• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.3
• /
• pp.255-263
• /
• 2014

Today, a number of CCTV on the highway is to observe the flow of traffics. There have been a number of studies where traffic data (e.g., the speed of vehicles and the amount of traffic on the road) are transferred back to the centralized server so that an appropriate action can be taken. This paper introduces a system that detects the changes of traffic flows caused by an accident or unexpected stopping (i.e., vehicle remains idle) by monitoring each lane separately. The traffic flows of each lane are level spacing curve that shows Wigner distribution for location vector. Applying calogero-moser system and Hamiltonian system, probability equation for each level-spacing curve is derived. The high level of modification of the signal means that the lane is in accident situation. This is different from previous studies in that it does more than looking for the signal from only one lane, now it is able to detect an accident in entire flow of traffic. In process of monitoring traffic flow of each lane, when camera recognizes a shadow of vehicle as a vehicle, it will affect the accident detecting capability. To prevent this from happening, the study introduces how to get rid of such shadow. The system using Basian network method is being compared for capability evaluation of the system of the study. As a result, the system of the study appeared to be better in performance in detecting the modification of traffic flow caused by idle vehicle.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.2
• /
• pp.48-61
• /
• 2020

The objective of this study is to estimate traffic capacity based on the heavy-vehicle ratio in a two-lane freeway work zone where one lane is blocked by construction. For this, closed circuit television (CCTV) video data of the freeway work zone was collected, and the congestion at an upstream point was observed. The traffic volume at a downstream point was analyzed after a bottleneck was created by the blockage due to the upstream congestion. A distribution model was estimated using observed-time headway, and the road capacity was analyzed using a goodness-of-fit test. Through this process, the general capacity and an equation for capacity based on the heavy-vehicle ratio passing through the work zone were presented. Capacity was estimated to be 1,181~1,422 passenger cars per hour per lane (pcphpl) at Yeongdong, and 1,475~1,589pcphpl at Jungbu Naeryuk. As the ratio of heavy vehicles increased, capacity gradually decreased. These findings can contribute to the proper capacity estimation and efficient traffic operation and management for two-lane freeway work zones that block one lane due to a work zone.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.3
• /
• pp.451-457
• /
• 2018

We propose an automatic detection and classification system of sewer damage database based on artificial intelligence and deep learning. In order to optimize the performance, we implemented a robust system against various environmental variations such as illumination and shadow changes. In our proposed system, a crack detection and damage classification method using a deep learning based Convolutional Neural Network (CNN) is implemented. For optimal results, 9,941 CCTV images with $256{\times}256$ pixel resolution were used for machine learning on the damaged area based on the CNN model. As a result, the recognition rate of 98.76% was obtained. Total of 646 images of $720{\times}480$ pixel resolution were extracted from various sewage DB for performance evaluation. Proposed system presents the optimal recognition rate for the automatic detection and classification of damage in the sewer DB constructed in various environments.

• Journal of Korea Water Resources Association
• /
• v.46 no.5
• /
• pp.477-490
• /
• 2013

A rip-current warning index function, which is estimated from the likelihood of rip current quantified based on numerical simulations under various sea environments and is varied according to real-time buoy-observations, was studied to help protect against rip current accidents at Haeundae beach. For the quantification, the definition of likelihood of rip current, which proposed by Choi et al. (2011, 2012b), was employed and estimated based on Boussinesq modelling. The distribution of likelihood of rip current was evaluated by using various simulations according to scenarios established based on physical quantities(i.e., wave parameters) of buoy-observations. To index the likelihood of rip current, empirical functions were derived based on the distribution and adjusted to observational environments. In this study, the observations from June to September in 2011 at Haeundae beach were applied to the rip-current index functions, and its applications into the real events found based on CCTV images were presented and investigated. In addition, limitations and improvements of the rip-current index function were discussed.

• Journal of Korea Water Resources Association
• /
• v.54 no.12
• /
• pp.1233-1242
• /
• 2021

In this study, a model was developed to determine whether flooding occurred using image data, which is unstructured data. CNN-based VGG16 and VGG19 were used to develop the flood classification model. In order to develop a model, images of flooded and non-flooded images were collected using web crawling method. Since the data collected using the web crawling method contains noise data, data irrelevant to this study was primarily deleted, and secondly, the image size was changed to 224×224 for model application. In addition, image augmentation was performed by changing the angle of the image for diversity of image. Finally, learning was performed using 2,500 images of flooding and 2,500 images of non-flooding. As a result of model evaluation, the average classification performance of the model was found to be 97%. In the future, if the model developed through the results of this study is mounted on the CCTV control center system, it is judged that the respons against flood damage can be done quickly.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.1
• /
• pp.77-91
• /
• 2019

