• Journal of Korea Multimedia Society
• /
• v.23 no.6
• /
• pp.729-737
• /
• 2020

Screening with low-dose spiral computed tomography (LDCT) has been shown to reduce lung cancer mortality by about 20% when compared to standard chest radiography. One of the problems arising from screening programs is that large amounts of CT image data must be interpreted by radiologists. To solve this problem, automated detection of pulmonary nodules is necessary; however, this is a challenging task because of the high number of false positive results. Here we demonstrate detection of pulmonary nodules using six off-the-shelf convolutional neural network (CNN) models after modification of the input/output layers and end-to-end training based on publicly databases for comparative evaluation. We used the well-known CNN models, LeNet-5, VGG-16, GoogLeNet Inception V3, ResNet-152, DensNet-201, and NASNet. Most of the CNN models provided superior results to those of obtained using customized CNN models. It is more desirable to modify the proven off-the-shelf network model than to customize the network model to detect the pulmonary nodules.

• Journal of Information Processing Systems
• /
• v.6 no.2
• /
• pp.177-184
• /
• 2010

Secure routing is vital to the acceptance and use of Wireless Sensor Networks (WSN) for many applications. However, providing secure routing in WSNs is a challenging task due to the inherently constrained capabilities of sensor nodes. Although a wide variety of routing protocols have been proposed for WSNs, most do not take security into account as a main goal. Routing attacks can have devastating effects on WSNs and present a major challenge when designing robust security mechanisms for WSNs. In this paper, we examine some of the most common routing attacks in WSNs. In particular, we focus on the wormhole routing attack in some detail. A variety of countermeasures have been proposed in the literature for such attacks. However, most of these countermeasures suffer from flaws that essentially render them ineffective for use in large scale WSN deployments. Due to the inherent constraints found in WSNs, there is a need for lightweight and robust security mechanisms. The examination of the wormhole routing attack and some of the proposed countermeasures makes it evident that it is extremely difficult to retrofit existing protocols with defenses against routing attacks. It is suggested that one of the ways to approach this rich field of research problems in WSNs could be to carefully design new routing protocols in which attacks such as wormholes can be rendered meaningless.

• Korean Journal of Biological Psychiatry
• /
• v.22 no.3
• /
• pp.95-100
• /
• 2015

Treatment of schizophrenia is one of the most challenging areas in the field of psychiatry. There has been much improvement in psycho-pharmacotherapy, and at present, psycho-pharmacotherapy along with milieu therapy and social rehabilitation is the standard first-line treatment for schizophrenia. Healing environment, a concept which has arisen from the architectural field, has similarities in meaning to milieu therapy in psychiatry. In other words, healing environment may be an encountering point between psychiatry and architecture. In this encountering, each field can understand each other and expand its concept to aid the treatment of schizophrenia and to plan the build-up of the entire environment considering its social and psychological effects. In this paper, we aim to establish the basic concept of healing environment to alleviate the psychopathologies in schizophrenic patients. We worked under the premise that physical setting affects human behavior and mind, and that physical setting should play a role as a medium with therapeutic potential for patients with medical problems. The aims of this paper are as follows. First, theoretical discussion of the concept and the constructs of healing environment : second, understanding of the schizophrenic symptoms that may be affected by supporting environment : and third, discussion of supporting environment that may alleviate the symptoms of schizophrenia.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.6
• /
• pp.2406-2412
• /
• 2015

Multi-user multi-input multi-output (MU-MIMO) system has attracted the 4^th generation wireless network as one of core technique for performance enrichment. In this system rate control is a challenging problem and another problem is optimization. Proper scheduling can resolve these problems by deciding which set of user and at which rate the users send their data. This paper proposes a new multi-parameter based scheduling (MPS) for downlink multi-user multiple-input multiple-output (MU-MIMO) system under space-time block coding (STBC) transmissions. Goal of this MPS scheme is to offer improved link level performance in terms of a low average bit error rate (BER), high packet delivery ratio (PDR) with improved resource utilization and service fairness among the user. This scheme allows the set of users to send data based on their channel quality and their demand rates. Simulation compares the MPS performance with other scheduling scheme such as fair scheduling (FS), normalized priority scheduling (NPS) and threshold based fair scheduling (TFS). The results obtained prove that MPS has significant improvement in average BER performance with improved resource utilization and fairness as compared to the other scheduling scheme.

