• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1512-1518
• /
• 2021

Since single cell RNA sequencing provides the expression profiles of individual cells, it provides higher cellular differential resolution than traditional bulk RNA sequencing. Using these single cell RNA sequencing data, clustering analysis is generally conducted to find cell types and understand high level biological processes. In order to effectively process the high-dimensional single cell RNA sequencing data fir the clustering analysis, this paper uses a variational autoencoder to transform a high dimensional data space into a lower dimensional latent space, expecting to produce a latent space that can give more accurate clustering results. By clustering the features in the transformed latent space, we compare the performance of various classical clustering methods for single cell RNA sequencing data. Experimental results demonstrate that the proposed framework outperforms many state-of-the-art methods under various clustering performance metrics.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.10
• /
• pp.39-46
• /
• 2013

Recently, the mobile industries grow up rapidly and their performances are improved. So the usage of mobile devices is increasing in our life. Also mobile devices equipped with a high-performance camera, so the image stitching can carry out on the mobile devices instead of the desktop. However the mobile devices have limited hardware to perform the image stitching which has a lot of computational complexity. In this paper, we have proposed improved feature descriptor extraction and matching method for efficient image stitching on mobile environment. Our method can reduce computational complexity using extension of orientation window and reduction of dimension feature descriptor when feature descriptor is generated. In addition, the computational complexity of image stitching is reduced through the classification of matching points. In our results, our method makes to improve the computational time of image stitching than the previous method. Therefore our method is suitable for the mobile environment and also that method can make natural-looking stitched image.

• The KIPS Transactions:PartA
• /
• v.10A no.4
• /
• pp.357-368
• /
• 2003

Evaluation of shape similarity for 3D models is essential in many areas - medicine, mechanical engineering, molecular biology, etc. Moreover, as 3D models are commonly used on the Web, many researches have been made on the classification and retrieval of 3D models. In this paper, we describe methods for 3D shape representation and major concepts of similarity evaluation, and analyze the key features of recent researches for shape comparison after classifying them into four categories including multi-resolution, topology, 2D image, and statistics based methods. In addition, we evaluated the performance of the reviewed methods by the selected criteria such as uniqueness, robustness, invariance, multi-resolution, efficiency, and comparison scope. Multi-resolution based methods have resulted in decreased computation time for comparison and increased preprocessing time. The methods using geometric and topological information were able to compare more various types of models and were robust to partial shape comparison. 2D image based methods incurred overheads in time and space complexity. Statistics based methods allowed for shape comparison without pose-normalization and showed robustness against affine transformations and noise.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.7C
• /
• pp.554-564
• /
• 2012

The vehicle recognition consists of two steps; the vehicle region detection step and the vehicle identification step based on the feature extracted from the detected region. Features using linear transformations have the effect of dimension reduction as well as represent statistical characteristics, and show the robustness in translation and rotation of objects. Among the linear transformations, the NMF(Non-negative Matrix Factorization) is one of part-based representation. Therefore, we can extract NMF features with sparsity and improve the vehicle recognition rate by the representation of local features of a car as a basis vector. In this paper, we propose a feature extraction using NMF suitable for the vehicle recognition, and verify the recognition rate with it. Also, we compared the vehicle recognition rate for the occluded area using the SNMF(sparse NMF) which has basis vectors with constraint and LVQ2 neural network. We showed that the feature through the proposed NMF is robust in the urban scene where occlusions are frequently occur.

• The Korean Journal of Financial Studies
• /
• v.5 no.1
• /
• pp.165-190
• /
• 1999

본 논문에서는 KOSPI 200 주가지수 선물의 만기효과와 베이시스의 행태를 체계적으로 헤지 의사결정에 반영하기 위한 몇 가지 방법을 실증분석하였다. 우선 베이시스의 동태적 운동형태를 명시적으로 설정하지 않고 통계적인 방법을 통하여 헤지해제시점이 선물만기에 접근함에 따라 베이시스가 변동되는 양상을 반영한 헤지비율을 산출한다. 그 다음에는 헤지기간 전체에 걸친 베이시스의 운동형태를 명시적으로 설정하여 이에 입각한 헤지비율을 계산한다. 명시적인 베이시스의 운동형태는 비확률적인 과정과 확률적인 과정으로 다시 구분하고, 이 각각에 입각하여 최적헤지활동을 결정한다. 모든 헤지활동은 가장 최근까지의 정보를 이용하여 사전적으로 미래 헤지기간에 대한 의사결정을 하게 된다. 그러한 헤지활동의 사후적인 결과는 베이시스 행태를 별도로 고려하지 않고 단순선형회귀분석만을 이용하여 산출된 헤지성과와 비교되고, 변동성 감소 및 손실감소의 측면에서 각 접근방법이 가지는 특징 및 효율성을 평가한다. 실증 분석 결과, 헤지의 성과를 제고하기 위하여 선물의 만기효과와 베이시스의 행태변화를 체계적으로 반영한 세 가지의 시도 중 어느 것도 위험-수익의 2차원적인 비교에서 베이시스의 행태변화를 명시적으로 반영하지 않은 전통적 단순회귀분석을 압도하지 못하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.111-114
• /
• 2008

