• Journal of Information Processing Systems
• /
• v.18 no.4
• /
• pp.575-586
• /
• 2022

Uneven light in real-world causes visual degradation for underexposed regions. For these regions, insufficient consideration during enhancement procedure will result in over-/under-exposure, loss of details and color distortion. Confronting such challenges, an unsupervised low-light image enhancement network is proposed in this paper based on the guidance of the unpaired low-/normal-light images. The key components in our network include super-resolution module (SRM), a GAN-based low-light image enhancement network (LLIEN), and denoising-scaling module (DSM). The SRM improves the resolution of the low-light input images before illumination enhancement. Such design philosophy improves the effectiveness of texture details preservation by operating in high-resolution space. Subsequently, local lightness attention module in LLIEN effectively distinguishes unevenly illuminated areas and puts emphasis on low-light areas, ensuring the spatial consistency of illumination for locally underexposed images. Then, multiple discriminators, i.e., global discriminator, local region discriminator, and color discriminator performs assessment from different perspectives to avoid over-/under-exposure and color distortion, which guides the network to generate images that in line with human aesthetic perception. Finally, the DSM performs noise removal and obtains high-quality enhanced images. Both qualitative and quantitative experiments demonstrate that our approach achieves favorable results, which indicates its superior capacity on illumination and texture details restoration.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.755-778
• /
• 2024

In recent years, the number of devices being connected to the internet has grown enormously, as has the intrusive behavior in the network. Thus, it is important for intrusion detection systems to report all intrusive behavior. Using deep learning and machine learning algorithms, intrusion detection systems are able to perform well in identifying attacks. However, the concern with these deep learning algorithms is their inability to identify a suitable network based on traffic volume, which requires manual changing of hyperparameters, which consumes a lot of time and effort. So, to address this, this paper offers a solution using the extended compact genetic algorithm for the automatic tuning of the hyperparameters. The novelty in this work comes in the form of modeling the problem of identifying attacks as a multi-objective optimization problem and the usage of linkage learning for solving the optimization problem. The solution is obtained using the feature map-based Convolutional Neural Network that gets encoded into genes, and using the extended compact genetic algorithm the model is optimized for the detection accuracy and latency. The CIC-IDS-2017 and 2018 datasets are used to verify the hypothesis, and the most recent analysis yielded a substantial F1 score of 99.23%. Response time, CPU, and memory consumption evaluations are done to demonstrate the suitability of this model in a fog environment.

• The Journal of Society for e-Business Studies
• /
• v.17 no.3
• /
• pp.105-116
• /
• 2012

In the system integration (SI) industry, improving relationships with outsourcing partners is one of the key challenges faced by SI companies that aim to increase productivity and reduce costs. In the SI industry, over 80% of project development work is performed by these outsourcing partners. Hence, large SI companies focus on fostering high-performing partners through improved cooperation and mutual relations. However, in the SI industry the existing process of managing outsourcing partners is unable to advance to the level of long-term partnerships due to inadequate information sharing and communication in the process of selecting and assessing a potential and existing partner. Based on the theory of partnership, this study identifies the success factors for new strategic partnerships. Further, this study presents a model for assessing partner companies to promote increased productivity and improve the relationship between an SI company and its partner. In addition, this model is used to assess partner relationships for a well-known SI company "A". This study explores the feasibility of a new framework model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.7
• /
• pp.1091-1097
• /
• 2022

In this paper, we implement the eXclusive OR-based Cast (XC) system that is a video streaming system using exclusive OR operations, and measure various performance metrics in wireless local area network (WLAN) environments. In addition, we investigate the performance improvement of the XC system considering various practical video streaming environments, while conventional studies analyzed the performance of XC through computer simulations in limited environments. To this end, we propose new control messages such as STR_REQ_MSG (SRM) that clients transmit to a video streaming server and STR_CON_MSG (SCM) that is used for the video streaming server to control the clients, and develop a new protocol by using the new control messages. According to the various measurement results using the implemented XC system, XC video streaming system can reduce the consumption of network bandwidth by 8.6% on average and up to 25% compared to the conventional video streaming system. In addition, the outage probability can be also reduced up to 76%.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.4
• /
• pp.1578-1585
• /
• 2018

This work investigates two different motor drive technologies, switched reluctance motor (SRM) and induction motor (IM). They are designed optimally to meet the desired performances for electric scooters. The comparison of both motors is described in terms of performances and material cost. With the similar constraint, induction motor performs slightly better than switched reluctance motor. But this must be traded-off with higher weight and cost. Both drive systems are, however, suitable for electric scooter application. Finally, the range simulations are conducted on a European urban driving cycle, ECE15 driving cycle and a more realistic cycle, Bangkok driving cycle. The e-scooter ranges are varied from 36 to 109 km depending on driving cycle, motor technology and number of passengers.

