• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.80-88
• /
• 2023

RGB-thermal (RGB-T) tracking using unmanned aerial vehicles (UAVs) involves challenges with regards to the similarity of objects, occlusion, fast motion, and motion blur, among other issues. In this study, we propose dynamic tracking aggregation (DTA) as a unified framework to perform object detection and data association. The proposed approach obtains fused features based a transformer model and an L1-norm strategy. To link the current frame with recent information, a dynamically updated embedding called dynamic tracking identification (DTID) is used to model the iterative tracking process. For object association, we designed a long short-term tracking aggregation module for dynamic feature propagation to match spatial and temporal embeddings. DTA achieved a highly competitive performance in an experimental evaluation on public benchmark datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1223-1237
• /
• 2024

Autonomous vehicles use onboard sensors to sense the surrounding environment. In complex autonomous driving scenarios, the detection and recognition capabilities are constrained, which may result in serious accidents. An efficient way to enhance the detection and recognition capabilities is establishing collaborations with the neighbor vehicles. However, the collaborations introduce additional challenges in terms of the data heterogeneity, communication cost, and data privacy. In this paper, a novel personalized federated learning framework is proposed for addressing the challenges and enabling efficient collaborations in autonomous driving environment. For obtaining a global model, vehicles perform local training and transmit logits to a central unit instead of the entire model, and thus the communication cost is minimized, and the data privacy is protected. Then, the inference similarity is derived for capturing the characteristics of data heterogeneity. The vehicles are divided into clusters based on the inference similarity and a weighted aggregation is performed within a cluster. Finally, the vehicles download the corresponding aggregated global model and train a personalized model which is personalized for the cluster that has similar data distribution, so that accuracy is not affected by heterogeneous data. Experimental results demonstrate significant advantages of our proposed method in improving the efficiency of collaborative perception and reducing communication cost.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.9
• /
• pp.2350-2364
• /
• 1999

In this paper, we propose a method of class library repository design for provide reuser the object-oriented C++ class component. To class library design, we started by studying the characteristics of a reusable component. We formally defined the reusable component model using an entity relationship model. This formal definition has been directly used as the database schema for storing the reusable component in a repository. The reusable class library may be considered a knowledge base for software reuse. Thus, we used that Enumerative classification of breakdown of knowledge based. And another used classification is clustering of based on class similarity. The class similarity composes member function similarity and member data similarity. Finally, we have designed class library for hierarchical inheritance mechanism of object-oriented concept Generalization, Specialization and Aggregation.

• Journal of Intelligence and Information Systems
• /
• v.13 no.3
• /
• pp.141-160
• /
• 2007

User interface has been regarded as a crucial issue to increase the acceptance of mobile services. In special, even though to what extent the machine as speaker communicates with human as listener in a timely and polite manner is important, fundamental studies to come up with these issues have been very rare. Hence, the purpose of this paper is to propose a methodology of estimating politeness level in a certain context-aware setting and then to design a context-aware system for polite mobile interface. We will focus on Korean language for the polite level estimation simply because the polite interface would highly depend on cultural and linguistic characteristics. Nested Minkowski aggregation model, which amends Minkowski aggregation model, is adopted as a privacy-preserving similarity evaluation for case retrieval under distributed computing environment such as ubiquitous computing environment. To show the feasibility of the methodology proposed in this paper, simulation-based experiment with drama cases has performed to show the performance of the methodology proposed in this paper.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.3
• /
• pp.416-420
• /
• 2002

Traditional queueing analyses are very useful for designing a network's capacity and predicting there performances, however most of the predicted results from the queueing analyses are quite different from the realistic measured performance. And recent empirical studies on LAN, WAN and VBR traffic characteristics have indicated that the models used in the traditional Poisson assumption can't properly predict the real traffic properties due to under estimation of the long range dependence of network traffic and self-similarity In this parer self-similar characteristics over statistical approaches and real time network traffic measurements are estimated It is also shown that the self- similar traffic reflects network traffic characteristics by comparing source model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.104-107
• /
• 2002

Traditional queueing analyses are very useful for designing a network's capacity and predicting there performances, however most of the predicted results from the queueing analyses are quite different from the realistic measured performance. And recent empirical studies on LAN, WAN and VBR traffic characteristics have indicated that the models used in the traditional Poisson assumption can't properly predict the real traffic properties due to under estimation of the long range dependence of network traffic and self-similarity. In this paper self-similar characteristics over statistical approaches and real time network traffic measurements are estimated. It is also shown that the self-similar traffic reflects network traffic characteristics by comparing source model.