본 논문에서는 고유영상 추출에 기반을 둔 동물표현의 수화인식 기법을 제안하였다. 여기서는 영상으로부터 구조적 특징을 효과적으로 추출하기 위하여 PCA를 이용한 방법을 제안한다. 이는 선형투영 및 차원의 감소에 따른 계산부하의 감소로 인식시간을 줄이기 위함이다. 제안된 기법을 10종류의 240*215 픽셀 동물표현의 수화(160개 : 1명*10종류*16동작)영상을 대상으로 City-block의 분류척도를 이용하여 실험한 결과 우수한 인식성능이 있음을 확인하였다.

• Geophysics and Geophysical Exploration
• /
• v.12 no.1
• /
• pp.85-98
• /
• 2009

The tau-p inversion algorithm is widely employed to generate starting models with most computer programs, which implement refraction tomography. This algorithm emphasises the vertical resolution of many layers, and as a result, it frequently fails to detect even large lateral variations in seismic velocities, such as the decreases which are indicative of shear zones. This study demonstrates the failure of the tau-p inversion algorithm to detect or define a major shear zone which is 50m or 10 stations wide. Furthermore, the majority of refraction tomography programs parameterise the seismic velocities within each layer with vertical velocity gradients. By contrast, the Generalized Reciprocal Method (GRM) inversion algorithms emphasise the lateral resolution of individual layers. This study demonstrates the successful detection and definition of the 50m wide shear zone with the GRM inversion algorithms. The existence of the shear zone is confirmed by a 2D analysis of the head wave amplitudes and by numerous closely spaced orthogonal seismic profiles carried out as part of a later 3D refraction investigation. Furthermore, an analysis of the shot record amplitudes indicates that a reversal in the seismic velocities, rather than vertical velocity gradients, occurs in the weathered layers. The major conclusion reached in this study is that while all seismic refraction operations should aim to provide as accurate depth estimates as is practical, those which emphasise the lateral resolution of individual layers generate more useful results for geotechnical and environmental applications. The advantages of the improved lateral resolution are obtained with 2D traverses in which the structural features can be recognised from the magnitudes of the variations in the seismic velocities. Furthermore, the spatial patterns obtained with 3D investigations facilitate the recognition of structural features such as faults which do not display any intrinsic variation or 'signature' in seismic velocities.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.2
• /
• pp.642-648
• /
• 2016

As 3D solid modeling prevails, a range of applications have become possible and intensive research on the integration of CAD/CAM has been conducted. As a consequence, methods to recognize the machining features from CAD models have been developed. On the other hand, generating a machining sequence using the machining features is still a problem due to a combinatorial problem with a large number of machining features. This paper proposes a new method that utilizes the precedence constraints through which the number of the combinations is reduced drastically. This method can automatically generate machining sequences requiring the lowest amount of machining time. An airplane part was used to test the usefulness of the proposed method.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.1
• /
• pp.50-55
• /
• 2001

This paper presents a korean isolated words recognition algorithm which used new endpoint detection method, auditory model, 2D-DCT and new distance measure. Advantages of the proposed algorithm are simple hardware construction and fast recognition time than conventional algorithms. For comparison with conventional algorithm, we used DTW method. At result, we got similar recognition rate for speaker dependent korean isolated words and better it for speaker independent korean isolated words. And recognition time of proposed algorithm was 200 times faster than DTW algorithm. Proposed algorithm had a good result in noise environments too.

• Proceedings of the KIEE Conference
• /
• 1999.11c
• /
• pp.757-759
• /
• 1999

여러 사람의 얼굴들 중에 특정 개인 얼굴을 찾는 문제나 인식하는 문제는 최근들어 법 집행이나 상업적 목적 등 여러 응용분야에서 요구되고 있어 학문적으로 활발히 연구되어지고 있다. 얼굴 인식 기술은 여러 가지 방법으로 연구되어 왔는데 그 중 'PCA를 이용한 얼굴 인식 방법'이 가장 효율적인 방법으로 알려졌다. 그러나 이 방법은 조명 변화에 따라 정확성이 떨어지는 단점이 있다. 그래서 Histogram equalization을 이용해 조명 변화에 영향을 줄였다. 그리고 인식의 정확성을 유지하면서 eigenface를 추출하는데 시간을 줄이기 위해 웨이블렛 변환을 이용해 저주파 성분이 포함된 영역만을 추출, 그 부분을 입력영상으로 사용해 입력 영상에서 처리해야하는 차원을 줄여 특징 추출하는데 시간을 감소시켰다. 그 결과 특징 추출하는데 시간을 크게 줄어든 반면, 심한 조명 변화에서도 90%이상의 높은 인식률을 유지할 수 있었다.