• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.4
• /
• pp.244-254
• /
• 2009

This paper describes a remote system to monitor the circadian behavioral patterns of elders who live alone. The proposed system was designed and implemented to provide more conveniently and reliably the required functionalities of a remote monitoring system for elders based on the development of first phase prototype[2]. The developed system is composed of an in-house sensing system and a server system. The in-house sensing system is a set of wireless sensor nodes which have pyroelectric infrared (PIR) sensor to detect a motion of elder. Each sensing node sends its detection signal to a home gateway via wireless link. The home gateway stores the received signals into a remote database. The server system is composed of a database server and a web server, which provides web-based monitoring system to caregivers (friends, family and social workers) for more cost effective intelligent care service. The improved second phase system can provide 'automatic diagnosis', 'going out detection', and enhanced user interface functionalities. We have evaluated the first and second phase monitoring systems from real field experiments of 3/4 months continuous operation with installation of 9/15 elders' houses, respectively. The experimental results show the promising possibilities to estimate the behavioral patterns and the current status of elder even though the simplicity of sensing capability.

• Journal of KIISE:Databases
• /
• v.28 no.4
• /
• pp.643-654
• /
• 2001

This paper discusses an effective processing of similarity search that supports time warping in large sequence database. Time warping enables finding sequences with similar patterns even when they are of different length, Previous methods fail to employ multi-dimensional indexes without false dismissal since the time warping distance does not satisfy the triangular inequality. They have to scan all the database, thus suffer from serious performance degradation in large database. Another method that hires the suffix tree also shows poor performance due to the large tree size. In this paper we propose a new novel method for similarity search that supports time warping Our primary goal is to innovate on search performance in large database without false dismissal. to attain this goal ,we devise a new distance function $D_{tw-Ib}$ consistently underestimates the time warping distance and also satisfies the triangular inequality, $D_{tw-Ib}$ uses a 4-tuple feature vector extracted from each sequence and is invariant to time warping, For efficient processing, we employ a distance function, We prove that our method does not incur false dismissal. To verify the superiority of our method, we perform extensive experiments . The results reveal that our method achieves significant speedup up to 43 times with real-world S&P 500 stock data and up to 720 times with very large synthetic data.

• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1261-1269
• /
• 1999

다층 퍼셉트론은 다양한 응용 분야에 성공적으로 적용되고 있는 대표적인 신경회로망 모델이다. 그러나 다층 퍼셉트론의 학습에서 나타나는 플라토에 기인한 느린 학습 속도와 지역 극소는 실제 응용문제에 적용함에 있어서 가장 큰 문제로 지적되어왔다. 이 문제를 해결하기 위해 여러 가지 다양한 학습알고리즘들이 개발되어 왔으나, 계산의 비효율성으로 인해 실제 문제에는 적용하기 힘든 예가 많은 등, 현재까지 만족할 만한 해결책은 제시되지 못하고 있다. 본 논문에서는 다층퍼셉트론의 베이시스 함수로 사용되는 시그모이드 함수를 보다 일반화된 형태로 정의하여 사용함으로써 학습에 있어서의 플라토를 완화하고, 지역극소에 빠지는 것을 줄이는 접근방법을 소개한다. 본 방법은 기존의 변형된 가중치 수정식을 사용한 학습 속도 향상의 방법들과는 다른 접근 방법을 택함으로써 기존의 방법들과 함께 사용하는 것이 가능하다는 특징을 갖고 있다. 제안하는 방법의 성능을 확인하기 위하여 간단한 패턴 인식 문제들에의 적용 실험 및 기존의 학습 속도 향상 방법을 함께 사용하여 시계열 예측 문제에 적용한 실험을 수행하였고, 그 결과로부터 제안안 방법의 효율성을 확인할 수 있었다. Abstract A multilayer perceptron is the most well-known neural network model which has been successfully applied to various fields of application. Its slow learning caused by plateau and local minima of gradient descent learning, however, have been pointed as the biggest problems in its practical use. To solve such a problem, a number of researches on learning algorithms have been conducted, but it can be said that none of satisfying solutions have been presented so far because the problems such as computational inefficiency have still been existed in these algorithms. In this paper, we propose a new learning approach to minimize the effect of plateau and reduce the possibility of getting trapped in local minima by generalizing the sigmoidal function which is used as the basis function of a multilayer perceptron. Adapting a new approach that differs from the conventional methods with revised updating equation, the proposed method can be used together with the existing methods to improve the learning performance. We conducted some experiments to test the proposed method on simple problems of pattern recognition and a problem of time series prediction, compared our results with the results of the existing methods, and confirmed that the proposed method is efficient enough to apply to the real problems.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.7
• /
• pp.1380-1386
• /
• 2007

