• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.714-720
• /
• 2021

The ITL (information theoretic learning) based on the kernel density estimation method that has successfully been applied to machine learning and signal processing applications has a drawback of severe sensitiveness in choosing proper kernel sizes. For the maximization of correntropy criterion (MCC) as one of the ITL-type criteria, several methods of adapting the remaining kernel size ( ) after removing the term have been studied. In this paper, it is shown that the main cause of sensitivity in choosing the kernel size derives from the term and that the adaptive adjustment of in the remaining terms leads to approach the absolute value of error, which prevents the weight adjustment from continuing. Thus, it is proposed that choosing an appropriate constant as the kernel size for the remaining terms is more effective. In addition, the experiment results when compared to the conventional algorithm show that the proposed method enhances learning performance by about 2dB of steady state MSE with the same convergence rate. In an experiment for channel models, the proposed method enhances performance by 4 dB so that the proposed method is more suitable for more complex or inferior conditions.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.34 no.9
• /
• pp.3-10
• /
• 2018

While computer environments have been dramatically developed in recent years, as the building structures become larger, the structural analysis models are also becoming more complex. So there is still a need to model one shear wall with one finite element. From the viewpoint of the concept of FEA, if one shear wall is modeled by one finite element, the result of analysis is not likely accurate. Shear wall may be modelled with various finite elements. Among them, considering the displacement compatibility condition with the beam element connected to the shear wall, plane stress element with in-plane rotational stiffness is preferred. Therefore, in order to analyze one shear wall with one finite element accurately, it is necessary to evaluate finite elements developed for the shear wall analysis and to develop various plane stress elements with rotational stiffness continuously. According to the above mentioned need, in this study, the theory about a plane stress element using hierarchical interpolation equation is reviewed and stiffness matrix is derived. And then, a computer program using this theory is developed. Developed computer program is used for numerical experiments to evaluate the analysis results using commercial programs such as SAP2000, ETABS, PERFORM-3D and MIDAS. Finally, the deflection equation of a cantilever beam with narrow rectangular section and bent by an end load P is derived according to the elasticity theory, and it is used to for comparison with theoretical solution.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.10
• /
• pp.159-168
• /
• 2022

The time series of foreign trade turnover is complex and variable and contains linear and nonlinear information. This paper proposes preprocessing the dataset by the EMD algorithm and combining the linear prediction advantage of the SARIMA model with the nonlinear prediction advantage of the EMD-LSTM model to construct the SARIMA-EMD-LSTM hybrid model by the weight assignment method. The forecast performance of the single models is compared with that of the hybrid models by using MAPE and RMSE metrics. Furthermore, it is confirmed that the weight assignment approach can benefit from the hybrid models. The results show that the SARIMA model can capture the fluctuation pattern of the time series, but it cannot effectively predict the sudden drop in foreign trade turnover caused by special reasons and has the lowest accuracy in long-term forecasting. The EMD-LSTM model successfully resolves the hysteresis phenomenon and has the highest forecast accuracy of all models, with a MAPE of 7.4304%. Therefore, it can be effectively used to forecast the Sino-Russia foreign trade turnover time series post-epidemic. Hybrid models cannot take advantage of SARIMA linear and LSTM nonlinear forecasting, so weight assignment is not the best method to construct hybrid models.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.50-56
• /
• 2020

With the rapid development of deep neural network theory and application research, the effect of face detection has been improved. However, due to the complexity of deep neural network calculation and the high complexity of the detection environment, how to detect face quickly and accurately becomes the main problem. This paper is based on the relatively simple model of the MTCNN model, using FDDB (Face Detection Dataset and Benchmark Homepage), LFW (Field Label Face) and FaceScrub public datasets as training samples. At the same time of sorting out and introducing MTCNN(Multi-Task Cascaded Convolutional Neural Network) model, it explores how to improve training speed and Increase performance at the same time. In this paper, the dynamic image pyramid technology is used to replace the traditional image pyramid technology to segment samples, and OHEM (the online hard example mine) function in MTCNN model is deleted in training, so as to improve the training speed.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.3
• /
• pp.215-222
• /
• 2021

