• Journal of Biosystems Engineering
• /
• v.15 no.4
• /
• pp.319-327
• /
• 1990

시스템의 적응 제어함수를 산출하는 네트워크인 소뇌모델 선형조합 회로망을 이용한 학습제어 기법은 시스템에 영향을 주는 제어인자들의 불확실성 및 모델링의 결여에도 불구하고 오히려 안정한 실시간 제어의 구현을 가능하게 함으로써 대단한 관심을 불러 일으켜 왔다. 그러나, 센서로부터의 정보처리와 인식 그리고 복잡한 비선형 시스템의 제어에 적용하기에는 회로망 자체의 내재적 문제점들이 여전히 남아있다. 소뇌모델 선형조합 회로망을 기지 또는 미지의 시스템 모델에 효과적으로 적용하기 위해서는 네트워크에 영향을 주는 제어인자가 시스템에 미치는 영향을 분석하는 것이 필수적이다. 분할 블럭의 크기, 학습이득, 입력편이 그리고 입력변수들의 영역과 같은 네트 제어인자들은 시스템의 학습 능률 및 소요 기억용량의 크기에 중대한 영향을 미침에도 불구하고 충분히 조사되지 못한 실태이다. 물론 이들 제어인자들의 결정에는 학습 대상이 되는 시스템 함수의 형태와 적용 학습 알고리즘이 반드시 고려되어야 한다. 본 논문에서는 학습 능률성에 미치는 이들 제어인자들의 상호영향도를 저자가 제안하였던 기본 학습 알고리즘에 의거하여 조사하였다. 분석적인 방법만으로 이러한 상호영향성을 조사하기는 매우 힘들거나 거의 불가능하다고 보아지기 때문에 학습 대상함수를 먼저 규정하여 다양한 컴퓨터 모의시험을 수행하였고 그 결과를 분석하였다. 컴퓨터 모의시험의 결과에 의하여 회로망의 시스템 적용시 고려할 설계 지침을 제시하였다.

• Annual Conference of KIPS
• /
• 2003.05a
• /
• pp.255-258
• /
• 2003

기존의 동적 문제 시스템은 사용자의 학습 수준을 획일적으로 평하여 사용자의 학습 수준 상태를 정확하게 분석하지 못하였고, 이에 다라 사용자에게 올바른 학습상태를 전달하여 주지 못하였다. 본 논문에서는 이러한 획일적인 평가에서 탈피하여 사용자의 학습 수준을 효율적으로 평가하여 문제를 출제하고 채점하는 시스템을 제안한다.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of KIISE:Software and Applications
• /
• v.34 no.1
• /
• pp.11-17
• /
• 2007

One of the popular methods used for pattern classification is the MBR (Memory-Based Reasoning) algorithm. Since it simply computes distances between a test pattern and training patterns or hyperplanes stored in memory, and then assigns the class of the nearest training pattern, it cannot explain how the classification result is obtained. In order to overcome this problem, we propose an incremental teaming algorithm based on RPA (Recursive Partition Averaging) to extract IF-THEN rules that describe regularities inherent in training patterns. But rules generated by RPA eventually show an overfitting phenomenon, because they depend too strongly on the details of given training patterns. Also RPA produces more number of rules than necessary, due to over-partitioning of the pattern space. Consequently, we present the IREA (Incremental Rule Extraction Algorithm) that overcomes overfitting problem by removing useless conditions from rules and reduces the number of rules at the same time. We verify the performance of proposed algorithm using benchmark data sets from UCI Machine Learning Repository.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.175-178
• /
• 2005

스마트 홈과 같은 유비쿼터스 환경은 다양한 센서 및 제어 네트워크가 밀집되어 있는 복잡한 시스템이다. 본 논문에서는 이러한 환경하에서 복잡한 인터페이스의 사용에 대한 사용자의 인지 부담(cognitive load)를 줄이고 개인화된(personalized) 서비스를 자율적으로 제공하기 위한 사용자 행동 패턴 선호도 학습 기법을 제안한다. 이를 위해 지식 발견(Knowledge Discovery)을 위한 평생 학습(life-long learning)의 관점에서 퍼지 귀납(Fuzzy Inductive)학습 방법론을 제안하며, 이것은 수치 데이터로부터 입력 공간에 대한 효율적인 퍼지 분할(fuzzy partition)을 얻어내고 일관성있는(consisitent) 퍼지 상관 룰(fuzzy association rule)을 얻어내도록 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.805-812
• /
• 2005