This paper proposed an automatic recognition system of shipping container identifiers using fuzzy-based noise removal method and ART2-based self-organizing supervised learning algorithm. Generally, identifiers of a shipping container have a feature that the color of characters is blacker white. Considering such a feature, in a container image, all areas excepting areas with black or white colors are regarded as noises, and areas of identifiers and noises are discriminated by using a fuzzy-based noise detection method. Areas of identifiers are extracted by applying the edge detection by Sobel masking operation and the vertical and horizontal block extraction in turn to the noise-removed image. Extracted areas are binarized by using the iteration binarization algorithm, and individual identifiers are extracted by applying 8-directional contour tacking method. This paper proposed an ART2-based self-organizing supervised learning algorithm for the identifier recognition, which improves the performance of learning by applying generalized delta learning and Delta-bar-Delta algorithm. Experiments using real images of shipping containers showed that the proposed identifier extraction method and the ART2-based self-organizing supervised learning algorithm are more improved compared with the methods previously proposed.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.10
• /
• pp.79-89
• /
• 2019

While most research focuses on the k-nearest neighbors (kNN) queries in the database community, an important type of proximity queries called k-farthest neighbors (kFN) queries has not received much attention. This paper addresses the problem of finding the k-farthest neighbors in road networks. Given a positive integer k, a query object q, and a set of data points P, a kFN query returns k data objects farthest from the query object q. Little attention has been paid to processing kFN queries in road networks. The challenge of processing kFN queries in road networks is reducing the number of network distance computations, which is the most prominent difference between a road network and a Euclidean space. In this study, we propose an efficient algorithm called FANS for k-FArthest Neighbor Search in road networks. We present a shared computation strategy to avoid redundant computation of the distances between a query object and data objects. We also present effective pruning techniques based on the maximum distance from a query object to data segments. Finally, we demonstrate the efficiency and scalability of our proposed solution with extensive experiments using real-world roadmaps.

• Structural Engineering and Mechanics
• /
• v.50 no.5
• /
• pp.589-603
• /
• 2014

Expressions for determining the value of the impact force as reported in the literature and incorporated into code provisions are essentially quasi-static forces for emulating deflection. Quasi-static forces are not to be confused with contact force which is generated in the vicinity of the point of contact between the impactor and target, and contact force is responsible for damage featuring perforation and denting. The distinction between the two types of forces in the context of impact actions is not widely understood and few guidelines have been developed for their estimation. The value of the contact force can be many times higher than that of the quasi-static force and lasts for a matter of a few milli-seconds whereas the deflection of the target can evolve over a much longer time span. The stiffer the impactor the shorter the period of time to deliver the impulsive action onto the target and consequently the higher the peak value of the contact force. This phenomenon is not taken into account by any contemporary codified method of modelling impact actions which are mostly based on the considerations of momentum and energy principles. Computer software such as LS-DYNA has the capability of predicting contact force but the dynamic stiffness parameters of the impactor material which is required for input into the program has not been documented for debris materials. The alternative, direct, approach for an accurate evaluation of the damage potential of an impact scenario is by physical experimentation. However, it can be difficult to extrapolate observations from laboratory testings to behaviour in real scenarios when the underlying principles have not been established. Contact force is also difficult to measure. Thus, the amount of useful information that can be retrieved from isolated impact experiments to guide design and to quantify risk is very limited. In this paper, practical methods for estimating the amount of contact force that can be generated by the impact of a fallen debris object are introduced along with the governing principles. An experimental-calibration procedure forming part of the assessment procedure has also been verified.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.63-70
• /
• 2021

Generating a super-resolution meteological data by using a high-resolution deep neural network can provide precise research and useful real-life services. We propose a new technique of generating improved training data for super-resolution deep neural networks. To generate high-resolution meteorological data with domain specific knowledge, Lambert conformal conic projection and objective analysis were applied based on observation data and ERA5 reanalysis field data of specialized institutions. As a result, temperature and humidity analysis data based on domain specific knowledge showed improved RMSE by up to 42% and 46%, respectively. Next, a super-resolution generative adversarial network (SRGAN) which is one of the aritifial intelligence techniques was used to automate the manual data generation technique using damain specific techniques as described above. Experiments were conducted to generate high-resolution data with 1 km resolution from global model data with 10 km resolution. Finally, the results generated with SRGAN have a higher resoltuion than the global model input data, and showed a similar analysis pattern to the manually generated high-resolution analysis data, but also showed a smooth boundary.

