• Journal of Multimedia Information System
• /
• 제6권2호
• /
• pp.91-98
• /
• 2019

Many researchers have attempted to recognize three-dimensional faces using feature points extracted from two-dimensional facial photographs. However, due to the limit of flat photographs, it is very difficult to recognize faces rotated more than 15 degrees from original feature points extracted from the photographs. As such, it is difficult to create an algorithm to recognize faces in multiple angles. In this paper, it is proposed a new algorithm to recognize three-dimensional face recognition based on feature points extracted from a flat photograph. This method divides into six feature point vector zones on the face. Then, the vector value is compressed and expanded according to the rotation angle of the face to recognize the feature points of the face in a three-dimensional form. For this purpose, the average of the compressibility and the expansion rate of the face data of 100 persons by angle and face zone were obtained, and the face angle was estimated by calculating the distance between the middle of the forehead and the tail of the eye. As a result, very improved recognition performance was obtained at 30 degrees of rotated face angle.

• 정보관리학회지
• /
• 제26권3호
• /
• pp.261-278
• /
• 2009

기계학습 기반 문서범주화 기법에 있어서 최적의 자질을 구성하는 것이 성능향상에 있어서 중요하다. 본 연구는 학술지 수록 논문의 필수적 구성요소인 저자 제공 키워드와 논문제목을 대상으로 자질확장에 관한 실험을 수행하였다. 자질확장은 기본적으로 선정된 자질에 기반하여 WordNet과 같은 의미기반 사전 도구를 활용하는 것이 일반적이다. 본 연구는 키워드와 논문제목을 대상으로 WordNet 동의어 관계 용어를 활용하여 자질확장을 수행하였으며, 실험 결과 문서범주화 성능이 자질확장을 적용하지 않은 결과와 비교하여 월등히 향상됨을 보여주었다. 이러한 성능향상에 긍정적인 영향을 미치는 요소로 파악된 것은 정제된 자질 기반 및 분류어 기준의 동의어 자질확장이다. 이때 용어의 중의성 해소 적용과 비적용 모두 성능향상에 영향을 미친 것으로 파악되었다. 본 연구의 결과로 키워드와 논문제목을 활용한 분류어 기준 동의어 자질 확장은 문서 범주화 성능향상에 긍정적인 요소라는 것을 제시하였다.

• 융합신호처리학회논문지
• /
• 제2권3호
• /
• pp.24-31
• /
• 2001

K-L 전개 방법은 특징의 차원을 효과적으로 압축하므로 인식 처리에서 계산량을 줄일 수 있는 방법으로 잘 알려져 있다. 본 논문에서는 한국어 인식 시스템의 인식 정도를 개선하기 위해, 음성의 특징 파라미터에 대하여 효과적으로 K-L전개를 적용하는 방법(K-L 계수)을 제안한다. 그리고 제안한 방법으로 얻어진 새로운 음성 특징 파라미터를 이용하여 화자 독립 연속 숫자음 인식실험을 수행하고, 기존의 Mel-cepstrum과 회귀계수의 인식 결과와 비 교, 분석하였다. 인식 실험 결과, 제안한 K-L 계수를 이용한 방법이 기존의 방법보다 높은 인식률을 얻어 제안한 방법의 유효성을 확인할 수 있었다.

• 정보과학회 논문지
• /
• 제43권9호
• /
• pp.1008-1014
• /
• 2016

트위터, 페이스북, 온라인 고객 리뷰 등은 신문기사처럼 정제된 글이 아닌 자유롭게 기술되는 비격식(informal) 텍스트 문서에 속한다. 이러한 비격식 문서에서 일관된 규칙이나 패턴을 찾는 일은 격식(formal) 문서 경우에 비해 용이하지 않기 때문에, 비격식 문서 분석을 위해서는 성능 개선을 위한 추가적인 접근 방법 필요다고 판단된다. 본 연구에서는 대표적 비격식 문서인 트위터 데이터를 열 가지 카테고리로 분류함에 있어 LDA(Latent Dirichlet allocation) 단어 분포를 사용하여 자질(feature)을 교정하고 확장한다. 토픽별로 상위에 랭크된 단어 자질들을 기반으로 다른 단어 자질들을 분해 및 병합하는 방식으로 유용한 자질 집합을 반복적으로 확장시킨다. 이렇게 생성된 자질로 문서 분류를 수행한 결과 자질 확장 이전에 비해 마이크로 평균 F1-score 7.11%p의 성능 개선 효과를 확인할 수 있었다.

• 전기학회논문지
• /
• 제57권4호
• /
• pp.560-566
• /
• 2008

Today, as the power trades between generation companies and power customer are liberalized, the uncertainty level of operated power system is rapidly increased. Therefore, transmission operators as decision makers for transmission expansion are required to establish a deliberate investment plan for effective operations of transmission facilities considering forecasted conditions of power system. This paper proposes the methodology for the optimal solution of transmission expansion in deregulated power system. The paper obtains the expected value of transmission congestion cost for various scenarios by using occurrence probability. In addition, the paper assumes that increasing rates of loads are the probability distribution and indicates the location of expanded transmission line, the time for transmission expansion with the minimum cost for the future by performing the Montecarlo simulation. To minimize the investment risk as the variance of the congestion cost, Mean-Variance Markowitz portfolio theory is applied to the optimization model by the penalty factor of the variance. By the case study, the optimal solution for transmission expansion plan considering the feature of market participants is obtained.