• Restorative Dentistry and Endodontics
• /
• v.47 no.2
• /
• pp.19.1-19.15
• /
• 2022

Objectives: The aim of this study was to evaluate and compare the apical constriction (AC) and apical canal morphology of maxillary first and second molars, using micro-computed tomography (micro-CT). Materials and Methods: The anatomical features of 313 root canals from 41 maxillary first molars and 57 maxillary second molars of patients with known age and sex were evaluated using micro-CT, with a resolution of 26.7 ㎛. The factors evaluated were the presence or absence of AC, the morphotypes, bucco-lingual dimension, mesio-distal dimension, and the profile (shape) of AC and the apical root canal. The apical root canal dimensions, location of the apical foramen (AF), AC to AF distance, and presence of accessory canals in the apical 5 mm were also assessed. Descriptive and analytical statistics were used for data evaluation. Results: AC was present in all 313 root canals. Patients' age and sex did not significantly impact either AC or the apical canal dimensions. The most common AC morphotype detected was the traditional (single) constriction (52%), followed by the parallel (29%) morphotype. The mean AC dimensions in maxillary first molars were not significantly different from those in maxillary second molars. Sixty percent of AF were located within 0.5 mm from the anatomic apex. Conclusions: The most common morphotype of AC detected was the traditional constriction. Neither patients' age nor sex had a significant impact on the dimensions of the AC or the apical root canal. The majority of AF (60%) were located within 0.5 mm from the anatomic apex.

• Journal of Magnetics
• /
• v.21 no.3
• /
• pp.393-398
• /
• 2016

This paper proposes a novel 16/10 segmented rotor switched reluctance motor (SSRM) for belt-driven starter generators (BSGs). Different from conventional SRMs, the stator of the proposed SSRM consists of two types of stator poles, i.e., exciting and auxiliary poles, and the rotor is constructed from a series of discrete segments. The topology and operation principle of this proposed SSRM are illustrated firstly, and then the design rules are listed. In addition, the finite element method (FEM) is employed to get the static and dynamic characteristics of the proposed SSRM. Finally, the simulation results are presented to show the validity of the proposed SSRM for BSGs.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2000.04a
• /
• pp.12-12
• /
• 2000

우리 나라는 일본상공의 비행을 피하기 위하여 제주도와 남해안 근해로 발사장 선정이 국한되는 지정학적인 위치로 볼 때 발사장 선택에 제한이 없는 공중발사에 대한 가능성 연구가 필요한 시점에 있다. 본 연구는 우리 나라와 같은 분단 된 특수상황 그리고 지정학적 위치에서의 발사장을 고려한 우주 발사체 개발의 필요성에 따라 F-4에 장착 가능한 3단형 공중발사 로켓을 설계하고 1/3의 축소 모형을 제작하였다. 2kg의 payload를 갖는 발사체의 1단은 LRM ( Lox/kerosene )을 사용하였고 2, 3단은 SRM ( HTPB/AP/Al )을 사용하였으며 발사고도는 11-l2km 상공에서 F-4에 의해서 발사되고 31km지점에서 1단 분리가 이루어지며 62km지점에서 2단 분리와 nose fairing을 분리하게 된다. 전장은 6.85m 이며 전체 무게는 560.6kg 이고 전체 발사체 시스템의 CAD 도면은 아래 그림 1과 같이 주어져 있다. 그림 2에서는 F-4E phantom의 장착성을 검토해 본 결과 장착이 가능함을 알 수 있었으며 추진제 양의 감소로 크기를 대폭 줄일 수 있었다.

• Journal of the Korea Safety Management & Science
• /
• v.6 no.2
• /
• pp.167-178
• /
• 2004

Supply chain management and e-procurement system built in the company, are set goal of internal process improvement. This paper show how SLV(Supplier Lifetime Value) can be used to structure the supplier in automotive part manufacturer, the method of selection is described and a detailed example of how SLV can be used and allocation of order Quantity for the selected suppliers also are provided, and aims at giving a method to realization profitable supplier relation using search for SL V system development which is making corporation with supplier, in the favor of decision supplier.

• Nuclear Engineering and Technology
• /
• v.34 no.6
• /
• pp.566-573
• /
• 2002

Neutron activation analysis based on $textsc{k}$$_{o}$-standardization method＃ ($textsc{k}$o-NAA) is Com as one of the most remarkable progresses of the NAA with advantages of experimental simplicity, high accuracy, excellent flexibility with respect to irradiation and counting conditions, and suitability for computerization. This study was carried out to determine the reactor neutron spectrum parameters, i.e. $\alpha$ and f as the main factors of irradiation quality at NAA ＃1 irradiation hole on HANARO research reactor, to evaluate peak detection efficiency of the gamma-ray spectrometer for the use in the $textsc{k}$$_{o}$ experiments and to compare the measured concentration results with the certified values of some SRMs applying the experimentally determined to-parameters.ers.