• 한국항해항만학회지
• /
• 제38권3호
• /
• pp.233-238
• /
• 2014

다양한 여가 활동 수요증가와 함께 국내에서도 여객선과 크루즈선의 이용자수가 증가하고 있다. 그러나 과거 10년간 연평균 15건 이상의 여객선 안전사고가 발생했음에도 불구하고 승조원과 승객의 피난행동특성과 관련된 연구는 매우 미흡한 실정이다. 본 논문은 승선생활 환경에 익숙한 3학년과 승선생활이 생소한 1학년을 대상으로 선박에서의 승선생활 숙련도에 따른 각 피난행동특성을 비교분석한 것이다. 연구결과를 정리하면 다음과 같다. 3학년 실험에서 승조원의 안전성 향상을 위해서는 다양한 재난대응 시나리오의 개발과 실천 교육이 필요하다는 사실을 알게 되었고, 1학년 실험에서는 신규 승조원과 승객을 대상으로 한 초기 피난안전교육과 선내구조에 익숙한 기존 승조원에 의한 피난안전 유도 임무의 중요성을 확인할 수 있다. 또한 재난상황과 그 전개과정에 대한 상황인식 공유가 전체 피난시간에 큰 영향을 미친다는 사실을 설명할 수 있으며, 피난상황 발생 시 승선자의 안전성 향상을 위해서는 피난계획기법 상 선장과 관련 승조원이 반드시 선내 모든 피난경로와 경로별 피난자수를 통제할 수 있어야만 한다는 사실을 확인하였다.

• 한국ITS학회 논문지
• /
• 제20권2호
• /
• pp.95-111
• /
• 2021

인구의 감소 및 고령화 사회가 진행되면서 운전자의 평균 연령은 높아지게 된다. 그에 따라 잠재적인 사고의 위험성이 높은 고령 운전자들은 자율 주행형 개인 이동체가 필요하게 된다. 이러한 이동체가 도로 주행 중에 안전성을 확보하기 위하여 여러 장애물에 대응할 기술이 요구된다. 그 중에서도 주행 중에 마주할 수 있는 차량, 자전거, 사람과 같은 동적 장애물뿐만 아니라 도로 노면의 불량 상태와 같은 정적 장애물을 인식하는 기술이 가장 우선적으로 필요하다. 이를 위해서 본 논문에서는 두 종류의 장애물을 동시에 탐지할 수 있는 심층 신경망 알고리즘을 제안했다. 이 알고리즘을 개발하기 위해서 1,418장의 영상을 이용하여 7종의 동적 장애물에 표기한 annotation data와 도로 노면 파손을 표시한 label 영상을 확보했다. 이를 이용하여 학습한 결과, 46.22%의 평균 정확도로 동적 장애물을 탐지하고 74.71%의 mean intersection over union으로 도로 노면 파손을 탐지했다. 또한 한 장의 영상을 처리하는데 평균 소요시간은 89ms로 일반 차량보다 느린 개인 이동 차량에 사용하기 적합한 알고리즘을 개발했다. 향후 주행 중 마주할 있는 도로 장애물을 탐지하는 기술을 활용하여 개인 이동 차량의 주행 안전성이 향상되길 기대한다.

• 한국항해항만학회지
• /
• 제35권3호
• /
• pp.213-218
• /
• 2011

본 연구는 운항실습선을 대상으로 선내 승선자의 군집유형에 따른 피난 시간과 특성을 시뮬레이션에 의해 비교한 것으로, 크루즈선 승객은 승선 시 다양한 선내 활동이 가능하다는 관점에서 출발하였다. 그리고 승선자와의 인터뷰를 통해 승선 시 행동 유형을 군집유형 A(지정선실 내 재실), 군집유형 B(모든 승조원이 정상활동 공간에 위치), 군집유형 C(모든 승조원이 소속 식당에 위치)의 3가지로 분류하여 비교 평가하였다. 연구의 성과를 정리하면 다음과 같다. 군집유형 B는 피난시간이 다른 유형에 비해 가장 빨랐고 피난 정체구간도 짧았다. 군집유형 C의 경우 승조원이 특정공간에 집중적으로 분포함에 따라 피난과정에서 Upper deck가 병목 구간으로 작용하였고 피난시간이 오래 걸렸다. 이상과 같은 결과로부터 동일한 선박 내에서도 화재 등의 재난 발생 시 선내 재실자의 분포특성에 맞는 피난 관리 및 대응책이 필요함을 알 수 있었다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.181-193
• /
• 2022

