• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.6
• /
• pp.60-68
• /
• 2015

Recent proliferation of advanced technologies such as wireless communication, mobile, sensor technology and so on has caused significant changes in a traffic environment. Human beings, in particular drivers, as well as roads and vehicles were advanced on information, intelligence and automation thanks to those advanced technologies; Intelligent Transport Systems (ITS) and autonomous vehicles are the results of changes in a traffic environment. This study proposed considerations when designing a simulation model for future transportation environments, which are difficult to predict the change by means of advanced technologies. First of all, approximability, flexibility and scalability were defined as a macroscopic concept for a simulation model design. For actual similarity, calibration is one of the most important steps in simulation, and Physical layer and MAC layer should be considered for the implementation of the communication characteristics. Interface, such as API, for inserting the additional models of future traffic environments should be considered. A flexible design based on compatibility is more important rather than a massive structure with inherent many functions. Distributed computing with optimized H/W and S/W together is required for experimental scale. The results of this study are expected to be used to the design of future traffic simulation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.15-23
• /
• 2017

The person Re-identification is the most challenging part of computer vision due to the significant changes in human pose and background clutter with occlusions. The picture from non-overlapping cameras enhance the difficulty to distinguish some person from the other. To reach a better performance match, most methods use feature selection and distance metrics separately to get discriminative representations and proper distance to describe the similarity between person and kind of ignoring some significant features. This situation has encouraged us to consider a novel method to deal with this problem. In this paper, we proposed an enhanced recurrent neural network with three-tier hierarchical network for person re-identification. Specifically, the proposed recurrent neural network (RNN) model contain an iterative expectation maximum (EM) algorithm and three-tier Hierarchical network to jointly learn both the discriminative features and metrics distance. The iterative EM algorithm can fully use of the feature extraction ability of convolutional neural network (CNN) which is in series before the RNN. By unsupervised learning, the EM framework can change the labels of the patches and train larger datasets. Through the three-tier hierarchical network, the convolutional neural network, recurrent network and pooling layer can jointly be a feature extractor to better train the network. The experimental result shows that comparing with other researchers' approaches in this field, this method also can get a competitive accuracy. The influence of different component of this method will be analyzed and evaluated in the future research.

• Journal of Intelligence and Information Systems
• /
• v.14 no.3
• /
• pp.59-76
• /
• 2008

In this paper, we propose an agent architecture called L-CAA that is quite effective in real-time dynamic environments. L-CAA is an extension of CAA, the behavior-based agent architecture which was also developed by our research group. In order to improve adaptability to the changing environment, it is extended by adding reinforcement learning capability. To obtain stable performance, however, behavior selection and execution in the L-CAA architecture do not entirely rely on learning. In L-CAA, learning is utilized merely as a complimentary means for behavior selection and execution. Behavior selection mechanism in this architecture consists of two phases. In the first phase, the behaviors are extracted from the behavior library by checking the user-defined applicable conditions and utility of each behavior. If multiple behaviors are extracted in the first phase, the single behavior is selected to execute in the help of reinforcement learning in the second phase. That is, the behavior with the highest expected reward is selected by comparing Q values of individual behaviors updated through reinforcement learning. L-CAA can monitor the maintainable conditions of the executing behavior and stop immediately the behavior when some of the conditions fail due to dynamic change of the environment. Additionally, L-CAA can suspend and then resume the current behavior whenever it encounters a higher utility behavior. In order to analyze effectiveness of the L-CAA architecture, we implement an L-CAA-enabled agent autonomously playing in an Unreal Tournament game that is a well-known dynamic virtual environment, and then conduct several experiments using it.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.6
• /
• pp.4021-4030
• /
• 2015

Pulse wave is the physiological responses through the autonomic nervous system such as ECG. It is relatively convenient because it can measure the signal just by applying a sensor on a finger. So, it can be usefully employed in the field of U-Healthcare. The objects of this study are acquiring the PPG (Photoplethysmography) one of the way of measuring the pulse waves in non-invasive way using the CMOS image sensor on a smartphone camera, developing the portable system judging stressful or not, and confirming the applicability in the field of u-Healthcare. PPG was acquired by using image data from smartphone camera without separate sensors and analyzed. Also, with that image signal data, HRV (Heart Rate Variability) and stress index were offered users by just using smartphone without separate host equipment. In addition, the reliability and accuracy of acquired data were improved by developing additional hardware device. From these experiments, we can confirm that measuring heart rate through the PPG, and the stress index for analysis the stress degree using the image of a smartphone camera are possible. In this study, we used a smartphone camera, not commercialized product or standardized sensor, so it has low resolution than those of using commercialized external sensor. However, despite this disadvantage, it can be usefully employed as the u-Healthcare device because it can obtain the promising data by developing additional external device for improvement reliability of result and optimization algorithm.

