• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.293-296
• /
• 2007

필드 워핑은 상대적으로 많은 연산량을 요구하는 것에 반해 제어선을 이용하여 워핑 결과를 좀 더 세밀하게 제어할 수 있는 것이 장점이다. 필드 영상 워핑 및 모핑의 동작 특성상, 국부적인 영상 확대와 축소 및 회전 등과 같은 다양한 기하학적인 변형이 복합적으로 발생하게 되는데 역방향 매핑 과정에서 소스 영상과 목적 영상의 화소가 정수 화소 단위로 대응되지 않을 경우, 목적 영상에 대응시킬 화소값을 산출하기 위해 적합한 영상 보간 기술이 필요하다. 다양한 보간 기술들 중에서 평균적으로 우수한 결과를 제공하는 양선형 보간이 보편적으로 사용되고 있으나, 이 보간 기술은 대각선 방향의 윤곽선 재현에 한계를 가지고 있다. 본 논문은 필드 워핑과 모핑을 위한 pseudomedian 필터 기반 영상보간법을 제안하고 주관적인 화질을 분석 평가함에 그 목적이 있다. 제안된 보간 방법은 양선형 보간과 윤곽선 재현에 우수한 특성을 보이는 pseudomedian 필터 보간을 효과적으로 결합한 것이다. 제안된 방법에 따르면 다른 보간 기법에 비해 상대적으로 자연스러운 워핑 및 모핑 결과를 얻을 수 있다. 제안된 방법의 타당성과 보편성을 검증하기 위해 서로 다른 분포 특성을 갖는 영상을 대상으로 주간적인 화질 측면에서 그 성능을 분석 평가하였다.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.93-110
• /
• 2023

Recently, attempts have been made to convert unstructured text into vectors and to analyze vast amounts of natural language for various purposes. In particular, the demand for analyzing texts in specialized domains is rapidly increasing. Therefore, studies are being conducted to analyze specialized and general-purpose documents simultaneously. To analyze specific terms with general terms, it is necessary to align the embedding space of the specific terms with the embedding space of the general terms. So far, attempts have been made to align the embedding of specific terms into the embedding space of general terms through a transformation matrix or mapping function. However, the linear transformation based on the transformation matrix showed a limitation in that it only works well in a local range. To overcome this limitation, various types of nonlinear vector alignment methods have been recently proposed. We propose a vector alignment model that matches the embedding space of specific terms to the embedding space of general terms through end-to-end learning that simultaneously learns the autoencoder and regression model. As a result of experiments with R&D documents in the "Healthcare" field, we confirmed the proposed methodology showed superior performance in terms of accuracy compared to the traditional model.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.11
• /
• pp.1254-1261
• /
• 2007

Mc-OFDM(Multi code-Orthogonal Frequency Division Multiplexing) modulation has been developed for high-speed data transmission over wireless channels. However, this system suffers critical problem from relatively high PAPR(Peak to Average Power Ratio). To reduce PAPR, we suggest a new technique, called Mapping using (N+1)PSK and demonstrate the performance of mapped Mc-OFDM system. On the other side, in OFDM system, each sub-carrier experiences different fading, thus some sub-carriers may be completely lost because of deep fade. To solve a above mention problem, we propose cooperative communications that improve the reliability of sub-carriers through spatial diversity.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.9 no.6 s.46
• /
• pp.59-65
• /
• 2005

In this paper a constitutive model for large cracks in concrete and other brittle materials subject to dynamic and cyclic leading is presented. The suggested model is based on the plastic-damage model for cyclic leading. A numerical formulation based on the three-step return-mapping algorithm for the proposed large crack model is also present. The numerical examples show that the present algorithm works appropriately under dynamic leading and should be used in large crack problems to prevent excessive tensive plastic strain development causing unrealistic results.

• The KIPS Transactions:PartD
• /
• v.17D no.1
• /
• pp.59-66
• /
• 2010

Traditional OCR engines are designed to the scanned documents in calibrated environment. Three dimensional perspective distortion and smooth distortion in images are critical problems caused by un-calibrated devices, e.g. image from smart phones. To meet the growing demand of character recognition of texts embedded in the photos acquired from the non-calibrated hand-held devices, we address the problem in three categorical aspects: rotational invariant method of text region extraction, scale invariant method of text line segmentation, and three dimensional perspective mapping. With the integration of the methods, we developed an OCR for camera-captured images.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.33 no.2
• /
• pp.81-93
• /
• 2020

