• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Journal of the Society of Disaster Information
• /
• v.11 no.4
• /
• pp.556-569
• /
• 2015

With about 75% of the population of Korea criticizing the government's disaster policy and a failure to respond to large-scale emergency like the Sewol ferry sinking means that there is a deep distrust in the government. In order to prevent dreadful disasters such as the Sewol ferry sinking, it is important to secure a prime time with respect to disaster safety. Improving crisis management skills and managerial role of police officers who are in close proximity to the people is necessary for the success of disaster management. With disaster management as one of the most essential missions of the police, as a part of a national crisis management, a step by step strengthening of the disaster safety management system of the police is necessary, as below. First, at the prevention phase, law enforcement officers were not injected into for profit large-scale assemblies or events, but in the future the involvement, injection should be based on the level of potential risk, rather than profitability. In the past and now, the priortiy was the priority was on traffic flow, traffic communication, however, the paradigm of traffic policy should be changed to a safety-centered policy. To prevent large-scale accidents, police investigators should root out improper routines and illegal construction subcontracting. The police (intelligence) should strengthen efforts to collect intelligence under the subject of "safety". Second, with respect to the preparatory phase, on a survey of police officers, the result showed that 72% of police officers responded that safety management was not related to the job descriptions of the police. This, along with other results, shows that the awareness of disaster safety must be adopted by, or rather changed in the police urgently. The training in disaster safety education should be strengthened. A network of experts (private, administrative, and police) in safety management should be established to take advantage of private resources with regard to crisis situtions. Third, with respect to the response phase, for rapid first responses to occur, a unified communication network should be established, and a real-time video information network should be adopted by the police and installed in the police situation room. Fourth, during the recovery phase, recovery teams should be injected, added and operated to minimize secondary damage.

• Journal of Digital Convergence
• /
• v.15 no.11
• /
• pp.261-269
• /
• 2017

The purpose of this paper is to provide a search system using a method of accessing the web in real time without using a database server in order to guarantee the up-to-date information in a single network, rather than using a plurality of bots connected by a wide area network Design. The method of the research is to design and analyze the system which can search the person and keyword quickly and accurately in crepe system. In the crepe server, when the user registers information, the body tag matching conversion process stores all the information as it is, since various styles are applied to each user, such as a font, a font size, and a color. The crepe server does not cause a problem of body tag matching. However, when executing the crepe retrieval system, the style and characteristics of users can not be formalized. This problem can be solved by using the html_img_parser function and the Go language html parser package. By applying queues and multiple threads to a general-purpose web crawler, rather than a web crawler design that targets a specific site, it is possible to utilize a multiplier that quickly and efficiently searches and collects various web sites in various applications.

• Journal of Korean Society of Disaster and Security
• /
• v.16 no.1
• /
• pp.105-113
• /
• 2023

Jeollanam-do has 25 land-to-island and island-to-island bridges, the largest number in Korea. It is a local government rich in specialized marine and tourism resources centered on the archipelago and the sea bridges connecting them. However, in the case of sea-crossing bridges, when strong winds or typhoons occur, there is an issue that increases anxiety among users and local residents due to excessive vibration of the bridge, apart from structural safety of the bridge. In fact, in the case of Cheonsa Bridge in Shinan-gun, which was recently opened in 2019, vehicle traffic restrictions due to strong winds and excessive vibrations frequently occurred, resulting in complaints from local residents and drivers due to increased anxiety. Therefore, based on the data measured using IoT measurement technology, it is possible to relieve local residents' anxiety about the safety management of marine bridges by providing quantitative and accurate bridge vibration levels related to traffic and wind conditions of bridges in real time to local residents. This study uses the existing measurement system and IoT sensor to constantly observe the wind speed and vibration of the marine bridge, and transmits it to local residents and managers to relieve anxiety about the safety and traffic of the sea-crossing bridge, and strong winds and to develop technologies capable of preemptively responding to large-scale disasters.