• Journal of Soil and Groundwater Environment
• /
• v.22 no.3
• /
• pp.1-9
• /
• 2017

The numerical analysis of groundwater flow is indispensable for predicting problems associated with water resource development, civil works, environmental hazards, and nuclear power plant construction. Korea lacks public regulatory procedures and guidelines for groundwater flow modeling, especially in nuclear facility sites, which makes adequate evaluation difficult. Feasible step-by-step guidelines are also unavailable. Consequently, reports on groundwater flow modeling have low-grade quality and often present controversial opinions. Additionally, without public guidelines, maintaining consistency in reviewing reports and enforcing laws is more challenging. In this study, the guidelines for groundwater flow modeling were reviewed for three countries - the United States (Documenting Groundwater Modeling at Sites Contaminated with Radioactive Substances), Canada (Guidelines for Groundwater Modelling to Assess Impacts of Proposed Natural Resource Development Activities), and Australia (Australian Groundwater Modelling Guidelines), with the aim of developing groundwater flow modeling regulatory guidelines that can be applied to nuclear facilities in Korea, in accordance with the Groundwater Act, Environmental Impact Assessment Act, and the Nuclear Safety Act.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.11 no.4
• /
• pp.173-188
• /
• 1999

Kim et al.(I999) presented a non-linear finite element formulation of spatial ocean cables using multiple noded cable elements. The initial equilibrium state of ocean cables subjected to self-weights, support motions, and current forces was determined using the load incremental method and free vibration analysis were performed considering added mass, In this paper, the methods to generate regular and irregular waves and calculate wave forces due to these waves are discussed and challenging example problems are presented in order to investigate dynamic non-linear behaviors of ocean cables subjected to wave loadings.

• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.197-204
• /
• 2020

In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.8
• /
• pp.50-57
• /
• 2003

Controlling rotorcraft to fly precisely through multiple, irregularly, and closely spaced waypoints is a common and practical mission. However, finding an optimal trajectory for this kind of mission is quite challenging. Usability of traditional approaches such as inverse control or direct methods to this kind of problem is limited because of either limitation on the specification of the constraints or requirement of extensive computation time. This paper proposes a method that can easily compute the full trajectory and control history for rotorcraft to pass through waypoints while satisfying other general constraints of states such as velocities and attitudes on each waypoint. The proposed method is applied to rotorcraft guidance problems of slalom and linear trajectory in the middle of general curved trajectory. The algorithm is test for various situations and demonstrates its usability.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.25 no.12
• /
• pp.60-66
• /
• 2011

There are several technological problems have to be resolved for LEDs to be used as a general purpose light source. In addition, there are several differences between existing luminaires and the general planer LED luminaire for the intensity distribution. Therefore, the optical engineer then faces the challenging a problem of designing for a spatially extended and non-uniform light source. In the previous studies on the optical design of luminaires, a lot of studies on reflectors and light source have been conducted but the ones on prisms and lenses are insufficient at present. This study developed the numerical model of planar prism LED luminaire to control luminous intensity distribution of LED luminaires. And this study presents an optical calculation process for the prism optical design of a planar prism LED luminaire and a comparison of the simulation results between the developed numerical model and Photopia 2.0 to verify the accuracy of the numerical model. In addition, this study showed a method for the control of various luminous intensity distribution from the developed numerical model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2667-2682
• /
• 2015

Due to the wide application of face recognition (FR) in information security, surveillance, access control and others, it has received significantly increased attention from both the academic and industrial communities during the past several decades. However, partial face occlusion is one of the most challenging problems in face recognition issue. In this paper, a novel method based on linear regression-based classification (LRC) algorithm is proposed to address this problem. After all images are downsampled and divided into several blocks, we exploit the evaluator of each block to determine the clear blocks of the test face image by using linear regression technique. Then, the remained uncontaminated blocks are utilized to partial occluded face recognition issue. Furthermore, an improved Distance-based Evidence Fusion approach is proposed to decide in favor of the class with average value of corresponding minimum distance. Since this occlusion removing process uses a simple linear regression approach, the completely computational cost approximately equals to LRC and much lower than sparse representation-based classification (SRC) and extended-SRC (eSRC). Based on the experimental results on both AR face database and extended Yale B face database, it demonstrates the effectiveness of the proposed method on issue of partial occluded face recognition and the performance is satisfactory. Through the comparison with the conventional methods (eigenface+NN, fisherfaces+NN) and the state-of-the-art methods (LRC, SRC and eSRC), the proposed method shows better performance and robustness.