• Journal of Multimedia Information System
• /
• v.7 no.3
• /
• pp.205-214
• /
• 2020

Crime has become one of the major problems in modern society. Even though visual surveillances through closed-circuit television (CCTV) is extensively used for solving crime, the number of crimes has not decreased. This is because there is insufficient workforce for performing 24-hour surveillance. In addition, CCTV surveillance by humans is not efficient for detecting dangerous situations owing to accuracy issues. In this paper, we propose the autonomous detection of dangerous situations in CCTV scenes using a deep learning model with relational inference. The main feature of the proposed method is that it can simultaneously perform object detection and relational inference to determine the danger of the situations captured by CCTV. This enables us to efficiently classify dangerous situations by inferring the relationship between detected objects (i.e., distance and position). Experimental results demonstrate that the proposed method outperforms existing methods in terms of the accuracy of image classification and the false alarm rate even when object detection accuracy is low.

• Journal of Preventive Medicine and Public Health
• /
• v.41 no.4
• /
• pp.209-213
• /
• 2008

Following the Anthrax bioterrorism attacks in the US in 2001, the Korean government established comprehensive countermeasures against bioterrorism. These measures included the government assuming management of all infectious agents that cause diseases, including smallpox, anthrax, plaque, botulism, and the causative agents of viral hemorrhagic fevers (ebola fever, marburg fever, and lassa fever) for national security. In addition, the Korean government is reinforcing the ability to prepare and respond to bioterrorism. Some of the measures being implemented include revising the laws and guidelines that apply to the use of infectious agents, the construction and operation of dual surveillance systems for bioterrorism, stockpiling and managing products necessary to respond to an emergency (smallpox vaccine, antibiotics, etc.) and vigorously training emergency room staff and heath workers to ensure they can respond appropriately. In addition, the government's measures include improved public relations, building and maintaining international cooperation, and developing new vaccines and drugs for treatments of infectious agents used to create bioweapons.

• Journal of Preventive Medicine and Public Health
• /
• v.55 no.3
• /
• pp.253-262
• /
• 2022

We aimed to review the current data composition of the Korean Tuberculosis and Post-Tuberculosis Cohort, which was constructed by linking the Korean Tuberculosis Surveillance System (KNTSS; established and operated by the Korean Disease Control and Prevention Agency since 2000) and the National Health Information Database (NHID; established by the National Health Insurance Service in 2012). The following data were linked: KNTSS data pertaining to patients diagnosed with tuberculosis between 2011 and 2018, NHID data of patients with a history of tuberculosis and related diseases between 2006 and 2018, and data (obtained from the Statistics Korea database) on causes of death. Data from 300 117 tuberculosis patients (177 206 men and 122 911 women) were linked. The rate of treatment success for new cases was highest in 2015 (86.7%), with a gradual decrease thereafter. The treatment success rate for previously treated cases showed an increasing trend until 2014 (79.0%) and decreased thereafter. In total, 53 906 deaths were confirmed among tuberculosis patients included in the cohort. The Korean Tuberculosis and Post-Tuberculosis Cohort can be used to analyze different measurement variables in an integrated manner depending on the data source. Therefore, these cohort data can be used in future epidemiological studies and research on policy-effect analysis, treatment outcome analysis, and health-related behaviors such as treatment discontinuation.

• Atmosphere
• /
• v.20 no.2
• /
• pp.153-160
• /
• 2010

In this paper, an attempt has been made to build and test self-configuring weather sensor networks and internet based observation system to gather atmospheric data. The aim is to provide integrated or real-time weather information in standard form using network data access protocol. This system was successfully developed to record weather information both digital as well as visual using sensor network and web-enabled surveillance cameras. These data were transformed by network based data access protocol to access and utilize for public domain. The competed system has been successfully utilized to monitor different types of weather. The results show that this is one of the most useful weather monitoring system.

• Journal of Digital Convergence
• /
• v.13 no.8
• /
• pp.473-481
• /
• 2015

The flood and environment surveillance network on riverside is a network requiring a way to efficiently manage the information from all kinds of sensors, along with an optical communication device that can deliver high-quality video information at high speed. Since on-site prompt recovery is very important especially for communication problems that occurred due to cut-off or aged network, various researches have been carried out on this. However, because the security against outside hacking or outside intrusion with illegal purposes is very important for environment surveillance network, such as the national backbone network, an efficient network maintenance and repair should be enabled while satisfying security and reliability at the same time. A characteristic of requirement is that when security is improved, the efficiency of maintenance and repair drops as they are conflicting to each other. Therefore, this research proposed a system in order to satisfy the conflicting requirement and improve security, by developing a one-way Bypass TAP and an android-based smartphone app that can enable efficient network maintenance and repair.

