• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.2
• /
• pp.97-106
• /
• 1999

This paper proposes the structure and the algorithm of the Fuzzy-Neural Controller(FNNC) which is able to adapt itself to unknown plant and the change of circumstances at the Fuzzy Logic Controller(FLC) with the Neural Network. This Learning Fuzzy Logic Controller is made up of Fuzzy Logic controller in charge of a main role and Neural Network of an adaptation in variable circumstances. This construct optimal fuzzy controller applied to the 2-area load frequency control of power system, and then it would examine fitness about parameter variation of plant or variation of circumstances. And it proposes the optimal Scale factor method wsint three preformance functions( E, , U) of system dynamics of load frequency control with error back-propagation learning algorithm. Applying the controller to the model of load frequency control, it is shown that the FNNC method has better rapidity for load disturbance, reduces load frequency maximum deviation and tie line power flow deviation and minimizes reaching and settling time compared to the Optimal Fuzzy Logic Controller(OFLC) and the Optimal Control for optimzation of performance index in past control techniques.

• Journal of Korea Game Society
• /
• v.15 no.3
• /
• pp.79-86
• /
• 2015

Effective feedback plays a key role in education environments to motivate the students and to maintain their flow. However, there usually lacks in providing effective feedbacks in education environments due to an excessive number of students and insufficient class times. The purpose of this paper is to analyze the educatees' response on game like feedback using App. In this paper, Class123 App developed and published by BravePops was used. During one semester, Class123 App had been used in engineering class and the educatees' responses on game like feedbacks using Class123 App are analyzed. Statistical analysis results show that the feedbacks using Class123 App made positive effects to motivate the students. Especially, the positive feedbacks were more effective than the negative feedbacks.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.981-984
• /
• 2014

개인화된 서술형 수학 문제(mathematics word problem)는 오랫동안 연구된 분야로 학생들의 학업 성취도와 수학에 대한 태도에 관심을 가져왔다. 본 연구에서는 2013년 도입된 스토리텔링 수학에 개인화된 콘텐츠를 접목하여 그 효과를 알아보고자 하였다. 초등학생 26명을 대상으로 하여 약 110분 동안 수업을 진행하였으며, 무게에 대한 새로운 개념을 배우는 데 그 목적을 두었다. 각각 13명씩 개인화 그룹과 비 개인화 그룹으로 나누어 수업을 진행하였다. 학업 성취도(Learning Achievement)에서는 사전 시험(pre-test) 점수가 너무 높아 두 그룹 간에 서로간의 유의한 차이점을 발견하지 못했다. 수학에 대한 태도 부분과 몰입도(Flow) 부분에서는 다소 개인화 그룹의 값이 높았지만, 통계적으로 유의한 정도는 차이는 아니었다. 하지만 정성적 분석에서는 차이가 있었다. 개인화 그룹(Personalized group)은 비 개인화 그룹(non-personalized group)에 비해 개인화(personalization)가 수업의 재미있는 요소로서 보다 중요한 작용을 했다고 느꼈다. 또한, 테스트나 측정(measure) 부분에서 생겼던 문제점을 개선하여 재 실험이 있을 시엔 유의미한 값을 나타낼 것으로 기대된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.10
• /
• pp.3668-3684
• /
• 2021

Video action recognition is widely used in video surveillance, behavior detection, human-computer interaction, medically assisted diagnosis and motion analysis. However, video action recognition can be disturbed by many factors, such as background, illumination and so on. Two-stream convolutional neural network uses the video spatial and temporal models to train separately, and performs fusion at the output end. The multi segment Two-Stream convolutional neural network model trains temporal and spatial information from the video to extract their feature and fuse them, then determine the category of video action. Google Xception model and the transfer learning is adopted in this paper, and the Xception model which trained on ImageNet is used as the initial weight. It greatly overcomes the problem of model underfitting caused by insufficient video behavior dataset, and it can effectively reduce the influence of various factors in the video. This way also greatly improves the accuracy and reduces the training time. What's more, to make up for the shortage of dataset, the kinetics400 dataset was used for pre-training, which greatly improved the accuracy of the model. In this applied research, through continuous efforts, the expected goal is basically achieved, and according to the study and research, the design of the original dual-flow model is improved.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.2
• /
• pp.45-52
• /
• 2019