As the importance of human-computer interaction (Human Computer Interface) technology grows and research on HCI is progressing, it is inferred about the research emotion inference or the computer reaction according to the user's intention, not the computer reaction by the standard input of the user. Stress is an unavoidable result of modern human civilization, and it is a complex phenomenon, and depending on whether or not there is control, human activity ability can be seriously changed. In this paper, we propose an intelligent emotional recommendation system using music as a way to relieve stress after measuring heart rate variability (HRV) and acceleration photoplethymogram (APG) increased through stress as part of human-computer interaction. The differential evolution algorithm was used to extract reliable data by acquiring and recognizing the user's biometric information, that is, the stress index, and emotional inference was made through the semantic web based on the obtained stress index step by step. In addition, by searching and recommending a music list that matches the stress index and changes in emotion, an emotional recommendation system suitable for the user's biometric information was implemented as an application.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.14 no.5
• /
• pp.121-133
• /
• 2004

In [15,16], Okeya and Sakurai showed that the randomized addition-subtraction chains countermeasures [18] are vulnerable to SPA attack. In this paper, we show that Okeya and Sakurai's attack algorithm [15,16] has two latent problems which need to be considered. We further propose new powerful concrete attack algorithms which are different from [15,16,19]. From our implementation results for standard 163-bit keys, the success probability for the simple version with 20 AD sequences is about 94% and with 30 AD sequences is about 99%. Also, the success probability for the complex version with 40 AD sequences is about 94% and with 70 AD sequences is about 99%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.15 no.5
• /
• pp.93-103
• /
• 2005

In this paper, we propose a RST resistant robust logo embedding technique for multimedia copyright protection Geometric manipulations are challenging attacks in that they do not introduce the quality degradation very much but make the detection process very complex and difficult. Watermark embedding in the normalized image directly suffers from smoothing effect due to the interpolation during the image normalization. This can be avoided by estimating the transform parameters using an image normalization technique, instead of embedding in the normalized image. Conventional RST resistant schemes that use full frame transform suffer from the absence of effective perceptual masking methods. Thus, we adopt $8\times8$ block DCT and calculate masking using a spatio-frequency localization of the $8\times8$ block DCT coefficients. Simulation results show that the proposed algorithm is robust against various signal processing techniques, compression and geometrical manipulations.

• Journal of Biomedical Engineering Research
• /
• v.42 no.3
• /
• pp.100-106
• /
• 2021

Measuring Intima-media thickness (IMT) with ultrasound images can help early detection of coronary artery disease. As a result, numerous machine learning studies have been conducted to measure IMT. However, most of these studies require several steps of pre-treatment to extract the boundary, and some require manual intervention, so they are not suitable for on-site treatment in urgent situations. in this paper, we propose to use deep learning networks U-Net, Attention U-Net, and Pretrained U-Net to automatically segment the intima-media complex. This study also applied the HE, HS, and CLAHE preprocessing technique to wireless portable ultrasound diagnostic device images. As a result, The average dice coefficient of HE applied Models is 71% and CLAHE applied Models is 70%, while the HS applied Models have improved as 72% dice coefficient. Among them, Pretrained U-Net showed the highest performance with an average of 74%. When comparing this with the mean value of IMT measured by Conventional wired ultrasound equipment, the highest correlation coefficient value was shown in the HS applied pretrained U-Net.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.1-6
• /
• 2021

Most of the match-3 puzzle games supports automatic play using the MCTS algorithm. However, implementing reinforcement learning agents is not an easy job because it requires both the knowledge of machine learning and the way of complex interactions within the development environment. This study proposes a method in which we can easily design reinforcement learning agents and implement game play agents by applying PPO(Proximal Policy Optimization) algorithms. And we could identify the performance was increased about 44% than the conventional method. The tools we used are the Unity 3D game engine and Unity ML SDK. The experimental result shows that agents became to learn game rules and make better strategic decisions as experiments go on. On average, the puzzle gameplay agents implemented in this study played puzzle games better than normal people. It is expected that the designed agent could be used to speed up the game level design process.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.10
• /
• pp.1255-1261
• /
• 2020

Currently, the spread of mobile devices is gradually increasing. Accordingly, various techniques using images or photos are actively being researched. However, image data generates noise for complex reasons, and the accuracy of image processing increases according to the performance of removing noise. Therefore, noise reduction is one of the essential steps. Salt and pepper noise is a typical impulse noise in the image, and various studies are being conducted to remove the noise. However, existing algorithms have poor noise rejection performance in high frequency areas, and average filters have blurring. Therefore, in this paper, we propose an algorithm that effectively removes salt and pepper noise in the high frequency region as well as the low frequency region using entropy. For objective and accurate judgment of proposed algorithms, MSE and PSNR were used to compare and analyze existing algorithms.