• 방송공학회논문지
• /
• 제19권2호
• /
• pp.166-173
• /
• 2014

최근 다양한 방송 및 영상 분야에서 사람의 행동을 인식하여는 연구들이 많이 이루어지고 있다. 영상은 다양한 형태를 가질 수 있기 때문에 제약된 환경에서 유용한 템플릿 방법들보다 특징점에 기반한 연구들이 실제 사용자 환경에서 더욱 관심을 받고 있다. 특징점 기반의 연구들은 영상에서 움직임이 발생하는 지점들을 찾아내어 이를 3차원 패치들로 생성한다. 이를 이용하여 영상의 움직임을 히스토그램에 기반한 descriptor(서술자)로 표현하고 학습기반의 판별기로 최종적으로 영상내에 존재하는 행동들을 인식하였다. 그러나 단일 판별기로는 다양한 행동을 인식하기에 어려움이 있다. 따라서 이러한 문제를 개선하기 위하여 최근에 다중 판별기를 활용한 연구들이 영상 판별 및 물체 검출 영역에서 사용되고 있다. 따라서 본 논문에서는 행동 인식을 위하여 support vector machine과 sparse representation을 이용한 decision-level fusion 방법을 제안하고자 한다. 제안된 논문의 방법은 영상에서 특징점 기반의 descriptor를 추출하고 이를 각각의 판별기를 통하여 판별 결과들을 획득한다. 이 후 학습단계에서 획득된 가중치를 활용하여 각 결과들을 융합하여 최종 결과를 도출하였다. 본 논문에 실험에서 제안된 방법은 기존의 융합 방법보다 높은 행동 인식 성능을 보여 주었다.

• 한국경영과학회지
• /
• 제22권3호
• /
• pp.37-58
• /
• 1997

An AI approach with tabu search is designed to solve multi-level knapsack problems. The approach performs intelligent actions with memories of historic data and learning effect. These action are developed ont only by observing the attributes of the optimal solution, the solution space, and its corresponding path to the optimal, but also by applying human intelligence, experience, and intuition with respect to the search strategies. The approach intensifies, or diversifies the search process appropriately in time and space. In order to create a good neighborhood structure, this approach uses two powerful choice rules that emphasize the impact of candidate variables on the current solution with respect to their profit contribution. "Pseudo moves", similar to "aspirations", support these choice rules during the evaluation process. For the purpose of visiting as many relevant points as possible, strategic oscillation between feasible and infeasible solutions around the boundary is applied. To avoid redundant moves, short-term (tabu-lists), intemediate-term (cycle-detection), and long-term (recording frequency and significant solutions for diversfication) memories are used. Test results show that among the 45 generated problems (these problems pose significant or insurmountable challenges to exact methods) the approach produces the optimal solutions in 39 cases.lutions in 39 cases.

• International Journal of Contents
• /
• 제12권1호
• /
• pp.25-43
• /
• 2016

An agent-centric event planning method is proposed for providing pedagogical experiences in an immersed environment. Two-level planning is required at in a macro-level (i.e., inter-event level) and an intra-event level to provide realistic experiences with the objective of learning declarative knowledge. The inter-event (horizontal) planning is based on search, while intra-event (vertical) planning is based on hierarchical decomposition. The horizontal search is dictated by several realistic types of association between events besides the conventional causality. The resulting schematic plan is further augmented by conditions associated with those agents cast into the roles of the events identified in the plan. Rather than following a main story plot, all the events potentially relevant to accomplishing an initial goal are derived in the final result of our planning. These derived events may progress concurrently or digress toward a new main goal replacing the current goal or event, and the plan could be merged or fragmented according to their respective lead agents' intentions and other conditions. The macro-level coherence across interconnected events is established via their common background world existing a priori. As the pivotal source of event concurrency and intricacy, agents are modeled to not only be autonomous but also independent, i.e., entities with their own beliefs and goals (and subsequent plans) in their respective parts of the world. Additional problems our method addresses for augmenting pedagogical experiences include casting of agents into roles based on their availability, subcontracting of subsidiary events, and failure of multi-agent event entailing fragmentation of a plan. The described planning method was demonstrated by monitoring implementation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1319-1323
• /
• 2004

