• Industry Promotion Research
• /
• v.8 no.4
• /
• pp.241-250
• /
• 2023

In investigating missing persons cases, the focus is on strengthening the efficiency of detectives' work in investigating missing persons cases. Disappearance cases are seen as a problem that directly affects social safety and individual well-being. The research has the following structure. The introduction presents the necessity, scope, and methodology of the study. Next, we analyze the definition, causes, types and actual conditions of disappearance cases. In terms of problems in the process of handling missing persons cases, the initial response process, problems in the response of related agencies after reporting, problems due to the prolongation of the case, and problems due to legal restrictions are analyzed. In the plan to improve the work efficiency of detectives for disappearance cases, the revitalization of public interest investigation networks, strengthening of capacity related to disappearance case investigation, professional public interest detective certification system, and establishment of exception provisions for detective activities are discussed. In the conclusion, we present what is necessary for the activities of public interest detectives specializing in missing persons cases.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.200-217
• /
• 2023

The purpose of this study is to analyze the urban spatial network of Busan in a time-series manner to examine the planning role of centeral area change and spatial management. The analysis method used the ASA(Angular Segment Analysis) technique of the space syntax theory, and the analysis period was divided into three periods from 1937 to 2022, when the Busan urban planning was established. As the result, the center area formed in Nampo-dong and Gwangbok-dong were differentiated and moved to Seomyeon, and then the process of redifferentiated into Yeonsan was confirmed in Seomyeon. It is also room for the center area to move to the western region in the future, and it was possible to identify the center area formed by the policy-set and amenity elements. In addition, we examined the system of urban spatial structure through the intelligibility analysis of space syntax theory, and found that the qualitative interpretation of existing studies and the interpretation from the perspective of space syntax theory were consistent with each other. Through this, it was possible to confirm that the role of the plan for space management.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.31-34
• /
• 2024

태양광 에너지는 탄소 중립 이행을 위한 주요 방안으로 많은 주목을 받고 있다. 태양광 발전량은 여러 환경적 요인에 따라 크게 달라질 수 있으므로, 정확한 발전량 예측은 전력 네트워크의 안정성과 효율적인 에너지 관리에 근본적으로 중요하다. 대표적인 인공지능 기술인 신경망(Neural Network)은 불안정한 환경 변수와 복잡한 상호작용을 효과적으로 학습할 수 있어 태양광 발전량 예측에서 우수한 성능을 도출하였다. 하지만, 신경망은 모델의 구조나 초매개변수(Hyperparameter)를 최적화하는 것은 복잡하고 시간이 많이 드는 작업이므로, 에너지 분야에서 실제 산업 적용에 한계가 존재한다. 본 논문은 2단계 신경망 최적화를 통한 태양광 발전량 예측 기법을 제안한다. 먼저, 태양광 발전량 데이터 셋을 훈련 집합과 평가 집합으로 분할한다. 훈련 집합에서, 각기 다른 은닉층의 개수로 구성된 여러 신경망 모델을 구성하고, 모델별로 Optuna를 적용하여 최적의 초매개변숫값을 선정한다. 다음으로, 은닉층별 최적화된 신경망 모델을 이용해 훈련과 평가 집합에서는 각각 5겹 교차검증을 적용한 발전량 추정값과 예측값을 출력한다. 마지막으로, 스태킹 앙상블 방식을 채택해 기본 초매개변숫값으로 설정해도 우수한 성능을 도출하는 랜덤 포레스트를 이용하여 추정값을 학습하고, 평가 집합의 예측값을 입력으로 받아 최종 태양광 발전량을 예측한다. 인천 지역으로 실험한 결과, 제안한 방식은 모델링이 간편할 뿐만 아니라 여러 신경망 모델보다 우수한 예측 성능을 도출하였으며, 이를 바탕으로 국내 에너지 산업에 이바지할 수 있을 것으로 기대한다.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.79-86
• /
• 2024

