• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.365-380
• /
• 2022

The target detection algorithm based on supervised learning is the current mainstream algorithm for target detection. A high-quality dataset is the prerequisite for the target detection algorithm to obtain good detection performance. The larger the number and quality of the dataset, the stronger the generalization ability of the model, that is, the dataset determines the upper limit of the model learning. The convolutional neural network optimizes the network parameters in a strong supervision method. The error is calculated by comparing the predicted frame with the manually labeled real frame, and then the error is passed into the network for continuous optimization. Strongly supervised learning mainly relies on a large number of images as models for continuous learning, so the number and quality of images directly affect the results of learning. This paper proposes a dataset STAR-24K (meaning a dataset for Space TArget Recognition with more than 24,000 images) for detecting common targets in space. Since there is currently no publicly available dataset for space target detection, we extracted some pictures from a series of channels such as pictures and videos released by the official websites of NASA (National Aeronautics and Space Administration) and ESA (The European Space Agency) and expanded them to 24,451 pictures. We evaluate popular object detection algorithms to build a benchmark. Our STAR-24K dataset is publicly available at https://github.com/Zzz-zcy/STAR-24K.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.3
• /
• pp.271-277
• /
• 2006

Injection molding process is the manufacturing process that can obtain a high quality products in large quantity to a low cost. Since there are many input factors in every situation that can influence part's quality, the method is difficult to save the exact simulation data. Latest, it deals with the CAE method that supports the experiment, it is applied to the Design of Experiment for the optimum injection molding process. If there are many factors, the interaction among those factors must be considered by applying Design of Experiment which is taken from the technique of minimizing the number of experiment. Without a real test, it is taken the simulation data using $Moldflow^(R)$ software. $Moldflow^(R)$ is used for the analysis of injection molding process, it is analyzed the factors that affect a warpage using the Taguchi method and then the optimal injection molding process is obtained.

• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.439-448
• /
• 2020

In this paper, we propose a modified optimization algorithm for point cloud matching of multi-view RGB-D cameras. In general, in the computer vision field, it is very important to accurately estimate the position of the camera. The 3D model generation methods proposed in the previous research require a large number of cameras or expensive 3D cameras. Also, the methods of obtaining the external parameters of the camera through the 2D image have a large error. In this paper, we propose a matching technique for generating a 3D point cloud and mesh model that can provide omnidirectional free viewpoint using 8 low-cost RGB-D cameras. We propose a method that uses a depth map-based function optimization method with RGB images and obtains coordinate transformation parameters that can generate a high-quality 3D model without obtaining initial parameters.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.3
• /
• pp.124-131
• /
• 2007

Research field on crowd animation can be classified into two major categories. One is to offer realism of the crowd motion and the other is to improve speed of the animation. For the last decade, a lot of research on realism and behavior of crowd have been presented. But lately, research on improving speed seems like more interesting. Therefore, in this paper, we conducted an experiment to analyze what is the main bottleneck of crowd animation. As the result, we find out one of the most important bottleneck is the number of joints transformed in each animation frame. In order to resolve this problem we propose a novel level-of-detail technique 'motion level-of-detail', which is a joint-reduction technique operated in the pre-processing time. We used a non-linear optimization, SQP (sequential quadric programming), to generate the low detailed motions.

• Earthquakes and Structures
• /
• v.17 no.2
• /
• pp.143-162
• /
• 2019

Motivated by a simpler and more compact hybrid active tuned mass damper (ATMD) system with wide frequency spacing (i.e., high robustness) but not reducing the effectiveness using the least number of ATMD units, the active tuned tandem mass dampers (ATTMD) have been proposed to attenuate undesirable oscillations of structures under the ground acceleration. Likewise, it is expected that the frequency spacing of the ATTMD is comparable to that of the active multiple tuned mass dampers (AMTMD) or the multiple tuned mass dampers (MTMD). In accordance with the mode generalised system in the specific vibration mode being controlled (simply referred herein to as the structure), the closed-form expression of the dimensionless displacement variances has been derived for the structure with the attached ATTMD. The criterion for the optimum searching may then be determined as minimization of the dimensionless displacement variances. Employing the gradient-based optimization technique, the effects of varying key parameters on the performance of the ATTMD have been scrutinized in order to probe into its superiority. Meanwhile, for the purpose of a systematic comparison, the optimum results of two active tuned mass dampers (two ATMDs), two tuned mass dampers (two TMDs) without the linking damper, and the TTMD are included into consideration. Subsequent to work in the frequency domain, a real-time Simulink implementation of dynamic analysis of the structure with the ATTMD under earthquakes is carried out to verify the findings of effectiveness and stroke in the frequency domain. Results clearly show that the findings in the time domain support the ones in the frequency domain. The whole work demonstrates that ATTMD outperforms two ATMDs, two TMDs, and TTMD. Thereinto, a wide frequency spacing feature of the ATTMD is its highlight, thus deeming it a high robustness control device. Furthermore, the ATTMD system only needs the linking dashpot, thus embodying its simplicity.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• pp.807-817
• /
• 2019

