• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.3004-3006
• /
• 2000

In this paper, we propose an optimal design method of Fuzzy-Neural Networks model with parallel structure for complex and nonlinear systems. The proposed model is consists of a multiple number of FNN connected in parallel. The proposed FNNs with parallel structure is based on Yamakawa's FNN and it uses simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rules. We use a HCM clustering and GAs to identify the structure and the parameters of the proposed model. Also, a performance index with a weighting factor is presented to achieve a sound balance between approximation and generalization abilities of the model. To evaluate the performance of the proposed model. we use the time series data for gas furnace and the numerical data of nonlinear function.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.654-656
• /
• 1998

This paper presents an optimal identification method of nonlinear and complex system that is based on fuzzy-neural network(FNN). The FNN used simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM Algorithm to find initial parameters of membership function. And then to obtain optimal parameters, we use the genetic algorithm. Genetic algorithm is a random search algorithm which can find the global optimum without converging to local optimum. The parameters such as membership functions, learning rates and momentum coefficients are easily adjusted using the genetic algorithms. Also, the performance index with weighted value is introduced to achieve a meaningful balance between approximation and generalization abilities of the model. To evaluate the performance of the FNN, we use the time series data for 9as furnace and the sewage treatment process.

• The Journal of Sustainable Design and Educational Environment Research
• /
• v.11 no.3
• /
• pp.46-54
• /
• 2012

The accuracy of cost estimation at an early stage in school building project is one of the critical factors for successful completion. So various of techniques are developed to predict the construction cost accurately and expeditely. Among the techniques, Support Vector Machine(SVM) has an excellent ability for generalization performance. Therefore, the purpose of this study is to construct the prediction model for construction cost of educational building project using support vector machine technique. And to verify the accuracy of prediction model for construction cost. The performance data used in this study are 217 school building project cost which have been completed from 2004 to 2007 in Gyeonggi-Do, Korea. The result shows that average error rate was 7.48% for SVM prediction model. So using SVM model on predicting construction cost of educational building project will be a considerably effective way at the early project stage.

• Proceedings of the KIPE Conference
• /
• 2019.07a
• /
• pp.459-460
• /
• 2019

The performance dynamics of battery is very sensitive to operating conditions (i.e temperature, load current, and state of charge). A model developed based on certain conditions may perform well under the similar conditions but can not accurately predict the performance for changing conditions. Thus, a generalized model is needed which can accurately emulate the battery dynamic behavior under all conditions. In addition, the components of the model should relate to the physicochemical processes that occur inside the battery. Electrochemical impedance curve shows better visible reflection of the processes inside battery as compared to voltage curve. The model trained for parameterization using neural network has better generalization than simple curve fitting. Thus, this study proposes recurrent neural network based parameterization of the Lithium ion battery model followed by impedance based identification.

• Journal of Korean Elementary Science Education
• /
• v.34 no.4
• /
• pp.502-518
• /
• 2015

In order to find out Elementary pre-service teachers' reflective thinking and experiment performance ability related with Photosynthesis Experiment in the Korea Elementary School Science Textbook, the research is conducted targeting Elementary pre-service teachers. They are asked to carry out the experiment and write their own report about the difficulties and solutions of exploration process. This study aims to analyze Elementary pre-service teachers' reflection and experiment performance ability on Photosynthesis experiment based on 10 groups' reports and presentation materials. Reflective thinking extracts 108 statements which is associated with the four types of the sentence 'Knowledge, Procedure, Orientation, Attitude' in 10 reports. There are many sentences about reflective thinking acquired through analysis of the photosynthesis experiment. reflective thinking about the newly discovered type or changed concepts through experimentation in Knowledge is at the highest frequency. 56 sentences in relation to the ability to perform experiments are extracted by adding 4 different types of reflective thinking in 10 groups shown the highest frequency group and the lowest frequency group's report through analyzing 4 steps 'Experimental preparation and safety accident prevention', 'Experiments performance', 'Experimental results and generalization', and 'Experimental results and feedback.' Results of the analysis showed that there are the biggest difference between the two groups in 'experiment results supplement and feedback step.' In the lowest group's report, there's no contents related with 'Computer-assisted information processing' in the 'Experimental results summary and generalization stage', 'Alternative reagents and materials research', and 'Devising alternative experiment methods'.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.10
• /
• pp.883-892
• /
• 2021

This paper provides a process for developing deep learning-based aerial object detection models that can run in realtime on onboard. To improve object detection performance, we pre-process and augment the training data in the training stage. In addition, we perform transfer learning and apply a weighted cross-entropy method to reduce the variations of detection performance for each class. To improve the inference speed, we have generated inference acceleration engines with quantization. Then, we analyze the real-time performance and detection performance on custom aerial image dataset to verify generalization.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.469-481
• /
• 2018

DL (Deep Learning) is getting popular in various fields to implement artificial intelligence that resembles human learning and cognition. DL based on complicate structure of the ANN (Artificial Neural Network) requires computing power and computation cost. Variety of DL models with improved performance have been developed with powerful computer specification. The main purpose of this paper is to detect buildings from aerial images and evaluate performance of Mask R-CNN (Region-based Convolutional Neural Network) developed by FAIR (Facebook AI Research) team recently. Mask R-CNN is a R-CNN that is evaluated to be one of the best ANN models in terms of performance for semantic segmentation with pixel-level accuracy. The performance of the DL models is determined by training ability as well as architecture of the ANN. In this paper, we characteristics of the Mask R-CNN with various types of the images and evaluate possibility of the generalization which is the ultimate goal of the DL. As for future study, it is expected that reliability and generalization of DL will be improved by using a variety of spatial information data for training of the DL models.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.19-26
• /
• 2020

Steel production requires high-quality surfaces with minimal defects. Therefore, the detection algorithms for the surface defects of steel strip should have good generalization performance. To meet the growing demand for high-quality products, the use of intelligent visual inspection systems is becoming essential in production lines. In this paper, we proposed a ShuffleDefectNet defect detection system based on deep learning. The proposed defect detection system exceeds state-of-the-art performance for defect detection on the Northeastern University (NEU) dataset obtaining a mean average accuracy of 99.75%. We train the best performing detection with different amounts of training data and observe the performance of detection. We notice that accuracy and speed improve significantly when use the overall architecture of ShuffleDefectNet.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.12
• /
• pp.88-93
• /
• 2016

This paper presents a method of constructing the highly performance switching circuit(HPSC) over finite fields. The proposed method is as following. First of all, we extract the input/output relationship of linear characteristics for the given digital switching functions, Next, we convert the input/output relationship to Directed Cyclic Graph using basic gates adder and coefficient multiplier that are defined by mathematical properties in finite fields. Also, we propose the new factorization method for matrix characteristics equation that represent the relationship of the input/output characteristics. The proposed method have properties of generalization and regularity. Also, the proposed method is possible to any prime number multiplication expression.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.377-382
• /
• 2007

This paper proposes a training algorithm for global optimization of the parameters of radial basis function networks. Since conventional training algorithms usually perform only local optimization, the performance of the network is limited and the final network significantly depends on the initial network parameters. The proposed hybrid simulated annealing algorithm performs global optimization of the network parameters by combining global search capability of simulated annealing and local optimization capability of gradient-based algorithms. Via experiments for function approximation problems, we demonstrate that the proposed algorithm can find networks showing better training and test performance and reduce effects of the initial network parameters on the final results.