Data-driven structural health monitoring (SHM) of civil infrastructure can be used to continuously assess the state of a structure, allowing preemptive safety measures to be carried out. Long-term monitoring of large-scale civil infrastructure often involves data-collection using a network of numerous sensors of various types. Malfunctioning sensors in the network are common, which can disrupt the condition assessment and even lead to false-negative indications of damage. The overwhelming size of the data collected renders manual approaches to ensure data quality intractable. The task of detecting and classifying an anomaly in the raw data is non-trivial. We propose an approach to automate this task, improving upon the previously developed technique of image-based pre-processing on one-dimensional (1D) data by enriching the features of the neural network input data with multiple channels. In particular, feature engineering is employed to convert the measured time histories into a 3-channel image comprised of (i) the time history, (ii) the spectrogram, and (iii) the probability density function representation of the signal. To demonstrate this approach, a CNN model is designed and trained on a dataset consisting of acceleration records of sensors installed on a long-span bridge, with the goal of fault detection and classification. The effect of imbalance in anomaly patterns observed is studied to better account for unseen test cases. The proposed framework achieves high overall accuracy and recall even when tested on an unseen dataset that is much larger than the samples used for training, offering a viable solution for implementation on full-scale structures where limited labeled-training data is available.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3904-3922
• /
• 2022

As a research hotspot, pedestrian detection has a wide range of applications in the field of computer vision in recent years. However, current pedestrian detection methods have problems such as insufficient detection accuracy and large models that are not suitable for large-scale deployment. In view of these problems mentioned above, a lightweight pedestrian detection and early warning method using a new model called you only look once (Yolov5) is proposed in this paper, which utilizing advantages of Yolov5s model to achieve accurate and fast pedestrian recognition. In addition, this paper also optimizes the loss function of the batch normalization (BN) layer. After sparsification, pruning and fine-tuning, got a lot of optimization, the size of the model on the edge of the computing power is lower equipment can be deployed. Finally, from the experimental data presented in this paper, under the training of the road pedestrian dataset that we collected and processed independently, the Yolov5s model has certain advantages in terms of precision and other indicators compared with traditional single shot multiBox detector (SSD) model and fast region-convolutional neural network (Fast R-CNN) model. After pruning and lightweight, the size of training model is greatly reduced without a significant reduction in accuracy, and the final precision reaches 87%, while the model size is reduced to 7,723 KB.

• International Journal of Computer Science & Network Security
• /
• 제23권8호
• /
• pp.177-189
• /
• 2023

Malware detection is an increasingly important operational focus in cyber security, particularly given the fast pace of such threats (e.g., new malware variants introduced every day). There has been great interest in exploring the use of machine learning techniques in automating and enhancing the effectiveness of malware detection and analysis. In this paper, we present a deep recurrent neural network solution as a stacked Long Short-Term Memory (LSTM) with a pre-training as a regularization method to avoid random network initialization. In our proposal, we use global and short dependencies of the inputs. With pre-training, we avoid random initialization and are able to improve the accuracy and robustness of malware threat hunting. The proposed method speeds up the convergence (in comparison to stacked LSTM) by reducing the length of malware OpCode or bytecode sequences. Hence, the complexity of our final method is reduced. This leads to better accuracy, higher Mattews Correlation Coefficients (MCC), and Area Under the Curve (AUC) in comparison to a standard LSTM with similar detection time. Our proposed method can be applied in real-time malware threat hunting, particularly for safety critical systems such as eHealth or Internet of Military of Things where poor convergence of the model could lead to catastrophic consequences. We evaluate the effectiveness of our proposed method on Windows, Ransomware, Internet of Things (IoT), and Android malware datasets using both static and dynamic analysis. For the IoT malware detection, we also present a comparative summary of the performance on an IoT-specific dataset of our proposed method and the standard stacked LSTM method. More specifically, of our proposed method achieves an accuracy of 99.1% in detecting IoT malware samples, with AUC of 0.985, and MCC of 0.95; thus, outperforming standard LSTM based methods in these key metrics.