In this paper, a problem of Thai sign language recognition using a neural network is considered. The paper addresses the problem in classifying certain signs conveying quantitative meaning, e.g., large or small. By treating those signs corresponding to different quantities as derived from different classes, the recognition error rate of the standard multi-layer Perceptron increases if the precision in recognizing different quantities is increased. This is due the fact that, to increase the quantitative recognition precision of those signs, the number of (increasingly similar) classes must also be increased. This leads to an increase in false classification. The problem is due to misinterpreting the amount of quantity the quantitative signs convey. In this paper, instead of treating those signs conveying quantitative attribute of the same quantity type (such as 'size' or 'amount') as derived from different classes, here they are considered instances of the same class. Those signs of the same quantity type are then further divided into different subclasses according to the level of quantity each sign is associated with. By using this two-level classification, false classification among main gesture classes is made independent to the level of precision needed in recognizing different quantitative levels. Moreover, precision of quantitative level classification can be made higher during the recognition phase, as compared to that used in the training phase. A standard multi-layer Perceptron with a back propagation learning algorithm was adapted in the study to implement this two-level classification of quantitative gesture signs. Experimental results obtained using an electronic glove measurement of hand postures are included.

• 한국멀티미디어학회논문지
• /
• 제5권3호
• /
• pp.290-297
• /
• 2002

웹의 등장은 멀티미디어 기술과 컴퓨터 통신 기술 개발의 가속화 및 이를 응용한 컨텐츠 개발에 촉진제 역할을 하게 되었다. 최근에는 교수-학습 활동에서의 새로운 형태인 웹을 기반으로 한 교육(WBI : Web-Based Instruction)이라는 교수 모형이 제시되기에 이르렀다. 또한, 개별 학습자의 학습 수준을 고려한 학습 및 평가 방식이 요구되고 있으며, 그에 따라 웹 기반 교육 시스템에 효율적이고 자동화된 교육 에이전트의 필요성이 인식되고 있다. 그러나 현재 연구되고 있는 많은 교육 시스템들은 학습자 성향에 맞는 학습 과정을 적절히 서비스해 주지 못할 뿐 아니라 지속적인 피드백과 학습자가 학습 과정에 따라 학습을 진행함에 있어서 취약한 부분을 재학습 할 수 있도록 도와주는 서비스를 원활히 제공하지 못하고 있다. 본 논문에서는 취약성 분석 알고리즘을 이용한 학습자 중심의 코스 스케줄링 멀티 모듈 시스템의 설계를 제안한다. 제안한 시스템은 먼저 학습자의 학습을 지속적으로 모니터링하고 평가하여 개인 학습자의 학습 성취도를 계산하며, 이 성취도 계산을 통해 나온 단원별 취약성을 에이전트의 스케줄에 적용하여 학습자에게 취약한 과목을 재학습 할 수 있는 학습 환경을 제공하고, 학습자는 이러한 학습 환경에 따라 반복된 학습을 통하여 완전학습을 수행하게 된다.

• 영어어문교육
• /
• 제11권4호
• /
• pp.365-389
• /
• 2005

The purpose of this study is to present the task model that is the most effective in English language methodology based on the investigation of task-based performance in Korean EFL classroom practice. The subjects were 538 high school students and 126 high school teachers, each of whom had common experiences using the materials of task-based activities for more than one year. To analyze the data, the program SPSS WIN 11.0 including frequency distribution and chi-square analysis was used. The results of the questionnaire analysis showed that both teachers and students had a comparatively high level of satisfaction in task rationale, but that they had some mixed responses in the fields of input data, settings, and activity types. To conclude, a few suggestions are made to provide some meaningful considerations for the EFL teachers and material developers: a) task goals and rationale that encourage the learner's positive motivation; b) authenticity of input data based on the real-world context; c) collaborative learning environment that enhances communicative interaction; d) proportional representation of the creative problem-solving activities related to discussions and decision-making processes; e) systematic introduction of integrated language skills. It also suggests that the multi-lateral task model, which has some positive assets compared to previous task models, be newly introduced and applied to the second language learning classrooms.

