• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.63-63
• /
• 2020

최근 머신러닝 기술의 발전에 따라 비선형 시계열자료에 대한 예측이 가능해졌으며, 기존의 과정기반모형을 대체하여 지하수, 하천수 예측 등 다양한 수문분야에 활용되고 있다. 본 연구에서는 기존의 연구들과 달리 과정기반모형을 이용한 하천수 모의결과를 개선하기 위해 과정기반모형과 결합하는 방식으로 머신러닝 기술을 활용하였다. 머신러닝 기술을 통해 관측값과 모의값 간의 차이를 예측하고 과정기반모형의 모의결과에 반영함으로써 관측값을 정확히 재현할 수 있도록 하는 시스템을 구축하고 평가하였다. 과정기반모형으로는 Weather Research and Forecasting model-Hydrological modeling system (WRF-Hydro)을 소양강 유역을 대상으로 구축하였다. 머신러닝 모형으로는 순환 신경망 중 하나인 Long Short-Term Memory (LSTM) 신경망을 이용하여 장기시계열예측이 가능하게 하였다(WRF-Hydro-LSTM). 머신러닝 모형은 2013년부터 2017년까지의 기상자료 및 유입량 잔차를 이용하여 학습시키고, 2018년 기상자료를 이용하여 예상되는 유입량 잔차를 모의하였다. 모의된 잔차를 WRF-Hydro 모의결과에 반영시켜 최종 유입량 모의값을 보정하였다. 또한, 연구에서 제안된 새로운 방법론의 성능을 비교평가하기 위해 머신러닝 단독 모형으로 유입량을 학습 후 모의하였다(LSTM-only). 상관계수와 Nash-Sutcliffe 효율계수(NSE)를 사용해 평가한 결과, LSTM을 이용한 두 방법(WRF-Hydro-LSTM과 LSTM-only) 모두 기존의 과정기반모형(WRF-Hydro-only)에 비해 높은 정확도의 하천수 모의가 가능했으며, PBIAS 지수를 사용하여 평가한 결과, LSTM을 단독으로 사용하였을 때보다 WRF-Hydro와 결합했을 때 더 관측값과 가까운 모의가 가능함을 확인할 수 있었다.

• Journal of Korean Society of Archives and Records Management
• /
• v.23 no.2
• /
• pp.95-115
• /
• 2023

This study designed a data model linked to Wikidata and examined its applicability to increase the utilization of the digital archive records of the National Debt Redemption Movement, registered as World Memory Heritage, and implications were derived by analyzing the existing metadata, thesaurus, and semantic network graph. Through analysis of the original text of the National Debt Redemption Movement records, key data model classes for linking with Wikidata, such as record item, agent, time, place, and event, were derived. In addition, by identifying core properties for linking between classes and applying the designed data model to actual records, the possibility of acquiring abundant related information was confirmed through movement between classes centered on properties. Thus, this study's result showed that Wikidata's strengths could be utilized to increase data usage in local archives where the scale and management of data are relatively small. Therefore, it can be considered for application in a small-scale archive similar to the National Debt Redemption Movement digital archive.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.5
• /
• pp.49-56
• /
• 2023

This study proposed a damage location estimation method for automated image-based port infrastructure inspection. Memory efficiency was improved by calculating the homography matrix using feature detection technology and outlier removal technology, without going through the 3D modeling process and storing only damage information. To develop an algorithm specialized for port infrastructure, the algorithm was optimized through ground-truth coordinate pairs created using images of port infrastructure. The location errors obtained by applying this to the sample and concrete wall were (X: 6.5cm, Y: 1.3cm) and (X: 12.7cm, Y: 6.4cm), respectively. In addition, by applying the algorithm to the concrete wall and displaying it in the form of an exterior damage map, the possibility of field application was demonstrated.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.456-477
• /
• 2024

