• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.444-447
• /
• 2022

최근 몇 년 동안 더 깊은 신경망 아키텍처로 인해 컴퓨터 비전 이미지 분류 작업이 더 빠르고 더 좋아졌다. 그러나 대부분의 이미지 분류 작업은 특정 이미지 모양(예: 고양이와 개 구별)을 기반으로 분류하도록 설계되었지만 낮과 밤 또는 사계절과 같은 기간을 구별하도록 훈련된 분류 모델은 많지 않다. 같은 장소의 사계절 이미지를 구분하기 위한 선행 연구는 있는 반면 일반 영상의 계절 분류 연구는 현재 부재한 실정이다. 그래서 본 논문에서는 일반 영상의 계절 분류 문제에 대한 다양한 접근 방식을 제시한다. 간단한 특징 추출부터 합성곱 신경망 구축, 전이 학습에 이르기까지 계절별 이미지 분류를 위한 세 가지 방법을 연구하고 정확도 결과를 비교, 분석하였다.

• Journal of Radiation Protection and Research
• /
• 제24권3호
• /
• pp.161-170
• /
• 1999

비정질 실리콘 방사선 촬영기와 기존 X-ray filmrh과의 영상질 비교를 통하여 시스템의 성능을 평가하였다. 다양한 영상질 평가를 위하여 MTF (Modulation Transfer Function), NPS (Noise Power Spectrum), Contrast를 측정하여 계산하였다. 실험 결과 DR의 MTF는 기존 film-screen system과 유사하였다. Noise 특성은 두 시스템 모두 quantum noise가 주를 이뤘으며, 특히 DR 에서는 전기적인 noise가 발견되었다. 또한 우수한 sensitivity 특성과 영상처리를 통하여 DR 시스템이 기존의 film-screen 시스템보다 놀은 대조도를 보였다. 이와 같이 DR에서는 기존 film-screen 시스템과 유사한 해상도와 영상처리를 통해 같은 촬영조건에서도 향상된 대조도의 영상을 얻을 수 있었다. 본 연구를 통하여 의학적 활용과 관련된DR 시스템의성능에 대한 유용한 정보를 제공할 수 있으리라 기대한다.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.158-168
• /
• 2024

There is hardly any person in modern times who has not taken soft drinks instead of drinking water. The rate of people taking soft drinks being surprisingly high, researchers around the world have cautioned from time to time that these drinks lead to weight gain, raise the risk of non-communicable diseases and so on. Therefore, in this work an image-based tool is developed to monitor the nutritional information of soft drinks by using deep convolutional neural network with transfer learning. At first, visual saliency, mean shift segmentation, thresholding and noise reduction technique, collectively known as 'pre-processing' are adopted to extract the location of drinks region. After removing backgrounds and segment out only the desired area from image, we impose Discrete Wavelength Transform (DWT) based resolution enhancement technique is applied to improve the quality of image. After that, transfer learning model is employed for the classification of drinks. Finally, nutrition value of each drink is estimated using Bag-of-Feature (BoF) based classification and Euclidean distance-based ratio calculation technique. To achieve this, a dataset is built with ten most consumed soft drinks in Bangladesh. These images were collected from imageNet dataset as well as internet and proposed method confirms that it has the ability to detect and recognize different types of drinks with an accuracy of 98.51%.

• 대한방사선기술학회지:방사선기술과학
• /
• 제27권3호
• /
• pp.25-29
• /
• 2004

방사선부문 디지털시스템 중 일반촬영 부문에서 가장 많이 사용하고 있는 Computed radiography(CR) 영상판(image plate)의 검출시스템은 형광체의 빛 확산 영향을 받는 시스템이므로 영상을 흐리게 만들기 마련이다. 본 연구에서는 X-ray 발생장치를 이용하여 computer radiography 영상판의 정보전달능(modulation transfer function, MTF) 측정을 통해 해상도와 선예도 등을 알아보기 위해 현재 A병원에서 사용중인 일반촬영용 IP 사용년수 3년 이상된 Fuji ST-Vn 모델과 사용년수 2년 미만인 Fuji ST-VI 모델을 대상으로 MTF를 측정하여 분석 비교하였다. 실험방법은 X선 발생장치를 이용하여 resolution chart법으로 시행하였으며, 레이저프린터로 출력하여 microdensitometer로 측정하였다. 실험결과, 2가지의 IP 모델은 차이가 있었으며, 해상도와 선예도 모두 Fuji ST-VI 모델이 우수하게 나타났다.

