• The KIPS Transactions:PartD
• /
• v.15D no.5
• /
• pp.609-620
• /
• 2008

The service handling the map data in the mobile device including navigation, LBS, Telematics, and etc., becomes various. The size of map data which is stored and managed in the mobile device is growing and reaches in several GB. The conventional navigation system has used the read-only PSF (physical storage format) in order to enhance the performance of system by maximum in the mobile device which has limited resources. So though a little part of the map data is changed the whole data must be updated. In general, it takes several ten minutes to write the 2 GB map data to a flash memory of mobile device. Therefore, we have developed the mobile spatial DBMS (database management system) to solve the problem which is that the partial map data couldn't be updated in the conventional navigation system. And we suggest the policy to guarantee the performance of the navigation system which is implemented using the spatial mobile DBMS and verify this by experiment.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.167-174
• /
• 2021

Recently, deep learning technology is widely used in various computer vision applications, such as object recognition, classification, and image generation. In particular, the deep learning-based super-resolution has been gaining significant performance improvement. Fast super-resolution convolutional neural network (FSRCNN) is a well-known model as a deep learning-based super-resolution algorithm that output image is generated by a deconvolutional layer. In this paper, we propose an FPGA-based convolutional neural networks accelerator that considers parallel computing efficiency. In addition, the proposed method proposes Optimal-FSRCNN, which is modified the structure of FSRCNN. The number of multipliers is compressed by 3.47 times compared to FSRCNN. Moreover, PSNR has similar performance to FSRCNN. We developed a real-time image processing technology that implements on FPGA.

• Journal of the Korea Convergence Society
• /
• v.12 no.7
• /
• pp.53-60
• /
• 2021

LOD-based recommender systems usually leverage the data available within LOD datasets, such as DBpedia, in order to recommend items(movies, books, music) to the end users. These systems use a semantic similarity algorithm that calculates the degree of matching between pairs of Linked Data resources. In this paper, we proposed a new approach to measuring semantic distance in an LOD-based recommender system by assigning weights converted from user ratings to links in the LOD graph. The semantic distance measurement model proposed in this paper is based on a processing step in which a graph is personalized to a user through weight calculation and a method of applying these weights to LDSD. The Experimental results showed that the proposed method showed higher accuracy compared to other similar methods, and it contributed to the improvement of similarity by expanding the range of semantic distance measurement of the recommender system. As future work, we aim to analyze the impact on the model using different methods of LOD-based similarity measurement.

• Journal of Advanced Navigation Technology
• /
• v.23 no.2
• /
• pp.158-165
• /
• 2019

The existing web page was downloaded and provided to the user every time the user requested the page. Therefore, if the same page is repeatedly requested by the user, only the download for the same resource is repeated. This is a factor that causes unnecessary consumption of data. We focus on reducing data consumption caused by unnecessary requests between users and servers, and improving content delivery speed. Therefore, in this paper, we propose a caching system and an algorithm that can reduce the data consumption while maintaining the latest cache by comparing the hash value using the hash function that can detect the change of the file requested by the user.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.10
• /
• pp.31-41
• /
• 2009

In the Semantic Web, it is possible to provide intelligent information retrieval and automated web services by defining a concept of information resource and representing a semantic relation between resources with meta data and ontology. It is very important to manage semantic data such as ontology and meta data efficiently for implementing essential functions of the Semantic Web. Thus we propose an index structure to support more accurate search results and efficient query processing by considering semantic and structural features of the semantic data. Especially we use a graph data model to express semantic and structural features of the semantic data and process various type of queries by using graph model based path expressions. In this paper the proposed index aims to distinguish our approach from earlier studies and involve the concept of the Semantic Web in its entirety by querying on primarily extracted structural path information and secondary extracted one through semantic inferences with ontology. In the experiments, we show that our approach is more accurate and efficient than the previous approaches and can be applicable to various applications in the Semantic Web.

• The Journal of Bigdata
• /
• v.9 no.1
• /
• pp.89-100
• /
• 2024

In the field of computer vision, models such as You Look Only Once (YOLO) and ResNet are widely used due to their real-time performance and high accuracy. However, to apply these models in real-world environments, factors such as runtime compatibility, memory usage, computing resources, and real-time conditions must be considered. This study compares the characteristics of three deep model runtimes: ONNX Runtime, TensorRT, and OpenCV DNN, and analyzes their performance on two models. The aim of this paper is to provide criteria for runtime selection for practical applications. The experiments compare runtimes based on the evaluation metrics of time, memory usage, and accuracy for vehicle license plate recognition and classification tasks. The experimental results show that ONNX Runtime excels in complex object detection performance, OpenCV DNN is suitable for environments with limited memory, and TensorRT offers superior execution speed for complex models.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.278-281
• /
• 2018

