• 한국멀티미디어학회논문지
• /
• 제12권3호
• /
• pp.451-459
• /
• 2009

본 논문은 심전도(ECG) 신호로부터 조기심실수축(PVC)을 자동 탐지하는 방법으로 이산 웨이블릿 변환과 퍼지 신경망을 이용하는 방안을 제시하고 있다. 심전도 신호를 이산 웨이블릿 변환(DWT)으로 특징을 추출한 후, 퍼지 신경망으로 학습하여 정상 비트와 PVC 비트를 분류한다. 윈도우 크기는 R파를 기준으로 $-31/360{\sim}+32/360$초를 사용하며, 웨이블릿 변환은 d3, d4, d5의 웨이블릿 계수 14개를 사용한다. 퍼지 신경망은 가중 퍼지소속함수 기반 신경망을 사용한다. 본 논문은 벤치마킹 데이터로 MIT-BIH 부정맥 데이터베이스를 사용하여 Shyu 실험군(7개 레코드)에서는 전체 분류율에서 97.04% 보다 높은 99.91%의 신뢰성 있는 결과를 나타내었고, Inan 실험군(40개 레코드)에서는 각각 SE는 82.57% 보다 높은 84.67%, SP는 98.33% 보다 높은 99.39%, 전체 분류율은 96.85% 보다 높은 98.01%의 신뢰성 있는 결과를 나타내었다.

• Current Optics and Photonics
• /
• 제3권5호
• /
• pp.408-414
• /
• 2019

In this paper we propose a method to generate a true color image in scanning white-light interferometry (SWLI). Previously, a true color image was obtained by using a color camera, or an RGB multichannel light source. Here we focused on acquiring a true color image without any hardware changes in basic SWLI, in which a monochrome camera is utilized. A Fourier transform method was used to obtain the spectral intensity distributions of the light reflected from the sample. RGB filtering was applied to the intensity distributions, to determine RGB values from the spectral intensity. Through color corrections, a true color image was generated from the RGB values. The image generated by the proposed method was verified on the basis of the RGB distance and peak signal-to-noise ratio analysis for its effectiveness.

• Journal of the Optical Society of Korea
• /
• 제5권3호
• /
• pp.93-98
• /
• 2001

In order to verify the performance of the two-stage iterative Fourier transform algorithm[Hankook Kwanghak Hoeji 11, 47 (2000)], a number of phase-only diffractive pattern elements which produce simple 16x16-pixel intensity patterns useful in the field of optical information processing have been designed and their performance has been compared with that from the nonlinear least-squares algorithm[Appl. Opt. 36, 7297(1977)] which is computationally intensive. for all intensity patterns, elements designed by the former algorithm show better overall signal-to noise ratio and uniformity, although they show essentially the same diffraction efficiency. In the case of continuous phase elements, they show far superior uniformity. Computationally,. the former algorithm is far more efficient than the latter.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제3권5호
• /
• pp.163-172
• /
• 2014

Scale Invariant Feature Transform (SIFT)은 높은 매칭 능력과 회전이나 스케일 조정 시 안정성으로 인해 이미지 특징 매칭을 위해 많은 응용에서 사용되어지고 있으며, 이러한 특성으로 인해 카피-무브 위조 검출을 위한 핵심 알고리즘으로 각광받고 있다. 하지만 SIFT 변환은 이미지 조작의 증거를 감출 수 있는 안티포렌식의 가능성이 높음에도 불구하고 이에 대한 연구는 거의 없으므로, 본 논문에서는 의미론적으로 허용될 수 있는 왜곡을 적용하여 SIFT 기반 카피-무브 위조 검출을 방해하기 위한 타켓 카운터-포렌식 기법을 제안한다. 제안 기법은 공격자가 유사성 매칭 절차를 속일 수 있는 동시에 SIFT 키포인트의 변형을 통한 추적을 방해하여 이미지 조작의 증거를 숨길 수 있는 방안을 제공한다. 또한 제안 기법은 의미론적 제약 하에서 가공된 이미지와 원본 이미지 간의 높은 충실도를 유지하는 특성을 가진다. 한편, 다양한 조건의 테스트 이미지에 대한 실험을 통해 제안 기법의 효율성을 확인하였다.

• 감성과학
• /
• 제5권3호
• /
• pp.53-59
• /
• 2002

The surface fibers on the fabric is one of decisive factors which affects human sensory evaluation as well as heat and moisture transfer characteristics. In this study the length and distribution of surface fibers that are extruded from the fabric surface of the wool/wool blend fabrics (14 wool fabrics and 10 wool blend fabrics) and its contribution to subjective sensory evaluation were investigated. In order to quantify the length and distribution of surface fibers, image analysis and wavelet transform technique were introduced. Instant warm-cool feeling of touch, Q$\_$max/, and contact area were also measured and related to the quantified surface fibers. To figure out the effect of surface characteristics on sensory evaluation, human sensory responses to three adjectives which represent surface characteristics and warm-cool feeling of touch were obtained and analyzed. The relationship between the quantified surface fibers assessed by wavelet energy and both warm-cool reeling of touch, Qmax, and human sensory response were discussed.