• 정보보호학회논문지
• /
• 제30권4호
• /
• pp.527-536
• /
• 2020

정보보호용 디바이스의 부채널 정보인 소비 전력 파형을 이용하면 내장된 비밀 키 뿐만 아니라 동작 명령어를 복구할 수 있음이 밝혀졌다. 최근에는 MLP 등과 같은 딥러닝 모델을 이용한 프로파일링 기반의 부채널 공격들이 연구되고 있다. 본 논문에서는 마이크로 컨트롤러 AVR XMEGA128-D4가 사용하는 명령어에 대한 역어셈블러를 구현하였다. 명령어에 대한 템플릿 파형을 수집하고 전처리하는 과정을 자동화하였으며 CNN 딥러닝 모델을 사용하여 명령-코드를 분류하였다. 실험 결과, 전체 명령어는 약 87.5%의 정확도로, 사용 빈도가 높은 주요 명령어는 99.6%의 정확도로 분류될 수 있음을 확인하였다.

• Journal of Information Processing Systems
• /
• 제14권5호
• /
• pp.1114-1135
• /
• 2018

A multiple classification system based on a new boosting technique has been approached utilizing different biometric traits, that is, color face, iris and eye along with fingerprints of right and left hands, handwriting, palm-print, gait (silhouettes) and wrist-vein for person authentication. The images of different biometric traits were taken from different standard databases such as FEI, UTIRIS, CASIA, IAM and CIE. This system is comprised of three different super-classifiers to individually perform person identification. The individual classifiers corresponding to each super-classifier in their turn identify different biometric features and their conclusions are integrated together in their respective super-classifiers. The decisions from individual super-classifiers are integrated together through a mega-super-classifier to perform the final conclusion using programming based boosting. The mega-super-classifier system using different super-classifiers in a compact form is more reliable than single classifier or even single super-classifier system. The system has been evaluated with accuracy, precision, recall and F-score metrics through holdout method and confusion matrix for each of the single classifiers, super-classifiers and finally the mega-super-classifier. The different performance evaluations are appreciable. Also the learning and the recognition time is fairly reasonable. Thereby making the system is efficient and effective.

• International Journal of Computer Science & Network Security
• /
• 제23권1호
• /
• pp.1-10
• /
• 2023

Open surface water body extraction is gaining popularity in recent years due to its versatile applications. Multiple techniques are used for water detection based on applications. Different applications of Radar as LADAR, Ground-penetrating, synthetic aperture, and sounding radars are used to detect water. Shortwave infrared, thermal, optical, and multi-spectral sensors are widely used to detect water bodies. A stereo camera is another way to detect water and different methods are applied to the images of stereo cameras such as deep learning, machine learning, polarization, color variations, and descriptors are used to segment water and no water areas. The Satellite is also used at a high level to get water imagery and the captured imagery is processed using various methods such as features extraction, thresholding, entropy-based, and machine learning to find water on the surface. In this paper, we have summarized all the available methods to detect water areas. The main focus of this survey is on water detection especially in small patches or in small areas. The second aim of this survey is to detect water hazards for unmanned vehicles and off-sure navigation.

• 대한임베디드공학회논문지
• /
• 제4권2호
• /
• pp.55-62
• /
• 2009

The power consumption of a high-end microprocessor increases very rapidly. High power consumption will lead to a rapid increase in the chip temperature as well. If the temperature reaches beyond a certain level, chip operation becomes either slow or unreliable. Therefore various approaches for Dynamic Thermal Management (DTM) have been proposed. In this paper, we propose a learning based temperature prediction scheme for a multi-core system. In this approach, from repeatedly executing an application, we learn the thermal patterns of the chip, and we control the temperature in advance through DTM. When the predicted temperature may go beyond a threshold value, we reduce the temperature by decreasing the operation frequencies of the corresponding core. We implement our temperature prediction on an Intel's Quad-Core system which has integrated digital thermal sensors. A Dynamic Frequency System (DFS) technique is implemented to have four frequency steps on a Linux kernel. We carried out experiments using Phoronix Test Suite benchmarks for Linux. The peak temperature has been reduced by on average $5^{\circ}C{\sim}7^{\circ}C$. The overall average temperature reduced from $72^{\circ}C$ to $65^{\circ}C$.