• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.565-568
• /
• 2009

This paper proposes an monitoring system of elder and feeble person's motion activity using an object's motion activity data. The proposed system used wireless 3-axis sensor module, product by Freescale(Wireless Sensing Triple Axis Reference Design Board (ZSTAR)). We distribute sensing data into three classes using Neural Network System SVM. We find performance of proposed system that simulate some case about walk, past walk, fallen. Classify result data and graph of sensing data present succes rate 80%.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06a
• /
• pp.10-12
• /
• 2012

얼굴인식은 보안 등 다수의 응용분야에서 중요하게 이용되는데, 얼굴인식을 위한 학습은 많은 계산시간이 소요되기 때문에 신속한 학습이 필요한 경우 가속화가 필요하다. 한편, 그래픽스 프로세서 유닛(GPU)은 대용량 정보처리를 빠르게 수행할 수 있어 최근 폭넓은 분야에서 널리 이용되고 있다. 본 논문에서는 주성분 기반의 얼굴인식 알고리즘을 GPU 에서 병렬 수행하여 가속하는 기법을 제안하였다. 주성분 기반의 얼굴인식 각각의 과정들의 병렬성을 분석하여 가속화 이득을 최대하였고, C/OpenCV[2]로 구현된 순차적인 버전[3]과 비교했을 때, 전체 학습시스템에서 최대 약 40 배의 성능이득을 얻었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.428-430
• /
• 2018

본 논문에서는 독감확산 예측을 위한 웨어러블 센서를 이용한 기침 감지 모델을 제안한다. 서로 상이한 기침 신체데이터를 사용하고 기침 감지 알고리즘의 구현없이 기계가 학습하는 방식인 멀티모달 DNN을 이용하여 설계하였다. 또한 웨어러블 센서를 통해 실생활의 기침 오디오 데이터와 기침 3축 가속도 데이터를 수집하였고, 두 개의 데이터중 하나의 데이터만으로도 감지를 위한 학습이 가능토록하기 위해 각각 MFCC와 FFT를 이용하여 특징 벡터를 추출하는 방법을 이용하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.7
• /
• pp.1759-1769
• /
• 1997

The Error BackPropagation (EBP) algorithm for multi-layered neural networks is widely used in various areas such as associative memory, speech recognition, pattern recognition and robotics, etc. Nevertheless, many researchers have continuously published papers about improvements over the original EBP algorithm. The main reason for this research activity is that EBP is exceeding slow when the number of neurons and the size of training set is large. In this study, we developed new learning speed acceleration methods using variable learning rate, variable momentum rate and variable slope for the sigmoid function. During the learning process, these parameters should be adjusted continuously according to the total error of network, and it has been shown that these methods significantly reduced learning time over the original EBP. In order to show the efficiency of the proposed methods, first we have used binary data which are made by random number generator and showed the vast improvements in terms of epoch. Also, we have applied our methods to the binary-valued Monk's data, 4, 5, 6, 7-bit parity checker and real-valued Iris data which are famous benchmark training sets for machine learning.

• The Journal of Society for e-Business Studies
• /
• v.22 no.4
• /
• pp.159-173
• /
• 2017

In order to provide technical financial support to small and medium-sized venture companies based on technology, the government implemented the TCB evaluation, which is a kind of technology rating evaluation, from the Kibo and a qualified private TCB. In this paper, we briefly review the current state of TCB evaluation and available indicators related to technology evaluation accumulated in the Korea Credit Information Services (TDB), and then use indicators that have a significant effect on the technology rating score. Multiple regression techniques will be explored. And the relative importance and classification accuracy of the indicators were calculated by applying the key indicators as independent features applied to the generalized boosting model, which is a representative machine learning classifier, as the class influence and the fitness of each model. As a result of the analysis, it was analyzed that the relative importance between the two models was not significantly different. However, GBM model had more weight on the InnoBiz certification, R&D department, patent registration and venture confirmation indicators than regression model.

• Journal of Educational Research in Mathematics
• /
• v.16 no.1
• /
• pp.59-78
• /
• 2006

As a theoretical background for this research, the literatures which focus on the rationale of teaching and learning of connecting with mathematics and science in calculus were investigated. And teaching and learning material of connecting with mathematics and science in calculus was developed. And then, based on the case study using this material, the research questions were analyzed in depth. Students could understand mean-velocity, instant-velocity, and acceleration in the experimenting process of constant acceleration movement. Also Students could understand fundamental ideas that instant-velocity means slope of the tangent line at one point on the time-displacement graph and rate of distance change means rate of area change under a time-velocity graph.