• Knowledge Management Research
• /
• v.19 no.1
• /
• pp.59-77
• /
• 2018

The tourism industry is now changing to smart tourism, which maximizes tourists' overall tourism experience with the use of advanced mobile technologies and emphasizes the utilization of tourism information. Despite the quantitative expansion of the tourism industry, there is a lack of academic and practical discussion on tourism safety. Especially, in the context of walking tourism, tourists are more likely to be exposed to natural or social disasters and emergencies. Therefore, it is necessary to build a system that can provide walking tourists with safety information not only on dangerous factors which are anticipated to be confronted during a walking trip in advance but also on specific dangers in real time. Under the circumstances, this study seeks to identify the types of tourism safety information that can be offered by using publicly available open data, drawing on the safety information framework on the walking tourism that is presented in Choi et al. (2017)'s study. More specifically, this study focuses on the use of open data which is provided by the Korean government. Furthermore, this study verifies the types of safety information that are most urgently needed in walking travel situations. Specifically, this study aims to derive the importance and priority of each type of safety information for a walking trip by applying the analytic hierarchy process (AHP) analysis. For this, we collected 35 questionnaires from walking tour operators (practitioners) and walking tourists. The main results are as follows. First, natural disaster information is the most important factor in the top-level factor of safety information for walking tourists, followed by social disaster, life safety, and exhibition (security crisis) information. Second, information on natural disasters, environmental pollution, and weather is considered to be important at the sub-level factor. Lastly, the noteworthy result of this study is that the importance of each type of safety information varies depending on the walking tour operators (practitioners) and the walking tourists. That is, there is a recognition difference between the operator (practitioner) and the user in the importance and priority of the safety information of the walking trip. Therefore, it is necessary to develop policies and services reflecting the opinions of potential users when providing safety information so that the most importantly recognized information can be provided first.

• Journal of the Korean Society for information Management
• /
• v.38 no.3
• /
• pp.23-39
• /
• 2021

This study aims to explore research trends in Blockchain studies in South Korea using dynamic topic modeling and network analysis. To achieve this goal, we conducted the university & institute collaboration network analysis, the keyword co-occurrence network analysis, and times series topic analysis using dynamic topic modeling. Through the university & institute collaboration network analysis, we found major universities such as Soongsil University, Soonchunhyang University, Korea University, Korea Advanced Institute of Science and Technology (KAIST) and major institutes such as Ministry of National Defense, Korea Railroad Research Institute, Samil PricewaterhouseCoopers, Electronics and Telecommunications Research Institute that led collaborative research. Next, through the analysis of the keyword co-occurrence network, we found major research keywords including virtual assets (Cryptocurrency, Bitcoin, Ethereum, Virtual currency), blockchain technology (Distributed ledger, Distributed ledger technology), finance (Smart contract), and information security (Security, privacy, Personal information). Smart contracts showed the highest scores in all network centrality measures showing its importance in the field. Finally, through the time series topic analysis, we identified five major topics including blockchain technology, blockchain ecosystem, blockchain application 1 (trade, online voting, real estate), blockchain application 2 (food, tourism, distribution, media), and blockchain application 3 (economy, finance). Changes of topics were also investigated by exploring proportions of representative keywords for each topic. The study is the first of its kind to attempt to conduct university & institute collaboration networks analysis and dynamic topic modeling-based times series topic analysis for exploring research trends in Blockchain studies in South Korea. Our results can be used by government agencies, universities, and research institutes to develop effective strategies of promoting university & institutes collaboration and interdisciplinary research in the field.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.157-175
• /
• 2013