• Asia pacific journal of information systems
• /
• 제19권3호
• /
• pp.151-179
• /
• 2009

The purpose of this study is to make an empirical analysis on the factors affecting the intention for the continuous use of the BSC system, which has been recently spread in the public sector. Because the object of acceptance is the performance management system based on BSC (Balanced Scorecard) implemented in the form of information systems, this study proposes a research model by applying TAM (Technology Acceptance Model). Independent variables are factors affecting the acceptance of BSC system such as training, communication, IS support, CEO support and personal innovativeness, and we examine their effects on the dependent variable, namely, intention on continuous use via mediating variables: perceived usefulness and perceived ease of use. A questionnaire survey was conducted with public institutions(firms) that had introduced and were operating the BSC system; 264 valid questionnaires are adopted. Collected data are analyzed using SPSS 16.0 and AMOS 7.0. Results of reliability test show that all analyzed data are reliable. In validity test, one item regarding communication was excluded; 9 latent variables and 34 observed variables are used in the final analysis. Based on the results of the hypothesis test through path analysis using a structural equation model, 10 out of 16 hypotheses are accepted. Factors affecting perceived usefulness are training and IS(Information System) support. The analysis results indicate that perceived ease of use is mainly affected by IS support, CEO support, and personal innovativeness among the factors related to the acceptance of the BSC system. This suggests that, contrary to the expectation that the BSC system may be used without difficulty, the management's active support is required in order to attain expected improvement in productivity and work efficiency. This was also pointed out in case studies on the construction of the BSC system in public sectors. On the other hand, perceived ease of use is found to affect perceived usefulness. This supports the results of previous researches on TAM. Perceived ease of use and perceived usefulness are found to affect the attitude towards the use of the system. The intention on continuous use is affected more by perceived usefulness than by the attitude towards the use of system. This result supports the results of previous researches on TAM, showing that the BSC system is utilized substantially in worksites. This study is considered meaningful in that it was actually performed on users at public institutions(firms) that had introduced the BSC system and that it empirically tested hypotheses on the acceptance of the BSC system by applying TAM to the research model.

• Journal of Multimedia Information System
• /
• 제4권1호
• /
• pp.9-18
• /
• 2017

Ubiquitous computing is a novel paradigm that is rapidly gaining in the scenario of wireless communications and telecommunications for realizing smart world. As rapid development of sensor technology, smart sensor system becomes more popular in automobile or vehicle. In this study, a new vehicle detection mechanism in real-time for blind spot area is proposed based on imaging sensors. To determine the position of other vehicles on the road is important for operation of driver assistance systems (DASs) to increase driving safety. As the result, blind spot detection of vehicles is addressed using an automobile detection algorithm for blind spots. The proposed vehicle verification utilizes the height and angle of a rear-looking vehicle mounted camera. Candidate vehicle information is extracted using adaptive shadow detection based on brightness values of an image of a vehicle area. The vehicle is verified using a training set with Haar-like features of candidate vehicles. Using these processes, moving vehicles can be detected in blind spots. The detection ratio of true vehicles was 91.1% in blind spots based on various experimental results.

• 품질경영학회지
• /
• 제44권2호
• /
• pp.451-466
• /
• 2016

Purpose: This study was to identify customers' demands in railway services system and then to seek the way to satisfy the customer expectations. Methods: We suggest a Quality Function Deployment(QFD)-based approach comprised of three stages. In first stage, SERVPERF survey was carried out to assess current positions of customer expectations in the market. Then, factor analysis was incorporated into SERVPERF to classify customer expectations for the house of quality. In the second stage, the analytic network process was used to prioritize the importance of the customer attributes. Finally, QFD was performed utilizing customer attributes and their weights. Results: The QFD identified the most important customer expectations as: accident prevention, swift reaction to accident, on-time arrivals and departures of the train. It also shows that driving capability, equipment for safety, and training for disaster are the most critical technical requirements. Conclusion: The results are useful for identifying the customers' demands in railway services systems, and they can contribute to the service quality and customer satisfaction.

• Journal of Preventive Medicine and Public Health
• /
• 제48권6호
• /
• pp.265-270
• /
• 2015

The recent Middle East respiratory syndrome coronavirus (MERS-CoV) outbreak has originated from a failure in the national quarantine system in the Republic of Korea as most basic role of protecting the safety and lives of its citizens. Furthermore, a number of the Korean healthcare system's weaknesses seem to have been completely exposed. The MERS-CoV outbreak can be considered a typical public health crisis in that the public was not only greatly terrorized by the actual fear of the disease, but also experienced a great impact to their daily lives, all in a short period of time. Preparedness for and an appropriate response to a public health crisis require comprehensive systematic public healthcare measures to address risks comprehensively with an all-hazards approach. Consequently, discussion regarding establishment of post-MERS-CoV improvement measures must focus on the total reform of the national quarantine system and strengthening of the public health infrastructure. In addition, the Korea Centers for Disease Control and Prevention must implement specific strategies of action including taking on the role of "control tower" in a public health emergency, training of Field Epidemic Intelligence Service officers, establishment of collaborative governance between central and local governments for infection prevention and control, strengthening the roles and capabilities of community-based public hospitals, and development of nationwide crisis communication methods.