• Journal of Korean Society of Steel Construction
• /
• v.11 no.1 s.38
• /
• pp.79-88
• /
• 1999

Damages of a steel box girder bridge are detected using neural networks. Damage detection using neural networks has increasing momentum in structural engineering. It is a new effort to overcome the limitations of the conventional analytical approaches and applied to the damage detection of a steel box-girder bridge. Data sets for training neural networks are obtained from the acceleration response of the bridge under moving load. Finite element model is first defined and damages of 5, 10, 15 and 20% are assumed in the model. Not only the trained damages but untrained damages are detected in the assessment stage. The untrained damages can be detected with acceptable errors. Because the number of damaged locations are limited to a few parts, more researches are needed to put this technique into practice.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.505-510
• /
• 2005

The RPO(randomized policy optimizer) algorithm, which utilizes probabilistic policy for the action selection, is a recently developed tool in the area of reinforcement learning, and has been shown to be very successful in several application problems. In this paper, we propose a modified RPO algorithm, whose critic network is adapted via RLS(Recursive Least Square) algorithm. In order to illustrate the applicability of the modified RPO method, we applied the modified algorithm to Kimura's robot and observed very good performance. We also developed a MATLAB-based animation program, by which the effectiveness of the training algorithms on the acceleration or the robot movement were observed.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.173-173
• /
• 2017

드론의 시장규모가 커짐에 따라 초창기 군사 목적에서 현재 민간부문으로 확대되고 있다. 현재 드론은 실외에서 사용될 목적으로 제작된 것이 많으나 실내에서도 드론의 활용 여부가 증가할 것으로 예상된다. 본 연구에서는 실외에서만 사용 가능한 GPS를 대신하여 영상 촬영으로 획득한 이미지를 CNN으로 학습을 시켜 자율고도제어비행을 하도록 한다. 첫 번째로 수동 조작하는 드론에 IMU센서를 부착하여 획득한 고도 데이터를 표로 제시함으로써 GPS를 사용하지 않는 드론의 실내주행에서 일정한 고도 유지는 다소 무리가 있음을 보여준다. 두 번째로 드론의 수동 조작은 일정하지 않은 고도 때문에 CNN의 학습할 영상 획득이 어렵다. 일정한 고도의 영상 획득을 위한 실험용 높이 조절 Base를 제작하여 고도별 영상을 획득한다. 획득한 영상을 통해 얻은 이미지를 CNN 학습을 시킨 후, 학습에 사용되지 않은 이미지를 사용하여 고도 판별을 확인한다. 대조군으로 실내장소를 바꾸어 미리 학습된 CNN으로 고도 판별을 확인한다. 학습에 사용된 이미지의 환경(생명공학관)과 대조군(제 2 공학관)이 촬영된 장소의 환경요소의 차이로 오차가 발생한다. 오차는 실내 장소의 총 높이의 차이 및 서로 상이한 천장 구조물에 따른 것으로 사료되며 Data crop을 통해 획득한 이미지의 천정 부분을 제거하여 노이즈를 줄여 고도 판별의 정확도를 높일 수 있을 것으로 예상한다. 세 번째, CNN으로 학습을 통해 Model을 도출하여 자율 고도 제어 프로세스를 제시한다. 그리고 해당 프로세스를 이용한 자율고도제어 주행과 수동조작을 통한 주행에서의 Z축 가속도 데이터의 표준편차를 비교하여 본 연구의 실효성을 보여준다

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.635-642
• /
• 2003

Reinforcement learning got successful results in a lot of applications such as control and scheduling. Various function approximation methods have been studied in order to improve the learning speed and to solve the shortage of storage in the standard reinforcement learning algorithm of Q-Learning. Most function approximation methods remove some special quality of reinforcement learning and need prior knowledge and preprocessing. Fuzzy Q-Learning needs preprocessing to define fuzzy variables and Local Weighted Regression uses training examples. In this paper, we propose a function approximation method, Fuzzy Q-Map that is based on on-line fuzzy clustering. Fuzzy Q-Map classifies a query state and predicts a suitable action according to the membership degree. We applied the Fuzzy Q-Map, CMAC and LWR to the mountain car problem. Fuzzy Q-Map reached the optimal prediction rate faster than CMAC and the lower prediction rate was seen than LWR that uses training example.