With information technology's rapid development, the Internet faces serious security problems. Studies have shown that malware has become a primary means of attacking the Internet. Therefore, adversarial samples have become a vital breakthrough point for studying malware. By studying adversarial samples, we can gain insights into the behavior and characteristics of malware, evaluate the performance of existing detectors in the face of deceptive samples, and help to discover vulnerabilities and improve detection methods for better performance. However, existing adversarial sample generation methods still need help regarding escape effectiveness and mobility. For instance, researchers have attempted to incorporate perturbation methods like Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), and others into adversarial samples to obfuscate detectors. However, these methods are only effective in specific environments and yield limited evasion effectiveness. To solve the above problems, this paper proposes a malware adversarial sample generation method (PixGAN) based on the pixel attention mechanism, which aims to improve adversarial samples' escape effect and mobility. The method transforms malware into grey-scale images and introduces the pixel attention mechanism in the Deep Convolution Generative Adversarial Networks (DCGAN) model to weigh the critical pixels in the grey-scale map, which improves the modeling ability of the generator and discriminator, thus enhancing the escape effect and mobility of the adversarial samples. The escape rate (ASR) is used as an evaluation index of the quality of the adversarial samples. The experimental results show that the adversarial samples generated by PixGAN achieve escape rates of 97%, 94%, 35%, 39%, and 43% on the Random Forest (RF), Support Vector Machine (SVM), Convolutional Neural Network (CNN), Convolutional Neural Network and Recurrent Neural Network (CNN_RNN), and Convolutional Neural Network and Long Short Term Memory (CNN_LSTM) algorithmic detectors, respectively.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.42 no.5
• /
• pp.1-12
• /
• 2014

Recently, the reenaction of place's memories that were considered sense of place based on its historical and structural characteristic were assumed important topics for recovering and making the park from industrial sites, such as factory, industrial complex, industrial city, labor residential development, landfill, etc., to the public all over the world. This research aimed to conduct some preliminary data for making the park's sense of place highly. So, after conducting the structural forms of sense of place for Seoul Forest and Seonyudo Park that were designed and built by considering its place memories actively, park planning and design implications that were considered its sense of place were suggested. The research results those were derived from the structural forms of sense of place for Seoul Forest and Seonyudo Park's are the following. First, the Structural Equation Modeling(SEM) on Seoul Forest and Seonyudo Park were determined that the natural-physical, artificial-physical, and experiential environmental factors' satisfaction of Seoul Forest and Seonyudo Park have an effect on forming the Sense of Place. In addition, the sense of place was affected by the willingness to visit. Second, according to its physical environment, the experiential factors, such as plays, exercises and etc., were more important relatively. Therefore, experiential factors should be considered significantly with physical factors to make the sense of place highly. Third, even the places were under similar category, the factors' and valuables' relative importance were changed. As the results of structural equation modeling said, specific valuables' related with each factors would be differentiated due to the places' characteristics. For example, the results were showed that natural-physical factor was more important than artificial-physical factor in Seoul Forest. On the other hand, artificial-physical factor was more important than the natural- physical factor. This research carries some significance for applying a quantitative research method(structural equation modeling) to various place to conduct the sense of place's structural model, for suggesting relative specific methods to make the sense of place, and for being a step forward to substance of sense of place. If further studies conduct focusing on various places to draw the forming models of sense of place that were based this research's analysis methods and results, those researches would contribute to make the urban place meaningful, characteristically and affectionately. Furthermore, those researches would contribute in making a humane and competitiveness city.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.15 no.2
• /
• pp.293-303
• /
• 2002

Recently, the use of the high strength materials and development of construction techniques have resulted in more flexible and longer spanning in the stadium systems. So the natural frequency of stadium structures are became low. Stadium stand could be led to significant dynamic response as like resonance due to spectator rhythmical activities. The accurate analysis of dynamic behavior of stadium systems and the precise investigation of the dynamic loads on stadium structures are demanded for effective design. It is desirable to apply measured dynamic loads created by spectator activities because these dynamic loads are not easy to express numerical formula. As the floor mesh of stadium stand is refined, the number of divided elements increases in numerical analysis. the rise of the number of elements makes the numbers of nodal points increased and numerous computer memory required. So it is difficult to analysis refine full model of stadium structures by using the commercial programs. In this study, the various dynamic loads induced by spectator movements are measured and analyzed. And a new modeling method that reduce the nodal points are introduced. Vibration analysis of stadium stands is executed to inspect accuracy and efficiency of proposed method in this paper.

