• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.3
• /
• pp.193-198
• /
• 2016

Lane change is an important issue in microscopic traffic flow simulations and active safety. Overtaking and changing lanes are dangerous driving maneuvers. This approach presents a lane-changing system based on speed and a minimum gap between vehicles in a vehicular ad hoc network (VANET). This paper proposes a solution to ensure the safety of drivers while changing lanes on highways. Efficient routing protocols could play a crucial role in VANET applications, safeguarding both drivers and passengers, and thus, maintaining a safe on-road environment. This paper focuses on the development of an intelligent transportation system that provides timely, reliable information to drivers and the concerned authorities. A test bed is created for the techniques used in the proposed system, where analysis takes place in an on-board embedded system designed for vehicle navigation. The designed system was tested on a four-lane road in Neemrana, India. Successful simulations were conducted with real-time network parameters to maximize quality of service and performance using Simulation of Urban Mobility and Network Simulator 2 (NS-2). The system implementation, together with the findings, is presented in this paper. Illustrating the approach are results from simulation using NS-2.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.2
• /
• pp.103-110
• /
• 2012

The population growth along with the urbanization, has caused more problems in many public areas, such as subway airport terminals, hospital, etc. Many surveillance systems have been installed in the public areas, but not all of those can be monitored in real-time, because the operators that observe the monitors are very small compared with the number of the monitors. For example, the observer can miss some crucial accidents or detect after considerable delays. Thus, intelligent surveillance system for preventing the accidents are needed, such as Intelligent Surveillance Systems. in this paper, we propose a new crowd density estimation method which aims at estimating moving crowd using images from surveillance cameras situated in outdoor locations. The moving crowd is estimated from the area where using optical flow. The edge information is also used as feature to measure the crowd density, so we improve the accuracy of estimation of crowd density. A multilayer neural network is designed to classify crowd density into 5 classes. Finally the proposed method is experimented with PETS 2009 images.

• The Korean Journal of Physiology and Pharmacology
• /
• v.24 no.1
• /
• pp.89-99
• /
• 2020

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.

• Journal of the Korean Society of Costume
• /
• v.67 no.3
• /
• pp.115-132
• /
• 2017

This study recognizes the importance of the social network platform as a new fashion media, and analyzes the significance of various digital fashion images, based on the 'Aura' theory of Walter Benjamin. The concept of "Disappearance of Artistic Aura" can be summarized into three discussions: 1) the change in the way of artistic perception, which is changes in value from worship to exhibition. 2) the change in the way of artistic acceptance, from personal to mass. 3) the emergence of new artistic concepts such as camera and film. By reviewing characteristics of the $21^{st}$ digital replication era, the study tried to discover and evaluate the expanded significance of the 'Aura' represented on digital fashion images, which are infinitely generated, modified, reproduced, transmitted, and shared in social network environments. The 'Burberry Instagram' was chosen as the subject of the study. The study reviewed around 2,500 images, which were uploaded from February 2011 to July 2016, and selected 200 images deemed the most representative of Burberry, and categorized and analyzed by the extended concept of 'Aura'. The study results as follows: First, the 'Aura' in digital fashion image appearing on social network platforms signifies the expansion of product value in fashion, and it also represents inherited traditions and modernization of images. Second, it also signifies the democratization and globalization of fashion through the open replication and sharing as well as the interaction of criticism and acceptance. Third, it signifies the personalized taste and fashion as everyday lifestyle, through personalized services, securing playful space, and real-time updates.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.6
• /
• pp.1121-1126
• /
• 2017

Recently, the importance of smart gateways that link these with the big data and the development of the Internet of things is getting bigger. The smart gateway includes a network function such as a router and a router, and a sensor network function that links various objects such as a sensor. As the internet market has expanded, network stability and security problems have arisen and VPN technology has been proposed as one of the ways to solve these security problems. Efficient design is needed to implement VPN in low-end smart gateway and SOHO-level Internet environment with poor line quality. In this paper, we propose the concept and principle of VPN tunneling implementation and real - time traffic shaping method according to internet line condition in the Smart Gateway that supports IOT developed based on OpenWRT, the implementation and measured performance indicators are presented.