• Food Science and Biotechnology
• /
• 제18권5호
• /
• pp.1199-1203
• /
• 2009

Melamine is a chemical intermediate to manufacture amino resins and plastics, which cannot be used as food additive since it can cause kidney stones. A qualitative determination method of melamine residue in powdered and liquid milk was developed using Fourier transform infrared (FTIR) spectroscopic technique. The calibration sets consisted of 21 standard melamine solutions, in which 1% trichloroacetic acid aqueous solution and acetonitrile (3:1, v/v) were used as solvent. The model was validated using 10 standard melamine solutions which were unused to build up the calibration set. Infrared (IR) absorbance peaks specific to almost all chemical groups in melamine molecule were shown in the spectral range between 1,100 and 1,800/cm. Combined partial least squares (PLS)-$2^{nd}$ derivative calibration model coupled with mean centering (MC) mathematical enhancement showed the highest correlation coefficients ($R^2$>0.99). In brief, the FTIR technique can be used for quantitative analysis of melamine in milk samples.

• Bulletin of the Korean Chemical Society
• /
• 제15권5호
• /
• pp.369-374
• /
• 1994

Self-assembled (SA) and Langmuir-Blodgett (LB) monolayers of arachidic acid on silver surfaces have been investigated by a reflection-absorption Fourier transform infrared spectroscopy. Arachidic acid was adsorbed on silver as carboxylate with its two oxygen atoms bound symmetrically to the surface. Although both the SA and LB monolayers consisted of fully extended trans zigzag carbon chains, a closer examination indicated that the SA monotayers should possess a more ordered crystalline structure than the LB monolayers. The infrared spectral data dictated that the extent of methyl group exposure at the air-film interface was greater in the SA monolayers than the LB monolayers, in agreement with the contact angle measurement. From a theoretical analysis, the alkyl chains in each monolayers seemed to be tilted away from the surface normal by less than $3.5^{\circ}$, but in opposite directions. Arachidic acid monolayers were concluded to have same structure as stearic acid monolayers.

• 제어로봇시스템학회논문지
• /
• 제3권5호
• /
• pp.511-519
• /
• 1997

This paper presents the automatic recognition algorithm of the license number in on vehicle image. The proposed algorithm uses the correlation coefficient and Hough transform to detect license plate. The m/n ratio reduction is performed to save time and memory. By the correlation coefficient between the standard pattern and the target pattern, licence plate area is roughly extracted. On the extracted local area, preprocessing and binarization is performed. The Hough transform is applied to find the extract outline of the plate. If the detection fails, a smaller or a larger standard pattern is used to compute the correlation coefficient. Through this process, the license plate of different size can be extracted. Two algorithms to each separate number are proposed. One segments each number with projection-histogram, and the other segments each number with the label. After each character is separated, it is recognized by the neural network. This research overlomes the problems in conventional methods, such as the time requirement or failure in extraction of outlines which are due to the processing of the entire image, and by processing in real time, the practical application is possible.

• 한국식품과학회지
• /
• 제48권5호
• /
• pp.418-423
• /
• 2016

근적외선분광분석기와 에너지 분산형 X선 형광분석기를 이용한 분석방법을 개발하여 각각 97.5, 98.0%의 높은 정확도의 판별식을 확립하였고, 시중 유통 시료를 분석하여 검증한 결과 각각 96.3, 95.0%의 판별 정확도를 확인하였다. 이상의 연구 결과를 통하여 근적외선분광분석기와 에너지 분산형 X선 형광분석기를 이용하여 청국장 원산지 판별이 가능함을 확인하였고 이는 유기성분 함량에 따른 근적외선 흡광도와 무기성분 함량에 따른 X선 형광에너지 강도가 국내산과 수입산 간에 차이가 있기 때문으로 사료된다.

• 한국지능시스템학회논문지
• /
• 제14권2호
• /
• pp.150-155
• /
• 2004

본 논문에서는 사람의 음성속에 내포된 6가지 기본 감정(기쁨, 슬픔, 화남, 놀람, 공포, 혐오)의 특징을 추출하고 인식하고자 한다. 제안한 감정인식 알고리즘은 웨이블렛 필터뱅크를 이용하여 각각의 감정별 코드북을 만들고, 인식단계에서 필터뱅크별 감정을 확인한 후 최종적으로 다중의사결정기법에 의해 감정을 인식하는 구조로 이루어져 있다. 이와 같은 웨이블렛 필터뱅크와 다중의사 결정기법에 기반을 둔 알고리즘의 유용성을 보이기 위해 실험에 사용된 음성은 20명의 화자로부터 6가지의 감정을 대상으로 각각 3번씩 발음한 감정음성을 녹음하여 총 360개의 데이터베이스로 구성하고 실험하였다. 이와 같이 제안한 알고리즘은 기존의 연구에 비해 5% 이상 향상된 인식률을 보였다.