• International Journal of Advanced Culture Technology
• /
• 제6권3호
• /
• pp.173-177
• /
• 2018

Masks are generated by adding different fonts of learning data characters in pixel unit, and pixel values belonging to each of the masks are divided into 3 groups. Using the directional expansion operators, we expand the text area of the test data character into 4 diagonal directions in order to create the boundary areas to distinguish it from the background area. A mask with a minimum average discordance is selected as the final recognition result by calculating the degree of discordance between the expanded test data and the masks. Image comparison using directional expansion operations more accurately recognizes test data through 4 subdivided recognition processes. It is also possible to expand the ranges of 3 groups of pixel values of masks more evenly such that new fonts can easily be added to the given learning data.

• 한국멀티미디어학회논문지
• /
• 제7권12호
• /
• pp.1680-1691
• /
• 2004

One of the representative methods of optical flow is a gradient method which estimates the movement of an object based on the differential of image brightness. However, the method is ineffective for large displacement of the object and many improved methods have been proposed to copy with such limitations. One of these improved techniques is the multigrid processing, which is used in many optical flow algorithms. As an alternative novel technique we have been proposing an orthogonal functional expansion method, where whole displacements are expanded from low frequency terms. This method is expected to be applicable to flow estimation with large displacement and deformation including expansion and contraction, which are difficult to cope with by conventional optical flow methods. In the orthogonal functional expansion method, the apparent displacement field is calculated iteratively by a projection method which utilizes derivatives of the invariant constraint equations of brightness constancy. One feature of this method is that differentiation of the input image is not necessary, thereby reducing sensitivity to noise. In this paper, we apply our method to several real images in which the objects undergo large displacement and/or deformation including expansion. We demonstrate the effectiveness of the orthogonal functional expansion method by comparing with conventional methods including our optimally scaled multigrid optical flow algorithm.

• 천문학회보
• /
• 제44권2호
• /
• pp.54.1-54.1
• /
• 2019

We use iterative smoothing reconstruction method along with exploring in the parameter space of the light curves of the JLA supernova compilation (Joint Light-curve Analysis) to simultaneously reconstruct the expansion history of the universe as well as putting constrains on the light curve parameters without assuming any cosmological model. Our constraints on the light curve parameters of the JLA from our model-independent analysis seems to be closely in agreement with results assuming ΛCDM cosmology or using Chevallier-Polarski-Linder (CPL) parametrization for the equation of state of dark energy. This implies that there is no hidden significant feature in the data that could be neglected by cosmology model assumption. The reconstructed expansion history of the universe and properties of dark energy seems to be in good agreement with expectations of the standard ΛCDM model. Our results also indicate that the data allows a considerable flexibility for expansion history of the universe.

• 대한의용생체공학회:의공학회지
• /
• 제43권6호
• /
• pp.441-447
• /
• 2022

In this paper, we propose a pill identification model using engraved text feature and image feature such as shape and color, and compare it with an identification model that does not use engraved text feature to verify the possibility of improving identification performance by improving recognition rate of the engraved text. The data consisted of 100 classes and used 10 images per class. The engraved text feature was acquired through Keras OCR based on deep learning and 1D CNN, and the image feature was acquired through 2D CNN. According to the identification results, the accuracy of the text recognition model was 90%. The accuracy of the comparative model and the proposed model was 91.9% and 97.6%. The accuracy, precision, recall, and F1-score of the proposed model were better than those of the comparative model in terms of statistical significance. As a result, we confirmed that the expansion of the range of feature improved the performance of the identification model.

• 한국음향학회지
• /
• 제15권2호
• /
• pp.40-48
• /
• 1996

본 논문은 단음절 전구간에 대해 4프레임폭과 7프레임폭을 결합하여 만든 40차원의 세그먼트를 K-L전개와 신경망으로 각각 10, 14, 20차원으로 압축하여 연속분포 HMM의 음성인식 특징파라미터로 사용하였다. 그리고 이산지속시간, 희귀계수 그리고 혼합분포를 특징 파라미터로 추가한 경우와 비교검토하였다. 단음절 100개에 대한 인식실험결과 연속분포 HMM의 인식률 85.19%에 비해 희귀계수를 부가한 경우 1.4%, 혼합분포를 이용한 경우 2.36%, 이산 지속시간제어를 한 경우 2.78%의 인식률이 향상되었다. 그리고 K-L전개에 의한 압축파라미터만 이용한 경우는 멜켑스트럼 + 희귀계수의 경우보다 인식률이 낮았으나, K-L전개에 의한 압축파라미터에 멜켑스트럼과 희귀계수를 부가한 경우는 동등한 결과를 얻을 수 있었다. 신경망에 의한 압축파라미터를 이용한 경우에는 비선형 변환인 시그모이드 함수를 사용하므로 음성의 동적변화가 잘 반영되어 K-L전개 및 다른 방법에 비해 향상된 인식결과를 얻을 수 있었다.