• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권8호
• /
• pp.2881-2894
• /
• 2014

In complicated environment, context information plays an important role in image segmentation/labeling. The recently proposed auto-context algorithm is one of the effective context-based methods. However, the standard auto-context approach samples the context locations utilizing a fixed radius sequence, which is sensitive to large scale-change of objects. In this paper, we present a scale invariant auto-context (SIAC) algorithm which is an improved version of the auto-context algorithm. In order to achieve scale-invariance, we try to approximate the optimal scale for the image in an iterative way and adopt the corresponding optimal radius sequence for context location sampling, both in training and testing. In each iteration of the proposed SIAC algorithm, we use the current classification map to estimate the image scale, and the corresponding radius sequence is then used for choosing context locations. The algorithm iteratively updates the classification maps, as well as the image scales, until convergence. We demonstrate the SIAC algorithm on several image segmentation/labeling tasks. The results demonstrate improvement over the standard auto-context algorithm when large scale-change of objects exists.

• 무역학회지
• /
• 제45권1호
• /
• pp.137-153
• /
• 2020

On February 14, 2019, the government of Korea formally decided to consider the feasibility of joining the Comprehensive and Progressive Trans-Pacific Partnership (CPTPP) and has since been conducting bilateral consultations with individual member countries. In terms of the impact estimation, the CPTPP is actually a Korea-Japan FTA, and the most sensitive issue in the FTA is the opening of the auto industry market to Japan. Despite these circumstances, previous studies have predicted that the auto industry will be a beneficiary industry when joining the CPTPP. However, the Korean auto industry is opposed to joining the CPTPP. In order to investigate the cause of this discrepancy, this paper examines the problems of previous studies in estimating the impact of joining the CPTPP and found that the preceding study did not consider the industrial characteristics of the auto sector, especially in the context of Japan-Korea trade, and was heavily dependent on the Armington elasticity (structure) in the demand function of the GTAP CGE model. As a result, the domestic auto sector could lower prices and increase exports when joining the CPTPP. This paper attempts to precisely re-estimate the impact of joining the CPTPP on the auto sector in a way that corrects these problems by changing the CGE model and reflecting the major characteristics of the industry, with policy implications for the negotiation of CPTPP accession.

• 충청수학회지
• /
• 제17권1호
• /
• pp.47-56
• /
• 2004

The nonlinearity of a Boolean function f on $GF(2)^n$ is the minimum hamming distance between f and all affine functions on $GF(2)^n$ and it measures the ability of a cryptographic system using the functions to resist against being expressed as a set of linear equations. Finding out the exact value of the nonlinearity of given Boolean functions is not an easy problem therefore one wants to estimate the nonlinearity using extra information on given functions, or wants to find a lower bound or an upper bound on the nonlinearity. In this paper we extend the notion of auto-correlations of Boolean functions to vector Boolean functions and obtain upper bounds and a lower bound on the nonlinearity of vector Boolean functions in the context of their auto-correlations. Also we can describe avalanche characteristics of vector Boolean functions by examining the extended notion of auto-correlations.

• Journal of Korean Neurosurgical Society
• /
• 제61권3호
• /
• pp.393-401
• /
• 2018

The prognosis of brain tumors in children has improved for last a few decades. However, the prognosis remains dismal in patients with recurrent brain tumors. The outcome for infants and young children in whom the use of radiotherapy (RT) is very limited because of unacceptable long-term adverse effect of RT remains poor. The prognosis is also not satisfactory when a large residual tumor remains after surgery or when leptomeningeal seeding is present at diagnosis. In this context, a strategy using high-dose chemotherapy and autologous stem cell transplantation (HDCT/auto-SCT) has been explored to improve the prognosis of recurrent or high-risk brain tumors. This strategy is based on the hypothesis that chemotherapy dose escalation might result in improvement in survival rates. Recently, the efficacy of tandem HDCT/auto-SCT has been evaluated in further improving the outcome. This strategy is based on the hypothesis that further dose escalation might result in further improvement in survival rates. At present, the number of studies employing tandem HDCT/auto-SCT for brain tumors is limited. However, results of these pilot studies suggest that tandem HDCT/auto-SCT may further improve the outcome. In this review, we will summarize our single center experience with tandem HDCT/auto-SCT for recurrent or high-risk brain tumors.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제10권1호
• /
• pp.19-24
• /
• 2010

The rise of music resources has led to a parallel rise in the need to manage thousands of songs on user devices. So users have a tendency to build playlist for manage songs. However the manual selection of songs for creating playlist is a troublesome work. This paper proposes an auto playlist generation system considering user context of use and preferences. This system has two separated systems; 1) the mood and emotion classification system and 2) the music recommendation system. Firstly, users need to choose just one seed song for reflecting their context of use. Then system recommends candidate song list before the current song ends in order to fill up user playlist. User also can remove unsatisfied songs from the recommended song list to adapt the user preference model on the system for the next song list. The generated playlists show well defined mood and emotion of music and provide songs that the preference of the current user is reflected.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1282-1290
• /
• 2017