• 한국가구학회지
• /
• 제17권3호
• /
• pp.79-85
• /
• 2006

Temperature of board surface was monitored during drying using an IR image measurement system. Boards were water-saturated and dried at the levels of four temperatures and three air velocities. At higher DB the surface temperature increased more steeply and level off period was significantly short. At the DB temperatures of 70, 80, $90^{\circ}C$ the period where the surface temperature was equivalent to WB temperature was constant regardless of air velocity while at $60^{\circ}C$ it decreased as air velocity increased. It was confirmed that a surface transfer coefficient increased with DB temperature. Variation of temperature profile on a wood surface increased with DB temperature and air velocity.

• 제어로봇시스템학회논문지
• /
• 제7권3호
• /
• pp.255-262
• /
• 2001

According to the advancement of web related technologies, many works on robots using these technologies, called web-based robots enables sharing of expensive equipments as well as control of remote robots. But none of the existing methods about web-based robots in-clude robot simulators in their web browser, which transfer appropriate information of a remote place to the local users. In this paper, a web-based robot simulator is proposed and developed to control a remote robot by using the web. The proposed simulator can transfer the 3D information about the remote robot to the local users by using 3D graphics, which has not been previously developed. Also, it sends the camera image of a remote place to the local users so that the users can use this camera image as well as 3D information in order to control the remote robot.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -1
• /
• pp.237-240
• /
• 2000

We have proposed a simple method based on IIR filters for realizing sound image localization. How-ever the nonlinearity of phase characteristics of the IIR filters, which are used for sound image localization, cause decrease of the localization accuracy. In this paper we investigate the influence of phase characteristics on the sound localization. Head-related transfer functions (HRTFs) of a dummy-head are approximated by the IIR filter. We carried out sound image localization experiment with 2-loudspeaker reproduction using the approximated HRTFs. Then the errors which obtained from experiments were compared with the theoretical values which were estimated from the phase shifts of the IIR filters. As a result there was little influence of the nonlinear phase characteristics of the IIR fitters in the localization on the horizontal plane.

• 대한디지털의료영상학회논문지
• /
• 제10권2호
• /
• pp.23-31
• /
• 2008

Computed radiography(CR) has been widely used in the field of diagnostic radiography since digital X-ray image was introduced. The imaging performance of CR system was studied by analyzing the digital image data of the CR images which are the outcomes of the whole imaging system composed of image plate(IP), laser digitizer, analoge-digital convertor, and a given image processing unit. In this study, we used a conventional CR system made by Agfa. From the flat field image of 150$\times$150 image pixels, signal-to-noise ratio(SNR) was calculated. SNR of the CR image increases in proportion to logarithm value of the X-ray exposure irradiated on the IP. SNR is less than about 6 at the exposure below 0.2mR and is more than 10 at the exposure above 0.54mR. In our study, most of images obtained by the smaller exposures less than 2.0mR can not be readable. In general, the minimum value of the SNR ranges from 3 to 5. We obtained modulation transfer function(MTF) by analyzing the bar pattern image which was made under conditions as follows: X-ray tube potential was 55kVp, the IP exposure was 0.54 mR, and the distance between X-ray source to IP was 2m, where bar pattern was located on the IP. MTF is 23% at 2.5lp/mm spatial frequency. Provided that the MTF of noise equivalent modulation is 10%, the CR system has the limiting spatial resolution of 3.2lp/mm. If the image sharpness is evaluated by the spatial frequency where MTF is 50%. the corresponding spatial frequency is 0.5$\sim$0.75lp/mm. MTFA(Modulation Transfer Function Area) is 1.0lp/mm. Compared with the Fuji CR whose MTFA is 1.1lp/mm, Agfa CR in this study shows almost same MTFA performance.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.114-117
• /
• 2000

In this paper, we proposed robust digital image watermarking based on modulation transfer function (MTF) of human visual system (HVS). Using the proposed method, robust watermarking is possible both in common image processing operations such as cropping and lossy compression and in geometrical transforms such as rotation, scaling, and translation, because it can embed watermark and template signal maximally using MTF of HVS. Experimental results show that the proposed watermarking method is more robust to several common image processing operations and geometrical transforms.

• 한국가시화정보학회:학술대회논문집
• /
• 한국가시화정보학회 2004년도 추계학술대회 논문집
• /
• pp.22-25
• /
• 2004

The present paper investigated the correlation between acoustic pressure and heat transfer augmentation in acoustic fields. The acoustic pressure predicted by numerical work and compared with the augmentation ratio of heat transfer coefficient was experimentally measured. Also, particle image velocimetry(PIV) was used for the visualization of velocity vectors and kinetic energy distribution inside liquid region. For the numerical work, SVS programed with Fortran language and based on a coupled FE-BEM was used. Results of the present study, the acoustic pressure is increased by $60\%$ and the largest augmentation of heat transfer about $28\%$ was measured. Finally, the profiles of acoustic pressure is consistent with that of augmentation of heat transfer. It is concluded that a correlation exists between the acoustic pressure and the heat transfer augmentation.