Since Korean government announced 'Smartwork promotion strategy' in 2010, Korean firms and government organizations have started to adopt smartwork. However, the smartwork has been implemented only in a few of large enterprises and government organizations rather than SMEs (small and medium enterprises). In USA, both Yahoo! and Best Buy have stopped their flexible work because of its reported low productivity and job loafing problems. In addition, according to the literature on smartwork, we could draw obstacles of smartwork adoption and categorize them into the three types: institutional, organizational, and technological. The first category of smartwork adoption obstacles, institutional, include the difficulties of smartwork performance evaluation metrics, the lack of readiness of organizational processes, limitation of smartwork types and models, lack of employee participation in smartwork adoption procedure, high cost of building smartwork system, and insufficiency of government support. The second category, organizational, includes limitation of the organization hierarchy, wrong perception of employees and employers, a difficulty in close collaboration, low productivity with remote coworkers, insufficient understanding on remote working, and lack of training about smartwork. The third category, technological, obstacles include security concern of mobile work, lack of specialized solution, and lack of adoption and operation know-how. To overcome the current problems of smartwork in reality and the reported obstacles in literature, we suggest a novel smartwork service model based on NFC(Near Field Communication). This paper suggests NFC-based Smartwork Service Model composed of NFC-based Smartworker networking service and NFC-based Smartwork space management service. NFC-based smartworker networking service is comprised of NFC-based communication/SNS service and NFC-based recruiting/job seeking service. NFC-based communication/SNS Service Model supplements the key shortcomings that existing smartwork service model has. By connecting to existing legacy system of a company through NFC tags and systems, the low productivity and the difficulty of collaboration and attendance management can be overcome since managers can get work processing information, work time information and work space information of employees and employees can do real-time communication with coworkers and get location information of coworkers. Shortly, this service model has features such as affordable system cost, provision of location-based information, and possibility of knowledge accumulation. NFC-based recruiting/job-seeking service provides new value by linking NFC tag service and sharing economy sites. This service model has features such as easiness of service attachment and removal, efficient space-based work provision, easy search of location-based recruiting/job-seeking information, and system flexibility. This service model combines advantages of sharing economy sites with the advantages of NFC. By cooperation with sharing economy sites, the model can provide recruiters with human resource who finds not only long-term works but also short-term works. Additionally, SMEs (Small Medium-sized Enterprises) can easily find job seeker by attaching NFC tags to any spaces at which human resource with qualification may be located. In short, this service model helps efficient human resource distribution by providing location of job hunters and job applicants. NFC-based smartwork space management service can promote smartwork by linking NFC tags attached to the work space and existing smartwork system. This service has features such as low cost, provision of indoor and outdoor location information, and customized service. In particular, this model can help small company adopt smartwork system because it is light-weight system and cost-effective compared to existing smartwork system. This paper proposes the scenarios of the service models, the roles and incentives of the participants, and the comparative analysis. The superiority of NFC-based smartwork service model is shown by comparing and analyzing the new service models and the existing service models. The service model can expand scope of enterprises and organizations that adopt smartwork and expand the scope of employees that take advantages of smartwork.

• Analytical Science and Technology
• /
• v.13 no.5
• /
• pp.608-615
• /
• 2000

A determination method of trace nickel and cobalt in water samples was studied and developed by adsorbing their complexes on ion exchange resin suspension. The analytical ions were formed as complexes with a ligand of APDC (ammonium pyrrolidinedithiocarbamate) and adsorbed on anion exchange resin of Dowex 2-X8. After the suspension was filtered out with membrane filter, the complexes were dissolved in HCl solution by an ultrasonic vibrator for ET-AAS determination. Several conditions were optimized as followings. pH of sample solution: 5.0, amount of ligand APDC: more than 430 times in mole ratio, the type and concentration of acid: 0.1 M HCl, and vibration time: 7 minutes. The addition of palladium in the HCl solution could improve the reproducibility and sensitivity by a matrix modification in the absorbance measurement. This procedure was applied for the analysis of three kinds of real water samples. The detection limits equivalent to 3 times standard deviation of blank were Co 0.36 ng/mL and Ni 0.27 ng/mL and recoveries in spiked samples were 99-102% for cobalt and 100-105% for nickel.

• The Korean Journal of Air & Space Law and Policy
• /
• v.30 no.2
• /
• pp.83-153
• /
• 2015