Since the mitigation of fear of crime significantly enhances the consumptions in a city, studies focusing on urban safety analysis have received much attention as means of revitalizing the local economy. In addition, with the development of computer vision and machine learning technologies, efficient and automated analysis methods have been developed. Previous studies have used global features to predict the safety of cities, yet this method has limited ability in accurately predicting abstract information such as safety assessments. Therefore we used a Convolutional Context Neural Network (CCNN) that considered "context" as a decision criterion to accurately predict safety of cities. CCNN model is constructed by combining a stacked auto encoder with a fully connected network to find the context and use it in the CNN model to predict the score. We analyzed the RMSE and correlation of SVR, Alexnet, and Sharing models to compare with the performance of CCNN model. Our results indicate that our model has much better RMSE and Pearson/Spearman correlation coefficient.

• 한국정보통신학회논문지
• /
• 제21권4호
• /
• pp.737-743
• /
• 2017

지능형 라이프로그 시스템은 언제(When), 어디서(Where), 누구와 함께(Who), 어떤 상황에서(What, How), 일어나는 정보, 즉 사용자의 일상에서 발생되는 시간, 인물, 장면, 연령대, 감정, 관계, 상태, 위치, 이동 경로 등의 다양한 상황정보들을 인식하여 태그를 달아 사용자의 일상생활을 기록하고 쉽고 빠르게 접근하도록 제공하는 것이다. 상황인식은 자동태킹(Auto-Tagging) 기술과 생체인식 기술인 얼굴인식을 이용해, 태그(Tag) 단위의 정보를 자동으로 생성하고 분류되어 상황정보 DB을 구축한다. 상황정보 DB에는 지리정보, 위치정보, 날씨정보, 감정정보 등을 포함하고 있다. 본 논문에서는 감정정보를 자동으로 기록하기 위해서 능동형태 모델 방법을 사용해 무표정과 웃는표정을 인식하는 어플리케이션을 개발하였다.

• 한국지능시스템학회논문지
• /
• 제12권3호
• /
• pp.261-268
• /
• 2002

This paper represents that auto tunings of membership functions and weights in the fuzzy neural networks are effectively performed by immune algorithm. A number of hybrid methods in fuzzy-neural networks are considered in the context of tuning of learning method, a general view is provided that they are the special cases of either the membership functions or the gain modification in the neural networks by genetic algorithms. On the other hand, since the immune network system possesses a self organizing and distributed memory, it is thus adaptive to its external environment and allows a PDP (parallel distributed processing) network to complete patterns against the environmental situation. Also, it can provide optimal solution. Simulation results reveal that immune algorithms are effective approaches to search for optimal or near optimal fuzzy rules and weights.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1990년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 26-27 Oct. 1990
• /
• pp.825-830
• /
• 1990

This paper reports the design procedure and testing result of a servo driven pan/tilt device which is capable of tracking arbitrary movement of a specified target object. In order to achieve real-time acquisition of feedback signal, a 2 degrees-of-freedom non-contact type displacement follower is used. The performance of the system is tested for different target velocities and control gains. The result of the research may provide an effective tool for visual transfer in the context of teleoperation.

• 디지털융복합연구
• /
• 제10권4호
• /
• pp.341-346
• /
• 2012

소셜 네트워크 서비스의 가장 대표적인 마이크로 블로깅 서비스를 효과적으로 제공하기 위해 사용자가 자신의 현재 상황정보를 간편하게 기록하고 그 정보를 바탕으로 다른 사람들과 네트워크를 형성하고 유지하도록 하는 것이 중요하다. 하지만 모바일 환경에서 사용자가 자신의 정보를 매번 모바일 디바이스를 통해 직접 입력하는 것은 매우 번거로운 작업이다. 본 논문에서는 획득된 사용자 컨텍스트를 이용하여 사용자의 현재 행동과 다음 목적지를 추론한 후, 자동으로 문장을 생성하여 블로깅을 해주는 컨텍스트 모델 기반 자동 블로깅 시스템을 제안한다. 컨텍스트 모델을 생성하기 위해 사용자의 행동 추론은 나이브 베이즈 분류기를 이용하고, 이동중인 사용자의 다음 목적지 추론은 시퀀스 매칭을 이용하였다. 생성된 컨텍스트 모델을 기반으로 5W1H 구조를 이용하여 상황에 적합한 문장을 생성하여 자동으로 블로깅하였다. 제안한 방법의 정확도를 평가한 결과 평균 88.73%의 정확도를 보여 자동 블로깅 서비스가 가능함을 보여주었다.