Smart home is one of the ubiquitous environment platforms with various complex sensor-and-control network. In this paper, a now learning methodology for learning user's behavior preference pattern is proposed in the sense of reductive user's cognitive load to access complex interfaces and providing personalized services. We propose a fuzzy inductive learning methodology based on life-long learning paradigm for knowledge discovery, which tries to construct efficient fuzzy partition for each input space and to extract fuzzy association rules from the numerical data pattern.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.1
• /
• pp.32-40
• /
• 2008

Fuzzy rules are very useful and efficient to describe classification rules especially when the attribute values are continuous and fuzzy in nature. However, it is generally difficult to determine membership functions for generating efficient fuzzy classification rules. In this paper, we propose a method of automatic generation of efficient fuzzy classification rules using evolutionary algorithm. In our method we generate a set of initial membership functions for evolutionary algorithm by supervised clustering the training data set and we evolve the set of initial membership functions in order to generate fuzzy classification rules taking into consideration both classification accuracy and rule comprehensibility. To reduce time to evaluate an individual we also propose an evolutionary algorithm with data partition evaluation in which the training data set is partitioned into a number of subsets and individuals are evaluated using a randomly selected subset of data at a time instead of the whole training data set. We experimented our algorithm with the UCI learning data sets, the experiment results showed that our method was more efficient at average compared with the existing algorithms. For the evolutionary algorithm with data partition evaluation, we experimented with our method over the intrusion detection data of KDD'99 Cup, and confirmed that evaluation time was reduced by about 70%. Compared with the KDD'99 Cup winner, the accuracy was increased by 1.54% while the cost was reduced by 20.8%.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1314-1321
• /
• 2010

In this paper, we study about the vehicle detection algorithm which is in the process of travelling from the road. An input image is segmented by means of split and merge algorithm. And two largest segmented regions are removed for reducing search region and speed up processing time. In order to detect the back side of the front vehicle considers a vertical/horizontal component, uses an integral image with to apply Haar-like methods which are the possibility of shortening a calculation time, classified with SVM. The simulation result of the method which is proposed appeared highly.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.227-233
• /
• 2019

Recently, along with the recent development of deep learning technique, neural networks are achieving success in computer vision filed. Convolutional neural network have shown outstanding performance in not only for a simple image classification task, but also for tasks with high difficulty such as object segmentation and detection. However many such deep learning models are based on supervised-learning, which requires more annotation labels than image-level label. Especially image semantic segmentation model requires pixel-level annotations for training, which is very. To solve these problems, this paper proposes a weakly-supervised semantic segmentation method which requires only image level label to train network. Existing weakly-supervised learning methods have limitations in detecting only specific area of object. In this paper, on the other hand, we use multi-classifier deep learning architecture so that our model recognizes more different parts of objects. The proposed method is evaluated using VOC 2012 validation dataset.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.579-580
• /
• 2023

본 논문에서는 동물의 십자인대 질환의 진단지표인 경골고원각도(TPA)를 자동으로 측정하는 딥러닝 소프트웨어 기법을 제안한다. 동물 X-ray 영상에서 나타나는 피사체의 위치와 형태에 대한 다양한 변이는 TPA(Tibial Plateau Angle) 지표 산출에 필요한 특징점 검출과정에서 학습 효율을 현저하게 저하시킨다. 이에 본 연구에서는 YOLO(You Only Look Once) 기반 모델을 사용하여 일차적으로 경골영역의 분할 단계를 수행하고, 이어서 경골 상단부의 과간융기와 복사뼈의 중심점을 찾는 과정을 Resnet 기반의 특징점 추출 모듈로서 구현함으로써 학습의 효율과 지표 검출의 정확도를 향상시켰다. 총 201 개의 실제 X-ray 영상을 사용하여 학습 속도와 영역 분할 및 특징점 추출의 정확도 측면을 고려함으로 제안된 이론의 타당성을 실험적으로 평가하였다.