2차 최적화 적용을 위한 Minibatch 단위 DNN 훈련 관점에서의 CNN 구현
송화전(Song, Hwa Jeon),정호영(Jung, Ho Young),박전규(Park, Jeon Gue) 한국음성학회 2016 말소리와 음성과학 Vol.8 No.2
This paper describes some implementation schemes of CNN in view of mini-batch DNN training for efficient second order optimization. This uses same procedure updating parameters of DNN to train parameters of CNN by simply arranging an input image as a sequence of local patches, which is actually equivalent with mini-batch DNN training. Through this conversion, second order optimization providing higher performance can be simply conducted to train the parameters of CNN. In both results of image recognition on MNIST DB and syllable automatic speech recognition, our proposed scheme for CNN implementation shows better performance than one based on DNN.