• Journal of the Korean Society of Safety
• /
• v.36 no.2
• /
• pp.49-57
• /
• 2021

In the case of military supplies, any potential failure and causes of failures must be considered. This study is aimed at examining the failure modes of a rotorcraft landing system to identify the priority items. Failure mode and effects analysis (FMEA) is applied to the rotorcraft landing system. In general, the FMEA is used to evaluate the reliability in engineering fields. Three elements, specifically, the severity, occurrence, and detectability are used to evaluate the failure modes. The risk priority number (RPN) can be obtained by multiplying the scores or the risk levels pertaining to severity, occurrence, and detectability. In this study, different weights of the three elements are considered for the RPN assessment to implement the FMEA. Furthermore, the FMEA is implemented using a fuzzy rule base, similarity aggregation model (SAM), and grey theory model (GTM) to perform a comparative analysis. The same input data are used for all models to enable a fair comparison. The FMEA is applied to military supplies by considering methodological issues. In general, the fuzzy theory is based on a hypothesis regarding the likelihood of the conversion of the crisp value to the fuzzy input. Fuzzy FMEA is the basic method to obtain the fuzzy RPN. The three elements of the FMEA are used as five linguistic terms. The membership functions as triangular fuzzy sets are the simplest models defined by the three elements. In addition, a fuzzy set is described using a membership function mapping the elements to the intervals 0 and 1. The fuzzy rule base is designed to identify the failure modes according to the expert knowledge. The IF-THEN criterion of the fuzzy rule base is formulated to convert a fuzzy input into a fuzzy output. The total number of rules is 125 in the fuzzy rule base. The SAM expresses the judgment corresponding to the individual experiences of the experts performing FMEA as weights. Implementing the SAM is of significance when operating fuzzy sets regarding the expert opinion and can confirm the concurrence of expert opinion. The GTM can perform defuzzification to obtain a crisp value from a fuzzy membership function and determine the priorities by considering the degree of relation and the form of a matrix and weights for the severity, occurrence, and detectability. The proposed models prioritize the failure modes of the rotorcraft landing system. The conventional FMEA and fuzzy rule base can set the same priorities. SAM and GTM can set different priorities with objectivity through weight setting.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.518-528
• /
• 2016

The searching of an extensive document database for documents that are locally similar to a given query document, and the subsequent detection of similar regions between such documents, is considered as an essential task in the fields of information retrieval and data management. In this paper, we present a framework for such a task. The proposed framework employs the method of short-read mapping, which is used in bioinformatics to reveal similarities between genomic sequences. In this paper, documents are considered biological objects; consequently, edit operations between locally similar documents are viewed as an evolutionary process. Accordingly, we are able to apply the method of evolution tracing in the detection of similar regions between documents. In addition, we propose heuristic methods to address issues associated with the different stages of the proposed framework, for example, a frequency-based fragment ordering method and a locality-aware interval aggregation method. Extensive experiments covering various scenarios related to the search of an extensive document database for documents that are locally similar to a given query document are considered, and the results indicate that the proposed framework outperforms existing methods.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.2
• /
• pp.79-90
• /
• 2006

The partitioning level of a catchment becomes an issue if the calculated results from different levels show the same performance regardless of the levels. This study aims to identify the proper processing level of spatial resolution for the SWMM model application in an urban area. Using GIS overlaying technique, the division of subcatchments as a hydrologic similarity unit (HSU) is achieved with a comprehensive consideration of surface slope conditions, flow directions of storm sewers, and current land cover situation. Three surface-sewer alternatives are made on the basis of three different levels of surface divisions as well as the number of sewer connections and used as runoff simulation fields for the application of SWMM. As the result, it is found that the effect of a spatial resolution on the surface runoff results is not significant. On the other hand, the accumulated pollution load from an unit subcatchment, which is built by aggregation of several unit subcatchments consisting of various land cover conditions is reduced through the deterioration of surface spatial resolution. Although overall runoff pattern and accumulated runoff are little affected by spatial resolution, the simulated runoff from sewer outlet shows slight difference at the peak appearance time. The gap between surface pollution load accumulated and it discharged from the sewer outlet in a surface-sewer alternative during runoff period is monitored but the level of error is less than 5-10% except the lowest spatial resolution case.

• Journal of Internet Computing and Services
• /
• v.20 no.6
• /
• pp.143-156
• /
• 2019

Discussion on OTT regulation has become fiercer, as OTT services' impacts on the global and domestic media market have been exponentially growing. In South Korea, it is argued that, on the basis of the similarity between television program and OTT's video content, OTT needs to be regulated in order to protect fair competition and to control sociocultural effects. In many of the discussions, developed countries' cases have been used for supporting OTT regulation. In this paper, we first analyzed the global OTT market trends based on our own categorization of OTT services. then we assessed the validity of the application of the foreign cases in the current OTT regulation debates in Korea. We proposed six OTT types (aggregation, mediation, mediation-aggregation, multi-screen, outlet, and outlet-linear) simultaneously considering service operator' origin, business model, content format, and content delivery. These services have been consistently evolving, and the OTT market has been increasingly competitive especially around content differentiation. Regulators must be wary of hastily introducing competition regulation to the dynamically innovating OTT market. The foreign cases, including the US, EU, the UK, and Japan, hardly seem to be appropriate bases for strengthening OTT regulation. Rather, they were focused more on promoting competition in the domestic media market and enriching the content ecosystem. Therefore, we need to consider revision of the outdated media regulation frameworks instead of fitting OTT under them, and to recognize the priority of securing practical jurisdiction on global service providers before capturing local players into the conventional regulation systems.