• Journal of the Korean Geosynthetics Society
• /
• v.20 no.3
• /
• pp.1-10
• /
• 2021

The purpose of this study is to present the possibility a utilization of the tertiary mudstone in Pohang as road subsoil material through pilot experiments on the road embankment structure. This mudstone is an unconsolidated rock that is distributed in the soft rock sedimentary layer, the tertiary layer of the Cenozoic, and causes physical problems such as slaking, swelling, and reduced shear strength and chemical problem like acid drainage. In order to solve various complex problems, an laboratory mixing test was conducted, and the optimal mixing conditions of the tertiary mudstone (90%), composite slag (steel making 70%, blast furnace 30%), and neutralization and coating agent treatment were derived. In order to prove its utilization, a real-scale road embankment structure was constructed and tests were conducted for each section. The pre-processing section is stable due to the design of optimal mixing conditions, while in post-processing section, natural weathering proceeded rapidly, and structural problems were concerned. Since the effect of neutralizing and coating agents was confirmed in temporary-staking section, the neutralizing and coating agents can be applied during the temporary storage period.

• Journal of Korean Theatre Studies Association
• /
• no.52
• /
• pp.5-59
• /
• 2014

This paper investigates the development and aesthetics of Korean political theatre from its quickening period 1920s to democratization era 1990s. Political theatre before 90s developed an antithesis resistant movement toward Korean modern history that had been scattered with suppressing political circumstances such as colonial era and dictatorial government, the movement has powerful activity and social influences. Just like the 20 century political theatre had been quickened under the influence of Marxism at Russia and Germany in 1920s, Korea's political theatre began in socialism theatre movement form around the same time. Proletarian theatre groups had been founded in Japan and Korea, and developed into practical movement with organized connection. However, the political theatre movement in Japanese colonial era was an empty vessel makes great sound but not much accomplishments. Most performance had been canceled or disapproved by suppression or censorship of the Japanese Empire. The political theatre in liberation era was the left drama inherited from Proletarian theatre of the colonial era. Korean Theatre alliance took lead the theatrical world unfold activities based on theatre popularization theory such as 'culture activists' taking a jump up the line and 'independent theatre' peeping into production spot as well as the important event, Independence Movement Day Memorial tournament theatre. Since 1947, US army military government in Korea strongly oppressed the left performances to stop and theatrical movement was ended due to many left theatrical people defection to North Korea. The political theatre in 1960s to 70s the Park regime, developed in dramatically different ways according to orthodox group and group out of power. The political theatre of institutional system handled judgment on sterile people and had indirect political theatre from that took history material and allegory technique because of censorship. In political theatre out of institution, it started outdoor theatre that has modernized traditional performance style and established deep relationship with labor spot and culture movement organizations. Madangguek(Outdoor theatre) is 'Attentive political theatre', satirizing and offending the political and social inconsistencies such as the dictatorial government's oppression and unbalanced distribution, alienation of general people, and foreign powers' pillage sharply as well as laughing at the Establishment with negative characters. The political theatre in 1980s is divided into two categories; political theatre of institutional system and Madangguek. Institutional Political theatre mainly performed in Korea Theatre Festival and the theatre group 'Yeonwoo-Moudae' led political theatre as private theatre company. Madangguek developed into an outdoor theatrical for indoor theatre capturing postcolonial historical view. Yeonwoo-Moudae theatre company produced representative political plays at 80s such as The chronicles of Han's, Birds fly away too, and so on by combining freewheeling play spirit of Madangguek and epic theatre. Political theatre was all the rage since the age of democratization started in 1987 and political materials has been freed from ban. However, political theatre was slowly declined as real socialism was crumbling and postmodernism is becoming the spirit of the times. After 90s, there are no more plays of ideology and propaganda that aim at politicization of theatre. As the age rapidly entered into the age of deideology, political theatre discourse also changed greatly. The concept 'the political' became influential as a new political possibility that stands up to neoliberalism system in the evasion of politics. Rather than reenact political issues, it experiments new political theatre that involves something political by deconstructing and reassigning audience's political sense with provocative forms, staging others and drawing discussion about it.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.