• Journal of Internet Computing and Services
• /
• v.10 no.5
• /
• pp.65-78
• /
• 2009

The most commonly adopted approach to find valuable information from tree data is to extract frequently occurring subtree patterns from them. Because mining frequent tree patterns has a wide range of applications such as xml mining, web usage mining, bioinformatics, and network multicast routing, many algorithms have been recently proposed to find the patterns. However, existing tree mining algorithms suffer from several serious pitfalls in finding frequent tree patterns from massive tree datasets. Some of the major problems are due to (1) modeling data as hierarchical tree structure, (2) the computationally high cost of the candidate maintenance, (3) the repetitious input dataset scans, and (4) the high memory dependency. These problems stem from that most of these algorithms are based on the well-known apriori algorithm and have used anti-monotone property for candidate generation and frequency counting in their algorithms. To solve the problems, we base a pattern-growth approach rather than the apriori approach, and choose to extract maximal frequent subtree patterns instead of frequent subtree patterns. The proposed method not only gets rid of the process for infrequent subtrees pruning, but also totally eliminates the problem of generating candidate subtrees. Hence, it significantly improves the whole mining process.

• Journal of Digital Convergence
• /
• v.15 no.12
• /
• pp.323-331
• /
• 2017

In this paper, four types of deposition control processes such as temperature, pressure, input/output(I/O), and gas were replaced by the Internet of Things(IoT) to analyze the data load and sequence procedure before and after the application of it. Through this analysis, we designed the load balancing in the sensing area of the deposition process by creating the sequence diagram of the deposition process. In order to do this, we were modeling of the sensor I/O according to the arrival process and derived the result of measuring the load of CPU and memory. As a result, it was confirmed that the reliability on the deposition processes were improved through performing some functions of the equipment controllers by the IoT. As confirmed through this paper, by applying the IoT to the deposition process, it is expected that the stability of the equipment will be improved by minimizing the load on the equipment controller even when the equipment is expanded.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.5 s.305
• /
• pp.85-94
• /
• 2005

Display characterization, deriving the relationship between digital input values and the corresponding CIEXYZ tri-stimulus values, is important to reproduce the accurate color in color management system. The relationship can be estimated from the nine channel TRCs(tone response curves) and the result of this characterization method is better than that of using three channel TRCs. However, the inverse display characterization using nine channel TRCs cannot be directly inverted because the CIEXYZ values corresponding to each of RGB values are inseparable. Accordingly, inverse display characterization is usually implemented by the 3D-LUT (look-up table) method. Although the result of 3B-LUT is accurate, creating the 3D-LUT requires a lot of memory space and considerable amount of measurements. Therefore the inverse characterization method is proposed based on the modeling of channel-dependent values and nine channel inverse process based on the GOG(gain, offset gamma) model. The proposed method enhances the accuracy of display characterization and reduces the complexity and the number of measurements data required for accuracy in 3-D LUT.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.3
• /
• pp.258-268
• /
• 2001

The compression method is necessarily used to send the high quality moving picture that contains a number of data in image processing. In the field of moving picture compression method, the motion estimation algorithm is used to reduce the temporal redundancy. Block matching algorithm to be usually used is distinguished partial search algorithm with full search algorithm. Full search algorithm be used in this paper is the method to compare the reference block with entire block in the search window. It is very efficient and has simple data flow and control circuit. But the bigger the search window, the larger hardware size, because large computational operation is needed. In this paper, we design the full search block matching motion estimator. Using the DCT DC values, we decide luminance. And we apply 3 bit compare-selector using bit plane to I(Intra coded) picture, not using 8 bit luminance signals. Also it is suggested that use the same selective bit for the P(Predicted coded) and B(Bidirectional coded) picture. We compare based full search method with PSNR(Peak Signal to Noise Ratio) for C language modeling. Its condition is the reference block 8$\times$8, the search window 24$\times$24 and 352$\times$288 gray scale standard video images. The result has small difference that we cannot see. And we design the suggested motion estimator that hardware size is proved to reduce 38.3％ for structure I and 30.7％ for structure II. The memory is proved to reduce 31.3％ for structure I and II.