• The Journal of the Acoustical Society of Korea
• /
• 제28권3E호
• /
• pp.109-117
• /
• 2009

Identity recognition in real environment with a reliable mode is a key issue in human computer interaction (HCI). In this paper, we present a robust person identification system considering score-based optimal reliability measure of audio-visual modalities. We propose an extension of the modified reliability function by introducing optimizing parameters for both of audio and visual modalities. For degradation of visual signals, we have applied JPEG compression to test images. In addition, for creating mismatch in between enrollment and test session, acoustic Babble noises and artificial illumination have been added to test audio and visual signals, respectively. Local PCA has been used on both modalities to reduce the dimension of feature vector. We have applied a swarm intelligence algorithm, i.e., particle swarm optimization for optimizing the modified convection function's optimizing parameters. The overall person identification experiments are performed using VidTimit DB. Experimental results show that our proposed optimal reliability measures have effectively enhanced the identification accuracy of 7.73% and 8.18% at different illumination direction to visual signal and consequent Babble noises to audio signal, respectively, in comparison with the best classifier system in the fusion system and maintained the modality reliability statistics in terms of its performance; it thus verified the consistency of the proposed extension.

• 수산해양기술연구
• /
• 제22권4호
• /
• pp.56-60
• /
• 1986

The purpose of this study is to find the light intensity which induced maximum gathering rate and to observe the variation of the gathering rate both in daytime and at night by using rock trout, He.'~agrammos otakii (Jordan et starks). An experimental tank (360L x SOW x 55H cm) was set up in a dark room. An illumination system was attached to the end of one side of the tank to control horizontal light intensity. Six artificial light sources were prepared by combination of two light bulbs (5 W, 150 W) and seven filters. During the experiment water depth was maintained 50 em level in the tank. The tank was marked into six longitudinal sections each being 60 em long to observe the distribution of fish. The fish were acclimatized in dark condition for 50 minutes prior to the main experiment. Upon turning on the light, the number of fish in each section was counted 40 times every 30 seconds. and the gathering rate was obtained from the average number of fish in each section. The light intensity inducing maximum gathering rate was O. 7 lux (0. 5~1. 1 lux) in the daytime and 5. 2/ux (3.2-7.7 lux) at night. The variation of the gathering rate of fish in illumination time was sma II and showed the decreasing trend.

• 전기학회논문지
• /
• 제64권5호
• /
• pp.755-765
• /
• 2015

This paper is concerned with Interval Type-2 Radial Basis Function Neural Network classifier realized with the aid of Census Transform(CT) and (2D)2LDA methods. CT is considered to improve performance of face recognition in a variety of illumination variations. (2D)2LDA is applied to transform high dimensional image into low-dimensional image which is used as input data to the proposed pattern classifier. Receptive fields in hidden layer are formed as interval type-2 membership function. We use the coefficients of linear polynomial function as the connection weights of the proposed networks, and the coefficients and their ensuing spreads are learned through Conjugate Gradient Method(CGM). Moreover, the parameters such as fuzzification coefficient and the number of input variables are optimized by Artificial Bee Colony(ABC). In order to evaluate the performance of the proposed classifier, Yale B dataset which consists of images obtained under diverse state of illumination environment is applied. We show that the results of the proposed model have much more superb performance and robust characteristic than those reported in the previous studies.

• 방송공학회논문지
• /
• 제24권1호
• /
• pp.87-96
• /
• 2019

저조도 영상의 개선에 관한 연구는 대부분 대비 개선을 목적으로 한다. 저저도 영상에서 밝기 개선, 대조 개선, 및 조명 성분 감쇠 등의 다양한 연구가 진행됐다. 최근에 인공신경망으로 상기 방법들을 대체하는 연구가 진행 중이다. 본 논문에서는 Retinex 이론에 기반하여 조명 광원이 존재하는 저저도 영상으로부터 조명 성분을 감쇠하고, 반사 성분만을 생성하는 기법을 심층신경망으로 대체하는 방법을 제안한다. 실험에서는 102장의 저저도 영상으로 학습시킨 인공신경망으로 반사 영상을 생성하였는데, PSNR=30.8682(db), SSIM=0.4345를 얻었다.

• 한국항해항만학회지
• /
• 제47권2호
• /
• pp.85-92
• /
• 2023

The hardware of an LED lighting control system for coastal lighting at coastal pier entrance consists of a power supply unit, an AVR control unit, a CLCD output unit, an LED control unit, a scenario selection switch unit, and an operation speed display unit. It is made of an 8-channel. The CPU used ATmega128 and the FET was used to control the current signal. To operate the CPU, DC 12V was converted to DC 5V using a regulator 7805. A heat sink was used to remove heat generated in the FET. By connecting the load LED module to the manufactured 8-channel LED lighting control system, the operation was confirmed through various production scenarios. In addition, a control system was designed to show the most suitable color for the atmosphere of the coastal pier according to the input value of temperature and illumination using a fuzzy control system. Computer simulation was then conducted. Results confirmed that fuzzy control did not need to store many data inputs due to characteristics of artificial intelligence and that it could efficiently represent many output values with simple fuzzy rules.