In this paper, a novel visual tracking method which can utilize the context of object regions is presented. Conventional methods have the inherent problem of treating all candidate regions independently, where the tracker could not successfully discriminate regions with similar appearances. This was due to lack of contextual modeling in a given scene, where all candidate object regions should be taken into consideration when choosing a single region. The goal of the proposed method is to encourage feature exchange between candidate regions to improve the discriminability between similar regions. It improves upon conventional methods that only consider a single region, and is implemented by employing the MLP-Mixer model for enhanced feature exchange between regions. By implementing channel-wise, inter-region interaction operation between candidate features, contextual information of regions can be embedded into the individual feature representations. To evaluate the performance of the proposed tracker, the large-scale LaSOT dataset is used, and the experimental results show a competitive AUC performance of 0.560 while running at a real-time speed of 65 fps.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.1
• /
• pp.11-17
• /
• 2024

Lung cancer segmentation in chest CT images is challenging due to the varying sizes of tumors and the presence of surrounding structures with similar intensity values. To address these issues, we propose a lung cancer segmentation network that incorporates deep supervision and utilizes UNet3+ as the backbone. Additionally, we propose a hybrid lesion focal loss function comprising three components: pixel-based, region-based, and shape-based, which allows us to focus on the smaller tumor regions relative to the background and consider shape information for handling ambiguous boundaries. We validate our proposed method through comparative experiments with UNet and UNet3+ and demonstrate that our proposed method achieves superior performance in terms of Dice Similarity Coefficient (DSC) for tumors of all sizes.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.6
• /
• pp.269-276
• /
• 2006

In this paper, we propose an emotion recognition system that can discriminate human emotional state into neutral or anger from the speech captured by a cellular-phone in real time. In general. the speech through the mobile network contains environment noise and network noise, thus it can causes serious System performance degradation due to the distortion in emotional features of the query speech. In order to minimize the effect of these noise and so improve the system performance, we adopt a simple MA (Moving Average) filter which has relatively simple structure and low computational complexity, to alleviate the distortion in the emotional feature vector. Then a SFS (Sequential Forward Selection) feature optimization method is implemented to further improve and stabilize the system performance. Two pattern recognition method such as k-NN and SVM is compared for emotional state classification. The experimental results indicate that the proposed method provides very stable and successful emotional classification performance such as 86.5%. so that it will be very useful in application areas such as customer call-center.

• Information Systems Review
• /
• v.21 no.2
• /
• pp.51-67
• /
• 2019

This study examines IPO underpricing and its interaction with investor attention as one of key factors that can affect post-IPO performance in the short- and long-run. With higher investor attention measured as Google searches around IPO dates, our empirical results show that IT firms are underpriced more severely than non-IT firms. We also demonstrate that investor attention is positively associated with IPO performance in the short-run for both IT and non-IT firms. However, the impact of investor attention is more sustained for IT firms over a longer time horizon when coupled with a high level of underpricing while its impact is neutralized for non-IT firms. Given the unique attributes such as network effects embedded in the IT industry, our findings suggest that IPO underpricing and its interplay with investor attention for IT firms play an important role in shaping long-run performance and ultimately affecting fundamental value.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.

• Journal of Space Technology and Applications
• /
• v.4 no.2
• /
• pp.169-183
• /
• 2024

Korea Astronomy and Space Science Institute (KASI) developed the Geochang satellite laser ranging (SLR) system for the scientific research on the space geodesy as well as for the national space missions including precise orbit determination and space surveillance. The operation system was developed based on the server-client communication structure, which controls the SLR subsystems, provides manual and automatic observation modes based on the observation algorithm, generates the range data between satellites and SLR stations, and carry out the post-processing to remove noises. In this study, we analyzed the requirements of operation system, and presented the development environments, the software structure and the observation algorithm, for the server-client communications. We also obtained laser ranging data for the ground target and the space geodetic satellite, and then analyzed the ranging precision between the Geochang SLR station and the International Laser Ranging Service (ILRS) network stations, in order to verify the operation system.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.4
• /
• pp.1-7
• /
• 2024

Traditional security model design presents two primary issues. First, these models have been developed and implemented with a technology-centered approach rather than considering human factors. Such structures can be undermined by cognitive vulnerabilities like psychological resistance within organizations and user errors. Second, these models are typically designed based on network perimeter security. This design is unsuitable for the boundary-less remote work environments rapidly becoming prevalent due to the Fourth Industrial Revolution and the COVID-19 pandemic. This paper proposes an approach to address these limitations by integrating human-centered threats within the Zero Trust security model, a state-of-the-art boundary-less security framework. By doing so, we suggest a robust security model design that can protect against both technical and human-centered threats.