Recent Linux operating systems having been increasingly used, ranging from automotive consoles, CCTV, IoT devices, and mobile devices to various versions of the kernel. Because these devices can be used as strong evidence in criminal investigations, there is a risk of destroying evidence through file deletion. Ext filesystem forensics has been studied in depth because it can recovery deleted files without depending on the kind of device. However, studies have been carried out without consideration of characteristics of file system which may vary depending on the kernel. This problem can lead to serious situations, such as those that can impair investigative ability and cause doubt of evidence ability, when an actual investigation attempts to analyze a different version of the kernel. Because investigations can be performed on various distribution and kernel versions of Linux file systems at the actual investigation site, analysis of the metadata changes that occur when files are deleted by Linux distribution and kernel versions is required. Therefore, in this paper, we analyze the difference of metadata according to the Linux kernel as a solution to this and recovery deleted file. After that, the investigating agency needs to consider the metadata change caused by the difference of Linux kernel version when performing Ext filesystem forensics.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.1
• /
• pp.464-469
• /
• 2019

A remote monitoring panel and control system was developed to control various valves and access control chambers, including gas shutoff valves used in CBR(Chemical, Biological and Radiological) facilities. The remote monitoring panel consisted of a main panel installed in the NBC (Nuclear, Biological and Chemical) control room and auxiliary panel installed in the clean room, and the size was divided into pure control and control including CCTV. This system can be monitored and controlled remotely according to the situation where an explosion door and gas barrier door can occur during war and during normal times. This system is divided into normal mode and war mode. In particular, it periodically senses the operation status of various valves, sensors, and filters in the CBR facilities to determine if each apparatus and equipment is in normal operation, and remotely alerts situation workers when repair or replacement is necessary. Damage due to the abnormal operation of each device in the situation can be prevented. This enables control of the blower, supply and exhaust damper, emergency generator, and coolant pump according to the state of shutoff valve and positive pressure valve in the occurrence of NBC, and prevents damage caused by abrupt inflow of conventional weapons and nuclear explosions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.9
• /
• pp.504-509
• /
• 2019

Recently, both wireless communications technology and the performance of small devices have developed exponentially, while the number of services using various types of Internet of Things (IoT) devices has also massively increased in line with the ongoing technological and environmental changes. Furthermore, ever more devices that were previously used in the offline environment-including small-size sensors and CCTV-are being connected to the Internet due to the huge increase in IoT services. However, many IoT devices are not equipped with security functions, and use vulnerable open source software as it is. In addition, conventional network equipment, such as switches and gateways, operates with vulnerabilities, because users tend not to update the equipment on a regular basis. Recently, the simple vulnerability of IoT devices has been exploited through the distributed denial of service (DDoS) from attackers creating a large number of botnets. This paper proposes a system that is capable of identifying Internet-connected devices quickly, analyzing and managing the vulnerability of such devices using Internet-wide scan technology. In addition, the vulnerability analysis rate of the proposed technology was verified through collected banner information. In the future, the company plans to automate and upgrade the proposed system so that it can be used as a technology to prevent cyber attacks.

• Journal of the Society of Disaster Information
• /
• v.15 no.3
• /
• pp.358-366
• /
• 2019

Purpose: This paper aims to clarify the problems and to examine the improvement methods by investigating the management condition of local-designated cultural property of which management is relatively poor in comparison with state-designated cultural heritage. Method: In order to grasp the management situation of the local-designated cultural heritage, a research on cultural heritage management situation and problems will be carried out with 35 cultual heritages in Goryeong-gun. Also, the improvement methods about the property type vulnerability on the basis of interview with cultual property managers, fire-fighting officers and civil servants, etc. Results: Local cultural heritages were investigated to be very vulnerable to the fire of wooden buildings, the theft of movable cultural heritages, and the effects of wind and water damage. It is because cultural heritages are scattered over wide areas fundamentally. As the result, it has difficulty in the patrols of police officers and fire fighters, and in the situation that it lacks disaster monitoring and CCTV for countermeasures to replace them, electronic security including fire hydrant, sensors, etc and fire extinguishing facilities and so on. It is difficult for local governments managing local-designated cultural heritages to enhance their management systems directly due to their lack of budget and manpower. Conclusion: In order to strengthen disaster and safety management system for the cultural heritages designated by local governments, they have to clarify disaster countermeasure task of fire fighting, police, and cultural heritage managers prepare their manuals, and systematize them through disaster drill mainly in local autonomous governments. Also, so as to establish a surveillance system every day, they have to enhance the community for local cultural heritage manage consisting of local volunteer fire departments, local voluntary disaster prevention organizations, volunteers, etc.