As our economy grows and the number of aircraft increase, the number of civil execution against aircraft cases are likely to increase as well in the future. The purpose of this article is to present the legal theory on the civil execution against aircrafts by drawing on the legal theory on the civil execution against vessels which constitute a relatively large number of cases thus observed. The provisions of the civil execution against immovables or vessel, shall basically apply mutatis mutandis to the civil execution against aircraft or light aircraft. The civil execution against ultra-light flying devices or a foreign aircraft shall be executed in conformity with the civil execution against movables. There are a compulsory auction, an auction to execute a security right to aircraft, and an auction under the right of retention, etc. in the civil execution against an aircraft. A compulsory execution against an aircraft means an execution carried out by a creditor against a debtor's aircraft to obtain satisfaction of claims for the purpose of payment of money. The court of execution of a compulsory execution against an aircraft shall be the district court having jurisdiction over the airport of stoppage or storage of such aircraft at the time of seizure. The forums of execution of a compulsory execution against an aircraft shall be exclusive forums. When a court has rendered an order on commencing an auction, it shall order an execution officer to receive a certificate of the aircraft's registration and other documents as required for its operation, and to submit them to the court. A court may revoke the procedures for a compulsory auction when an execution officer fails to obtain a transfer of the aircraft's registration certificate, etc. and the location of the aircraft is not evident, not later than an elapse of 2 months from the date on which an order on commencing an auction has been rendered. In the case where it is deemed that there exists a business-related need or other based on proper reasoning, the court may permit the aircraft's operation, upon the motion submitted by the debtor. In this case, there shall be a consent from the creditor, the highest bidder, the next highest bidder and successful bidder. A court may, upon a motion submitted by the creditor, make the dispositions required for observing and preserving the aircraft. When a debtor has submitted the documents under subparagraph 2 or 4 of the Article 49 of the Civil Execution Act, and furnished the guarantee equivalent to the claims of the execution creditors and the creditors demanding a distribution and to the costs for execution, before a declaration of bid, the court shall, upon request, revoke other procedures than those for distribution. The provisions of a obligatory auction against vessel or aircraft and an auction to execute a security right to real estate or vessel, shall apply mutatis mutandis to an auction to execute the security right to aircraft. In an auction to execute the security right to aircraft case, an executive title is not necessary. An executory exemplification is not necessary in an application for an auction to execute the security right to aircraft. A court should examine the existence of security right and claim secured. No order on commencing an auction procedure shall be issued with non-existence or invalidity of the security right and absence or extinguishment of the claim secured. Furthermore, these prohibitions are the reason of a decision on non-permit for sale, the court overlooked these prohibitions, and the decision on a permit for sale became final and conclusive, the successful bidder who paid the price and registered of ownership could not acquire ownership of the aircraft sold. A court may render a ruling to put plural aircrafts up for a blanket auction, only when they are in restraint and related matter (Supreme Court Order 2001Ma3688 dated on August 22, 2001). A righter of retention on aircraft may file a request for an auction against the aircraft. The provisions of an auction to execute a security right to aircraft shall apply mutatis mutandis to the formal auction. Airport facility fee and an aircraft are not in restraint and related matter, so an airport management corporation does not hold the right of retention on the aircraft (Supreme Court Decision 2011Da29291 decided on April 10, 2014). In an auction in accordance with the right of retention, all encumbrances (e.g., mortgages) on the sold aircraft shall be extinguished by a sale under the legal conditions for sale. Not only creditors who have claims for preferential payment but also general creditors could demand for distribution. The precedence of the claim of the right of retention on aircraft and that of general creditor's claims are equal.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.125-141
• /
• 2012

These days, the malicious attacks and hacks on the networked systems are dramatically increasing, and the patterns of them are changing rapidly. Consequently, it becomes more important to appropriately handle these malicious attacks and hacks, and there exist sufficient interests and demand in effective network security systems just like intrusion detection systems. Intrusion detection systems are the network security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. Conventional intrusion detection systems have generally been designed using the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. However, they cannot handle new or unknown patterns of the network attacks, although they perform very well under the normal situation. As a result, recent studies on intrusion detection systems use artificial intelligence techniques, which can proactively respond to the unknown threats. For a long time, researchers have adopted and tested various kinds of artificial intelligence techniques such as artificial neural networks, decision trees, and support vector machines to detect intrusions on the network. However, most of them have just applied these techniques singularly, even though combining the techniques may lead to better detection. With this reason, we propose a new integrated model for intrusion detection. Our model is designed to combine prediction results of four different binary classification models-logistic regression (LOGIT), decision trees (DT), artificial neural networks (ANN), and support vector machines (SVM), which may be complementary to each other. As a tool for finding optimal combining weights, genetic algorithms (GA) are used. Our proposed model is designed to be built in two steps. At the first step, the optimal integration model whose prediction error (i.e. erroneous classification rate) is the least is generated. After that, in the second step, it explores the optimal classification threshold for determining intrusions, which minimizes the total misclassification cost. To calculate the total misclassification cost of intrusion detection system, we need to understand its asymmetric error cost scheme. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, total misclassification cost is more affected by FNE rather than FPE. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 10,000 samples from them by using random sampling method. Also, we compared the results from our model with the results from single techniques to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell R4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on GA outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that the proposed model outperformed all the other comparative models in the total misclassification cost perspective. Consequently, it is expected that our study may contribute to build cost-effective intelligent intrusion detection systems.