Airborne Gamma Ray Spectrometry (AGRS) with its important applications such as gathering radiation information of ground surface, geochemistry measuring of the abundance of Potassium, Thorium and Uranium in outer earth layer, environmental and nuclear site surveillance has a key role in the field of nuclear science and human life. The Broyden-Fletcher-Goldfarb-Shanno (BFGS), with its advanced numerical unconstrained nonlinear optimization in collaboration with Artificial Neural Networks (ANNs) provides a noteworthy opportunity for modern AGRS. In this study a new AGRS system empowered by ANN-BFGS has been proposed and evaluated on available empirical AGRS data. To that effect different architectures of adaptive ANN-BFGS were implemented for a sort of published experimental AGRS outputs. The selected approach among of various training methods, with its low iteration cost and nondiagonal scaling allocation is a new powerful algorithm for AGRS data due to its inherent stochastic properties. Experiments were performed by different architectures and trainings, the selected scheme achieved the smallest number of epochs, the minimum Mean Square Error (MSE) and the maximum performance in compare with different types of optimization strategies and algorithms. The proposed method is capable to be implemented on a cost effective and minimum electronic equipment to present its real-time process, which will let it to be used on board a light Unmanned Aerial Vehicle (UAV). The advanced adaptation properties and models of neural network, the training of stochastic process and its implementation on DSP outstands an affordable, reliable and low cost AGRS design. The main outcome of the study shows this method increases the quality of curvature information of AGRS data while cost of the algorithm is reduced in each iteration so the proposed ANN-BFGS is a trustworthy appropriate model for Gamma-ray data reconstruction and analysis based on advanced novel artificial intelligence systems.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.529-533
• /
• 2019

Combined cycle power plants are often used to produce power. These days prediction of power plant output based on operating parameters is a major concern. This paper presents an approach to using computational intelligence technique to predict the output power of combined cycle power plant. Computational intelligence techniques have been developed and applied to many real world problems. In this paper, tree architectures of fuzzy neural networks are considered to predict the output power. Tree architectures of fuzzy neural networks have an advantage of reducing the number of rules by selecting fuzzy neurons as nodes and relevant inputs as leaves optimally. For the optimization of the networks, two-step optimization method is used. Genetic algorithms optimize the binary structure of the networks by selecting the nodes and leaves as binary, and followed by random signal-based learning further refines the optimized binary connections in the unit interval. To verify the effectiveness of the proposed method, combined cycle power plant dataset obtained from the UCI Machine Learning Repository Database is considered.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.59-76
• /
• 2023

Since the COVID-19 era, the rise in apartment prices has been unconventional. In this uncertain real estate market, price prediction research is very important. In this paper, a model is created to predict the actual transaction price of future apartments after building a vast data set of 870,000 from 2015 to 2020 through data collection and crawling on various real estate sites and collecting as many variables as possible. This study first solved the multicollinearity problem by removing and combining variables. After that, a total of five variable selection algorithms were used to extract meaningful independent variables, such as Forward Selection, Backward Elimination, Stepwise Selection, L1 Regulation, and Principal Component Analysis(PCA). In addition, a total of four machine learning and deep learning algorithms were used for deep neural network(DNN), XGBoost, CatBoost, and Linear Regression to learn the model after hyperparameter optimization and compare predictive power between models. In the additional experiment, the experiment was conducted while changing the number of nodes and layers of the DNN to find the most appropriate number of nodes and layers. In conclusion, as a model with the best performance, the actual transaction price of apartments in 2021 was predicted and compared with the actual data in 2021. Through this, I am confident that machine learning and deep learning will help investors make the right decisions when purchasing homes in various economic situations.

• Interaction and multiscale mechanics
• /
• v.5 no.3
• /
• pp.211-228
• /
• 2012

This paper reports the development of a generalized inverse analysis formulation for the parameter estimation of four-parameter Burger model. The analysis is carried out by formulating the problem as a mathematical programming formulation in terms of identification of the design vector, the objective function and the design constraints. Thereafter, the formulated constrained nonlinear multivariable problem is solved with the aid of fmincon: an in-built constrained optimization solver module available in MatLab. In order to gain experience, a synthetic case-study is considered wherein key issues such as the determination and setting up of variable bounds, global optimality of the solution and minimum number of data-points required for prediction of parameters is addressed. The results reveal that the developed technique is quite efficient in predicting the model parameters. The best result is obtained when the design variables are subjected to a lower bound without any upper bound. Global optimality of the solution is achieved using the developed technique. A minimum of 4-5 randomly selected data-points are required to achieve the optimal solution. The above technique has also been adopted for real-time settlement of four oil refineries with encouraging results.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.3
• /
• pp.19-29
• /
• 2010

A marketing model is changed from a customer acquisition to customer retention and it is being moved to a way that enhances the quality of customer interaction to add value to our customers. Such personalization is emerging from this background. The Web site is accelerate the adoption of a personalization, and in contrast to the rapid growth of data, quantitative analytical experience is required. For the automated analysis of large amounts of data and the results must be passed in real time of personalization has been interested in technical problems. A recommendation algorithm is an algorithm for the implementation of personalization, which predict whether the customer preferences and purchasing using the database with new customers interested or likely to purchase. As recommended number of users increases, the algorithm increases recommendation time is the problem. In this paper, to solve this problem, a recommendation system based on clustering and dimensionality reduction is proposed. First, clusters customers with such an orientation, then shrink the dimensions of the relationship between customers to low dimensional space. Because finding neighbors for recommendations is performed at low dimensional space, the computation time is greatly reduced.