The fatigue characteristics of glass fiber reinforced plastic (GFRP) composites were studied under repeated loads using the finite element method (FEM). To realize the material characteristics of GFRP composites, Digimat, a mean-field homogenization tool, was employed. Additionally, the micro-structures and material models of GFRP composites were defined with it to predict the fatigue behavior of composites more realistically. Specifically, the fatigue characteristics of polybutylene terephthalate with short fiber fractions of 30wt% were investigated with respect to fiber orientation, stress ratio, and thickness. The injection analysis was conducted using Moldflow software to obtain the information on fiber orientations. It was mapped over FEM concerned with fatigue specimens. LS-DYNA, a typical finite element commercial software, was used in the coupled analysis of Digimat to calculate the stress amplitude of composites. FEMFAT software consisting of various numerical material models was used to predict the fatigue life. The results of coupled analysis of linear and nonlinear material models of Digimat were analyzed to identify the fatigue characteristics of GFRP composites using FEMFAT. Neuber's rule was applied to the linear material model to analyze the fatigue behavior in LCF regimen. Additionally, to evaluate the morphological and mechanical structure of GFRP composites, the coupled and fatigue analysis were conducted in terms of thickness.

• Proceedings of the KAIS Fall Conference
• /
• 2004.11a
• /
• pp.247-251
• /
• 2004

본 논문은 필드 기반 영상 워핑을 위한 Pseudomedian 필터 기반 영상 보간법을 제안하고 주관적인 화질을 분석$\cdot$평가함에 그 목적이 있다. 대표적인 영상 워핑 기술 중 하나인 필드 기반 워핑은 상대적으로 많은 연산량을 요구하는 것에 반해 제어선을 이용하여 좀 더 세밀하게 워핑 결과를 제어할 수 있는 것이 장점이다. 필드 기반 영상 워핑의 동작 특성상, 국부적인 영상 확대와 축소 및 회전등과 같은 다양한 기하학적인 변형이 복합적으로 발생하게 되는데 역방향 매핑 과정에서 소스영상과 목적 영상의 화소가 정수 화소 단위로 대응되지 않을 경우, 목적 영상에 대응시킬 화소값을 산출하기 위해 적합한 영상 보간 기술이 필요하다. 다양한 보간 기술들 중에서 평균적으로 우수한 결과를 제공하는 양선형 보간이 보편적으로 사용되고 있으나, 이 보간 기술은 대각선 방향의 윤곽선 재현에 한계를 가지고 있다. 본 논문에서는 필드 기반 영상 워핑을 수행할 시에 윤곽선 재현에 우수한 특성을 보이는 Pseudomedian 필터 보간과 양선형 보간을 효과적으로 결합하여 양자의 장점을 보간 결과에 반영함으로써 양선형 보간만을 이용하는 필드 기반 워핑 기술에 비해 자연스러운 워핑 결과를 제공할 수 있었다. 제안된 방법의 타당성과 보편성을 검증하기 위해 서로 다른 분포 특성을 갖는 영상을 대상으로 주간적인 화질 측면에서 그 성능을 분석$\cdot$평가하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.10
• /
• pp.881-886
• /
• 2013

Even though a multi-view camera system is able to capture multiple images at different viewpoints, the color distributions of captured multi-view images can be inconsistent. This problem decreases the quality of multi-view images and the performance of post-image processes. In this paper, we propose an improved polynomial model for effectively correcting the color inconsistency problem. This algorithm is fully automatic without any pre-process and considers occlusion regions of the multi-view image. We use the 5th order polynomial model to define a relative mapping curve between reference and source views. Sometimes the estimated curve is seriously distorted if the dynamic range of extracted correspondences is quite low. Therefore we additionally estimate the first order polynomial model for the bottom and top regions of the dynamic range. Afterwards, colors of the source view are modified via these models. The proposed algorithm shows the good subjective results and has better objective quality than the conventional color correction algorithms.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.104-112
• /
• 2012

Recently, the HDR imaging technique that mimics human eye is incorporated with LCD/LED display devices to deal with mismatch between the real world scene and the displayed image. However, HDR image has a veiling glare limit as well as a scale of the local contrast problem. In order to overcome these problems, several color correction methods, CSR(center/surround Retinex), MSR (Multi-Scale Retinex), tone-mapping method, iCAM06 and so on, are proposed. However, these methods have a dominated color throughout the entire resulting image after performing color correction. Accordingly, this paper presents a new color correction method using dynamic cone response function. The proposed color correction method consists of tone-mapping and dynamic cone response. The tone-mapping is obtained by using a linear interpolation between chromatic and achromatic. Thereafter, the resulting image is processed through the dynamic cone response function, which estimates the dynamic responses of human visual system as well as deals with mismatch between the real scene image and the rendered image. The experiment results show that the proposed method yields better performance of color correction over the conventional methods.