The early Web technology was to show text information through a browser. However, as web technology advances, it is possible to show large amounts of multimedia data through browsers. Web technologies are being applied in a variety of fields such as sensor network, hardware control, and data collection and analysis for big data and AI services. As a result, the standard has been prepared for the Internet of Things, which typically controls a sensor via HTTP communication and provides information to users, by installing a web browser on the interface of the Internet of Things. In addition, the recent development of web-assembly enabled 3D objects, virtual/enhancing real-world content that could not be run in web browsers through a native language of C-class. Factors that evaluate the performance of existing Web applications include performance, network resources, and security. However, since there are many areas in which web applications are applied, it is time to revisit and review these factors. In this thesis, we will conduct an analysis of the factors that assess the performance of a web application. We intend to establish an indicator of the development of web-based applications by reviewing the analysis of each element, its main points, and its needs to be supplemented.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.5
• /
• pp.39-48
• /
• 2016

One of the major trends of automotive networking architecture is the introduction of automotive Ethernet. Ethernet is already used in single automotive applications (e.g. to connect high-data-rate sources as video cameras), it is expected that the ongoing standardization at IEEE (IEEE802.3bw - 100BASE-T1, respectively IEEE P802.3bp - 1000BASE-T1) will lead to a much broader adoption in future. Those applications will not be limited to simple point-to-point connections, but may affect Electric/Electronic(EE) Architectures as a whole. It is agreed that IP based traffic via Ethernet could be secured by application of well-established IP security protocols (e.g., IPSec, TLS) combined with additional components like, e.g., automotive firewall or IDS. In the case of safety and real-time related applications on resource constraint devices, the IP based communication is not the favorite option to be used with complicated and performance demanding TLS or IPSec. Those applications will be foreseeable incorporate Layer-2 based communication protocols as, e.g., currently standardized at IEEE[13]. The present paper reflects the state-of-the-art communication concepts with respect to security and identifies architectural challenges and potential solutions for future Ethernet Switch-based EE-Architectures. It also gives an overview and provide insights into the ongoing security relevant standardization activities concerning automotive Ethernet. Furthermore, the properties of non-automotive Ethernet security mechanisms as, e.g., IEEE 802.1AE aka. MACsec or 802.1X Port-based Network Access Control, will be evaluated and the applicability for automotive applications will be assessed.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.4
• /
• pp.3-11
• /
• 2016

In the environment of Internet of Things (IoT), IoT devices are limited by physical components such as power supply and memory, and also limited to their network performance in bandwidth, wireless channel, throughput, payload, etc. Despite these limitations, resources of IoT devices are shared with other IoT devices. Especially, remote management of the information of devices and patients are very important for the IoT healthcare service, moreover, providing the interoperability between the healthcare device and healthcare platform is essential. To meet these requirements, format of the message and the expressions for the data information and data transmission need to comply with suitable international standards for the IoT environment. However, the ISO/IEEE 11073 PHD (Personal Healthcare Device) standards, the existing international standards for the transmission of health informatics, does not consider the IoT environment, and therefore it is difficult to be applied for the IoT healthcare service. For this matter, we have designed and implemented the IoT healthcare system by applying the oneM2M, standards for the Internet of Things, and ISO/IEEE 11073 DIM (Domain Information Model), standards for the transmission of health informatics. For the implementation, the OM2M platform, which is based on the oneM2M standards, has been used. To evaluate the efficiency of transfer syntaxes between the healthcare device and OM2M platform, we have implemented comparative performance evaluation between HTTP and CoAP, and also between XML and JSON by comparing the packet size and number of packets in one transaction.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.6
• /
• pp.205-220
• /
• 2018

Software defect prediction is helpful for allocating valuable project resources effectively for software quality assurance activities thanks to focusing on the identified fault-prone modules. If historical data collected within a company is sufficient, a Within-Project Defect Prediction (WPDP) can be utilized for accurate fault-prone module prediction. In case a company does not maintain historical data, it may be helpful to build a classifier towards predicting comprehensible fault prediction based on Cross-Project Defect Prediction (CPDP). Since CPDP employs different project data collected from other organization to build a classifier, the main obstacle to build an accurate classifier is that distributions between source and target projects are not similar. To address the problem, because it is crucial to identify effective similarity measure techniques to obtain high performance for CPDP, In this paper, we aim to identify them. We compare various similarity measure techniques. The effectiveness of similarity weights calculated by those similarity measure techniques are evaluated. The results are verified using the statistical significance test and the effect size test. The results show k-Nearest Neighbor (k-NN), LOcal Correlation Integral (LOCI), and Range methods are the top three performers. The experimental results show that predictive performances using the three methods are comparable to those of WPDP.