• ETRI Journal
• /
• 제45권6호
• /
• pp.1022-1034
• /
• 2023

In this study, the impact of varying lighting conditions on recognition and decision-making was considered. The luminosity approach was presented to increase gesture recognition performance under varied lighting. An efficient framework was proposed for sensor-based sign language gesture identification, including picture acquisition, preparing data, obtaining features, and recognition. The depth images were collected using multiple Microsoft Kinect devices, and data were acquired by varying resolutions to demonstrate the idea. A case study was designed to attain acceptable accuracy in gesture recognition under variant lighting. Using American Sign Language (ASL), the dataset was created and analyzed under various lighting conditions. In ASL-based images, significant feature points were selected using the scale-invariant feature transformation (SIFT). Finally, an artificial neural network (ANN) classified hand gestures using specified characteristics for validation. The suggested method was successful across a variety of illumination conditions and different image sizes. The total effectiveness of NN architecture was shown by the 97.6% recognition accuracy rate of 26 alphabets dataset with just a 2.4% error rate.

• 대한소아치과학회지
• /
• 제34권2호
• /
• pp.183-191
• /
• 2007

본 연구는 Digital Imaging Fiber-Optic Trans-Illumination(DIFOTI)이나 Laser Fluorescence(LF)를 이용한 방법이 평활면 법랑질 초기우식증의 재광화 정도를 정확하게 monitoring 할 수 있는지의 여부를 평가하기 위하여 진행되었다. 우치 법랑질로부터 얻어진 인공 우식 절편을 특별히 제작된 의치에 식립한 다음, 10명의 실험 참가자들이 구강 내에 장착하고 3주 동안 불소 양치용액을 사용하도록 한 다음 Confocal Laser Scanning Microscopy(CLSM)로 측정된 병소 깊이를 gold standard로 사용하여 DIFOTI와 LF로 측정 된 우식 절편의 재광화 정도를 비교 평가한 후 다음과 같은 결론을 얻었다. 1. DIFOTI에 측정된 광도 백분율이 재광화 기간이 지남에 따라 증가하였으며, CLSM에서 측정된 병소 깊이와 유의한 역상관관계를 보였다(r=-0.683, p<0.01). 2. LF 측정치는 재광화 기간이 경과함에 따라 증가하였으며 CLSM에서 측정된 병소 깊이와 유의한 상관관계를 보였다(p<0.05). 3. CLSM 촬영 결과 500 ppm 불소 양치군이 0 ppm 불소 양치군보다 빠른 속도로 병소 깊이가 감소하는 양상을 보여 주었다.

• 한국태양에너지학회 논문집
• /
• 제24권2호
• /
• pp.51-62
• /
• 2004

Architects may evaluate building models to see how a building will shadow itself and its neighbors at various times. A heliodon, a tilt-table which is a machine that imitates the rotation and orbit of the Earth, helps architects wanting to analyze patterns of shadow patch, passive solar heating options, site solar panels, or control solar heat gain. The heliodon swivels in three directions for setting latitude, season, and time of day. Using the device, an architect first clamps a model to the tabletop, then turns the table to the coordinates of interest. Usually, the winter and summer solstices receive strong attention, for they represent extreme cases, A more recent installation at a university adds to its heliodon a set of lamps to recreate the illumination level and more accurate patterns of shadow patch. The table holds the building model at various angles to a spotlight, which mounts in the pole. The set of scale model measurement describes the validity of various electric lamps as an artificial sun to approximate the sun's parallel rays, helping designers to distinguish between illuminated areas in and around a building and those regions falling in the shadows.

• 한국태양에너지학회 논문집
• /
• 제35권6호
• /
• pp.1-7
• /
• 2015

This paper deals with the design and operation of a hybrid daylighting system that uses natural and artificial lighting to enhance visual comfort in buildings. The system was developed using an AVR micro controller for solar tracking in conjunction with dimming controls, which, acting together, enables the maximum use of natural daylight and also improves energy efficiency in buildings. Experimental results clearly demonstrates the usefulness of the present system capable of enhancing indoor lighting conditions when sufficient daylight is available and distributed appropriately in harmony with artificial lighting.

• 조명전기설비학회논문지
• /
• 제19권3호
• /
• pp.1-9
• /
• 2005

주거공간에서 빛은 인간이 삶을 영위하고 창조활동을 하기 위해서는 꼭 필요한 환경요소이다. 특히 현대인들은 많은 시간을 인공적으로 꾸며진 실내환경 속에서 인공적인 빛을 밝히고 생활하는데 익숙해 있다. 현재 주택에서의 빛환경은 공간마다 명시성 위주의 생활행위에만 기준을 두고 있어 평균조도가 증가되기는 하였다. 그러나 주거공간은 다양한 용도와 목적으로 여러 생활행위가 행해지므로 이것에 대응시킬 수 있는 다양한 조도설정이 필요하다. 본 연구에서는 주거공간에서의 빛의 본질을 파악하고, 공동주택을 대상으로 실내공간별 빛환경을 측정하였다. 이를 통해 공동주택의 조명평면 및 문제점을 분석하여 우리나라 주거조명환경에 적합한 조도기준을 제안하고자 하였다.