• Journal of Digital Convergence
• /
• v.19 no.9
• /
• pp.189-199
• /
• 2021

In order to increase the self-reliance rate of new and renewable energy in order to respond to the mandatory domestic zero-energy buildings, the taller the building, the more limited the site area, and installing PV modules on the roof is not enough. Therefore, BIPV (Building integrated photovoltaic, hereinafter BIPV) is the industry receiving the most attention as a core energy source that can realize zero-energy buildings. Therefore, this study conducted a survey on the problems of the BIPV industry in a self-discussing method for experts with more than 10 years of experience of designers, builders, product manufacturers, and maintainers in order to suggest the right direction and revitalize the BIPV industry. Industrial problems of BIPV adjustment are drawn extention range of standard and certification for products, range improvement for current small condition of various kind productions, need to revise standards for capable of accomodating roof-type, color-module and louver-module, necessary of barrier in flow of foreign modules into korea through domestic certification mandatory, difficulty in obtaining BIPV information, request to prevent confusion among participants by exact guidelime about architectural application part of BIPV, and lack of the BIPV definition clearness, support policy, etc. Based on the improvements needed for the elements, giving change and competitiveness impacts aims to present and propose counter measures and direction.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.2
• /
• pp.43-57
• /
• 2021

The Forward Collision Warning, a representative technology of the Advanced Driver Assistance Systems, was selected as the target technology. The cognitive response time, deceleration, and impact were selected as the measures of effectiveness. And the amount of change with and without the Forward Collision Warning was measured. The experimental scenarios included a sudden stop event (1) of the vehicle in front of the driver and an event (2) in which the vehicle intervened in the next lane. All experiments were divided into day and night. As a result of the analysis, response time and the deceleration rate decreased when the forward collision warning system was installed. It was analyzed that the driver's risk situation could be detected quickly and the number of front-end collisions could be reduced as a result. Reflecting the driver's operating habits and diversifying the experimental scenarios will increase the installation effectiveness of ADAS and be used to estimate the effectiveness of other technologies.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.3
• /
• pp.73-82
• /
• 2019

In order to build a smart factory, building a CPPS (Cyber Physical Product System) is an important system that must be accompanied. Through the CPPS, it is the reality of smart factories to move physical factories to a digital-based cyber world and to intelligently and autonomously monitor and control them. But The existing CPPS architectures present only an abstract modeling architecture, and the research that applied the OPC UA Framework (Open Platform Communication Unified Architecture), an international standard for data exchange in the smart factory, as the basic system of CPPS It was insufficient. Therefore, it is possible to implement CPPS that can include both cloud and IoT by collecting field data distributed by CPPS architecture applicable to actual factories and concentrating data processing in a centralized In this study, we implemented CPPS architecture through central OPC UA Server based on OPC UA conforming to central processing OPC UA Framework, and how CPPS logical process and data processing process are automatically generated through OPC UA modeling processing We have proposed the CPPS architecture including the model factory and implemented the model factory to study its performance and usability.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Management & Information Systems Review
• /
• v.38 no.3
• /
• pp.151-170
• /
• 2019

This study deals with the rent-seeking behavior that exist in cultural voucher from the viewpoint of culture and arts management. Art organizations open to consumers, producers and governments of the arts are dependent on the internal and external influence of an open system. Researcher has found rent-seeking in the course of introducing policies and legalization of the cultural voucher business which has been promoted in the direction of democratization of culture or cultural democracy. Cultural voucher business is a legal term. Although the government has increased the budget or tried to change the policies of the cultural voucher business, the implementation of the cultural voucher business has been opposed to the diversity of consumption and equity as the consumption of genre and the concentration of the capital region have increased. These results were structurally related to the process of legalization and rent-seeking behavior in bureaucracy. This study reaffirms that the efficient operation standard of the cultural voucher business is a balance between the choice of the beneficiary, the competition of the supplier, and access to the cultural voucher. And the theory of rent-seeking was applied as a criterion to analyze this balance. Thus, it is suggested that the criteria of evaluation and improvement to check the conservativeness of bureaucrats are needed to establish a legal system applied to the purpose of 'cultural democracy' and 'democratization of culture' ideology and to guarantee individual creativity and autonomy.

• Convergence Security Journal
• /
• v.21 no.1
• /
• pp.177-189
• /
• 2021

Defense R&D is a key process for securing weapons systems determined by mid- and long-term needs to cope with changing future battlefield environments. In particular, the test and evaluation provides information necessary to determine whether or not to switch to mass production as the last gateway to research and development of weapons systems and plays an important role in ensuring performance linked to the life cycle of weapons systems. Meanwhile, if you look at the recent changes in the operational environment of the Korean Peninsula and the defense acquisition environment, you can see three main characteristics. First of all, continuous safety accidents occurred during the operation of the weapon system, which increased social interest in the safety of combatants, and the efficient execution of the limited defense budget is required as acquisition costs increase. In addition, strategic approaches are needed to respond to future battlefield environments such as robots, autonomous weapons systems (RAS), and cyber security test and evaluation. Therefore, in this study, we would like to present strategies for improving the testing and evaluation of weapons systems by considering the characteristics of the security environment that has changed recently. To this end, the improvement strategy was derived by analyzing the complementary elements of the current weapon system operational test and evaluation system in a multi-dimensional model and prioritized through the hierarchical analysis method (AHP).