• Journal of Satellite, Information and Communications
• /
• v.12 no.3
• /
• pp.54-62
• /
• 2017

China has invested for military satellite technology development to construct the space-based surveillance system from existing land-based and aerostat surveillance system since 1960s to react rapidly for deployment of marine force of United States and surrounding nations in west Pacific, south China sea and Indian ocean. China has also launched about 40 the Yaogan military intelligence satellites series for EO, SAR and ELINT fields since 2006 after the required technique with several technical experiment satellites launch and operational test. ELINT satellites transmit data from satellite to earth station in real time with construction space-based network around it. Those data are simultaneously delivered to Anti-Ship Ballistic Missile(ASBM) connected land-based C4ISR network for marine target attack. Therefore China has enhanced surveillance and attack capability to the surrounding marine nations with space-based network around it. In the future, It is considered that China will increase accurate location search, signal processing and analysis ability through a further study on its technology.

• Journal of the Korea Convergence Society
• /
• v.8 no.2
• /
• pp.15-20
• /
• 2017

In this paper, Esp8266, Node.js, and TCP / IP socket communication are used to design a compact data integration device. This device is designed to configure server and client using Esp8266 module that supports Wifi connection and to support bidirectional data transmission using TCP / IP socket communication. The server is configured using the Node.js operating system, and the database is integrated using Mysql. The network is designed to have a separate IP address by assigning a private IP address to the router, such as a home network. This device can transmit data bidirectionally, store individual client data on the server side, and can check the flow of data transmitted bidirectionally through wire-shark, so that it can be used as a compact real-time data integration and convergence device.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.891-902
• /
• 2020

The Re-Identification(Re-ID) is one of the most popular researches in the field of computer vision due to a variety of applications. To achieve a high-level re-identification performance, recently other methods have developed the deep learning based networks that are specialized for only person or vehicle. However, most of the current methods are difficult to be used in real-world applications that require re-identification of both person and vehicle at the same time. To overcome this limitation, this paper proposes a deep neural network learning method that combines triplet and softmax loss to improve performance and re-identify people and vehicles simultaneously. It's possible to learn the detailed difference between the identities(IDs) by combining the softmax loss with the triplet loss. In addition, weights are devised to avoid bias in one-side loss when combining. We used Market-1501 and DukeMTMC-reID datasets, which are frequently used to evaluate person re-identification experiments. Moreover, the vehicle re-identification experiment was evaluated by using VeRi-776 and VehicleID datasets. Since the proposed method does not designed for a neural network specialized for a specific object, it can re-identify simultaneously both person and vehicle. To demonstrate this, an experiment was performed by using a person and vehicle re-identification dataset together.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.8 no.1
• /
• pp.165-172
• /
• 1988

An optimal network design method using continuous form of design variables is considered. Modified Hooke-and-Jeeves algorithm has been implemented in order to solve nonlinear progamming problem which is approximately equivalent to the real network design problem (NDP) with system. efficiency criteria(i. e. travel time and costs) and construction cost as objective function. Various forms of construction cost function, locations of initial solution, and dimension of initial step size of link improvement are taken into account to show the validity of this approach. The results obtained are quite promising in terms of the numbers of evaluations in solving NDP, and the speed of convergence. Finally, some techniques in choosing efficient intial solution, initial step size and approximation are given.

• Journal of Internet Computing and Services
• /
• v.17 no.4
• /
• pp.43-50
• /
• 2016

In this paper, we propose the protocol of Ethernet that will receive a popular interesting in the automotive intelligent network, it also attempts to implementation and verification through simulation and experiments to propose a fault tolerance algorithm when the data transfer on it. It has proven the usefulness of the system in order to apply toward an existing automotive communication system. In the case of actual real-time data for automotive industry, we generated a randomly-generated data which is the set of payload into a standard format to complete the experiment. Among the implemented existing algorithms performance, we confirmed the effectiveness of all range from a single data to mixed (Hybrid-type) data, to verify the proposed algorithm.