• Journal of Veterinary Clinics
• /
• v.34 no.2
• /
• pp.146-151
• /
• 2017

Six major outbreaks of highly pathogenic avian influenza (HPAI) occurred from 2003 to 2016 in Korea. Epidemiological investigations of each outbreak revealed that migratory birds were the primary source of the HPAI virus. During the last five years, the geographic transmission pattern of domestic HPAI seems to have extended from local to nationwide; therefore, it is necessary to identify specific locations in which poultry farms are at elevated risk for HPAI outbreak to enable targeted surveillance and other mitigation strategies. Here, a geographical information system (GIS)-based analysis was used to identify geographic areas at high risk for future HPAI incidents in Korea based on historical outbreak data collected between December 2003 and April 2016. To accomplish this, seven criteria were used to identify areas at high-risk for HPAI occurrence. The first three criteria were based on defined spatial criteria buffering of 200 bird migration sites to some defined extents and the historical incidence of HPAI outbreaks at the buffering sites. The remaining criteria were based on combined attribute information such as number of birds or farms at district levels. Based on the criteria established for this study, the most-likely areas at higher risk for HPAI outbreak were located in Chungcheong, Jeolla, Gyeonggi, and Gyeongnam provinces, which are densely populated poultry regions considered major poultry-production areas that are located along bird migration sites. The proportion of areas at risk for HPAI occurrence ranged from 4.5% to 64.9%. For the worst criteria, all nine provinces, including Jeju Island, were found to be at risk of HPAI. The results of this study indicate that the number of poultry farms at risk for HPAI outbreaks is largely underestimated by current regulatory risk assessment procedures conducted for biosecurity authorization. The HPAI risk map generated in this study will enable easy use of information by policy makers to identify surveillance zones and employ targeted surveillance to reduce the impact of HPAI transmission.

• Journal of information and communication convergence engineering
• /
• v.12 no.3
• /
• pp.168-172
• /
• 2014

In this paper, we propose an algorithm that can efficiently detect and monitor multiple ships in real-time. The proposed algorithm uses morphological operations and edge information for detecting and tracking ships. We used smoothing filter with a $3{\times}3$ Gaussian window and luminance component instead of RGB components in the captured image. Additionally, we applied Sobel operator for edge detection and a threshold for binary images. Finally, object labeling with connectivity and morphological operation with open and erosion were used for ship detection. Compared with conventional methods, the proposed method is meant to be used mainly in coastal surveillance systems and monitoring systems of harbors. A system based on this method was tested for both stationary and non-stationary backgrounds, and the results of the detection and tracking rates were more than 97% on average. Thousands of image frames and 20 different video sequences in both online and offline modes were tested, and an overall detection rate of 97.6% was achieved.

• Journal of the Korean Society of Safety
• /
• v.31 no.6
• /
• pp.84-92
• /
• 2016

Since the mankind started its space mission, the number of artificial space objects has been increasing exponentially. It contains not just the space machines which are in use but the machines which are out of order. Meantime, those dead machines are being a serious danger, a real threat to human's lives and property because of it could re-enter into the earth's atmosphere and fall to the ground causing mega-disaster. As the number of space activities gets growing so far, the re-entry of the space objects will be a lot more happened in the future. Therefore, not just natural space object like asteroids but the artificial space object like artificial satellite and space station can cause the disaster by falling to the ground. To protect our nation and our property, the government has set up the space disaster management center in Korea astronomy and Space science Institute. In this study, we surveyed public's recognition of the space object's re-entry situation and analyzed it to contribute building national space disaster management policy.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.124-125
• /
• 2017

레이더 센서보다 가격이 저렴한 초음파 센서를 사용하여 센서 운용에 있어 가격을 줄였다. 후측면 사각감시 시스템이 경보를 못하거나 오 경보할 수 있는 상황을 후방 카메라의 객체 인식, 탐지를 통해 초음파 센서가 미치지 못하는 영역에 대해 감지를 한다. 또한 후방 카메라와 후측면 사각감시 시스템이 ECU 내부의 통신을 하여 오류를 줄인 정보를 운전자에게 전달 할 수 있다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.139-140
• /
• 2015

본 논문에서는 교통 감시 시스템에서 차량추적방법을 제안한다. 교통 감시 카메라를 이용한 차량추적시스템은 차량 감시, 사고감지 및 교통정보를 확인할 수 있게 하는 시스템이다. 차량추적을 위하여 먼저 가우스 혼합 모델(Gaussian Mixture Model)을 이용하여 배경과 전경을 분리하고 형태학적 필터링을 이용하여 차량을 검출한다. 검출된 차량으로부터 SURF(Speed Up Robust Features) 매칭을 통하여 차량추적방법을 제안한다.