BEMS(Building Energy Management Systems) have been applied to office buildings and collect relevant building energy data, e.g. temperatures, mass flow rates and energy consumptions of building mechanical systems and indoor spaces. The aforementioned measured data can be beneficially utilized for developing data-driven machine learning models which can be then used as part of MPC(Model Predictive Control) and/or optimal control strategies. In this study, the authors developed ANN(Artificial Neural Network) models of an AHU (Air Handling Unit) and a chiller for a real-life office building using BEMS data. Based on the ANN models, the authors developed optimal control strategies, e.g. daily operation schedule with regard to optimal start and stop of the AHU and the chiller (500 RT). It was found that due to the optimal start and stop of the AHU and the chiller, 4.5% and 16.4% of operation hours of the AHU and the chiller could be saved, compared to an existing operation.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• Journal of Korea Multimedia Society
• /
• v.24 no.1
• /
• pp.21-29
• /
• 2021

With the recent increase in the number of households raising pets, various engineering studies have been underway for pets. The final purpose of this study is to automatically generate situation-sensitive captions that can express implicit intentions based on the behavior and sound of cats by embedding the already mature behavioral detection technology of pets as basic element technology in the video capturing research. As a pilot project to this end, this paper proposes a high-level capturing system using optical-flow, RGB, and sound information of cat videos. That is, the proposed system uses video datasets collected in an actual breeding environment to extract feature vectors from the video and sound, then through hierarchical LSTM encoder and decoder, to identify the cat's behavior and its implicit intentions, and to perform learning to create context-sensitive captions. The performance of the proposed system was verified experimentally by utilizing video data collected in the environment where actual cats are raised.

• Electronics and Telecommunications Trends
• /
• v.36 no.2
• /
• pp.32-42
• /
• 2021

The rapid growth of deep-learning applications has invoked the R&D of artificial intelligence (AI) processors. A dedicated software framework such as a compiler and runtime APIs is required to achieve maximum processor performance. There are various compilers and frameworks for AI training and inference. In this study, we present the features and characteristics of AI compilers, training frameworks, and inference engines. In addition, we focus on the internals of compiler frameworks, which are based on either basic linear algebra subprograms or intermediate representation. For an in-depth insight, we present the compiler infrastructure, internal components, and operation flow of ETRI's "AI-Ware." The software framework's significant role is evidenced from the optimized neural processing unit code produced by the compiler after various optimization passes, such as scheduling, architecture-considering optimization, schedule selection, and power optimization. We conclude the study with thoughts about the future of state-of-the-art AI compilers.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.292-298
• /
• 2022

Coding education and algorithm education are essential in the era of the fourth industrial revolution. Text-oriented algorithm textbooks are perceived as difficult by students who are new to coding and algorithms. There is a need to develop educational content so that students can easily understand the principles of complex algorithms. This paper has implemented basic sorting algorithms as augmented reality contents for students who are new to algorithm education. To make it easier to understand the concept and principles of sorting algorithms, sorting data was expressed as a 3D box and the comparison of values according to the algorithms and the movement of values were produced as augmented reality contents in the form of 3D animations. In order to help with the understanding of sorting algorithms in C language, the change of variable values and the exchange of data were shown as animations according to the execution order of the code and the flow of the loop. Students can conveniently use contents through a smart phone without special equipment by being produced in a marker-based manner. Interest and immersion, as well as understanding of classes of sorting algorithms can be increased through educational augmented reality-based educational contents.

• International journal of advanced smart convergence
• /
• v.12 no.3
• /
• pp.104-108
• /
• 2023

This paper deals with research on innovative systems using Python-based artificial intelligence technology in the field of plant growth monitoring. The importance of monitoring and analyzing the health status and growth environment of plants in real time contributes to improving the efficiency and quality of crop production. This paper proposes a method of processing and analyzing plant image data using computer vision and deep learning technologies. The system was implemented using Python language and the main deep learning framework, TensorFlow, PyTorch. A camera system that monitors plants in real time acquires image data and provides it as input to a deep neural network model. This model was used to determine the growth state of plants, the presence of pests, and nutritional status. The proposed system provides users with information on plant state changes in real time by providing monitoring results in the form of visual or notification. In addition, it is also used to predict future growth conditions or anomalies by building data analysis and prediction models based on the collected data. This paper is about the design and implementation of Python-based plant growth monitoring systems, data processing and analysis methods, and is expected to contribute to important research areas for improving plant production efficiency and reducing resource consumption.