• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• Geomechanics and Engineering
• /
• v.34 no.2
• /
• pp.155-171
• /
• 2023

Aiming at the grouting treatment of water inflow in karst conduits, a visualized experiment system for conduit-type grouting blocking was developed. Through the improved water supply system and grouting system, and the optimized multisource information monitoring system, the real-time observation of diffusion and deposition of slurry, and the data acquisition of pressure and velocity during the whole process of grouting were realized, which breaks through the problem that the monitoring element is easy to fail due to slurry adhesion in conventional test system. Based on the grouting experiments in static and flowing water, the diffusion and deposition behavior of the quick-setting slurry under different working conditions were analyzed. The temporal and spatial variation behavior of the pressure and velocity were studied, and the blocking mechanism of the grouting were further revealed. The results showed that: (1) Under the flowing water condition, the counter-flow diffusion distance of slurry was negatively correlated with the flow water velocity and the volume ratio of cement and sodium silicate (C-S ratio), and positively correlated with the grouting volume. The slurry deposition thickness was negatively correlated with the flowing water velocity, and positively correlated with the grouting volume and C-S ratio. (2) The pressure increased slowly before blocking of the flowing water and rapidly after blocking in karst conduits. (3) With the continuous progress of grouting, the flowing water velocity decreased slowly first, then significantly, and finally tended to be stable. According to the research results, some engineering recommendations were put forward for the grouting treatment of the conduit-type water inflow disaster, which has been successfully applied in the treatment project of the China Resources Cement (Pingnan) Limestone Mine. This study provided some guidance and reference for the parameter optimization of grouting for the treatment projects of water inflow in karst conduits.

• Journal of the HCI Society of Korea
• /
• v.9 no.1
• /
• pp.23-28
• /
• 2014

In this study we focus on the effect of the interaction between humans and reactive television when emotion recognition through facial expression mechanism is used. Most of today's user interfaces in electronic products are passive and are not properly fitted into users' needs. In terms of the user centered device, we propose that the emotion based reactive television is the most effective in interaction compared to other passive input products. We have developed and researched next generation cognitive TV models in user centered. In this paper we present a result of the experiment that had been taken with Fraunhofer IIS $SHORE^{TM}$ demo software version to measure emotion recognition. This new approach was based on the real time cognitive TV models and through this approach we studied the relationship between humans and cognitive TV. This study follows following steps: 1) Cognitive TV systems can be on automatic ON/OFF mode responding to motions of people 2) Cognitive TV can directly select channels as face changes (ex, Neutral Mode and Happy Mode, Sad Mode, Angry Mode) 3) Cognitive TV can detect emotion recognition from facial expression of people within the fixed time and then if Happy mode is detected the programs of TV would be shifted into funny or interesting shows and if Angry mode is detected it would be changed to moving or touching shows. In addition, we focus on improving the emotion recognition through facial expression. Furthermore, the improvement of cognition TV based on personal characteristics is needed for the different personality of users in human to computer interaction. In this manner, the study on how people feel and how cognitive TV responds accordingly, plus the effects of media as cognitive mechanism will be thoroughly discussed.

• Journal of KIISE:Information Networking
• /
• v.31 no.1
• /
• pp.91-100
• /
• 2004

This paper presents a receiver-driven TCP flow control mechanism, which is adaptive to the wireless condition, for memory constrained mobile receiver. A receiver-driven TCP flow control is, in general, achieved by adjusting the size of advertised window at the receiver. The proposed method constantly measures at the receiver both the available wireless bandwidth and the packet round-trip time. Depending on the measured values, the receiver adjusts appropriately the size of advertised window. Constrained by the adjusted window which reflects the current state of the wireless network, the sender achieves an improved TCP throughput as well as the reduced round-trip packet delay. Its implementation only affects the protocol stack at the receiver and hence neither the sender nor the router are required to be modified. The mechanism has been implemented in real environments. The experimental results show that in CDMA2000 1x networks the TCP throughput of the proposed method has improved about 5 times over the conventional method when the receiver's buffer size is limited to 2896 bytes. Also, with 64Kbytes of buffer site, the packet round-trip time of the proposed method has been reduced in half, compared the case with the conventional method.

• The Journal of Society for e-Business Studies
• /
• v.4 no.2
• /
• pp.1-22
• /
• 1999

In this paper, we review and classify shopping information services for electronic commerce in Korea. Shopping information services for electronic commerce can be divided into shopping mall information service and product information service. Shopping mall information service simply shows and suggests information on the shopping malls that it has in its data store. But product information service provides information about the items of the shopping malls the service has and therefore gives more powerful and convenient interfaces to users. Cutting-edge product information services can perform online comparison shopping facility, which searches and compares the items what the users want on the reasonable basis of economical criteria, such as price. Books, CDs, and Personal Data Assistant(PDA)s are the products of which the specification is clear enough to be described on the Web. And the comparison of the products by the information service is performed on the basis of "price" of the product item. The approach to design and implement product information service can be divided into two categories. The one is an application of data warehousing mechanism, which stores the product information on the regular basis of time and provide it when the users query over the Web-database gateway. The other approach is a shopping agent mechanism, which stores information on "how to shop" and the shopping agent collects the information of product items just after users query about the product and provide the information in real time or notify them by alerting service. Thirty nine shopping information services are compared and classified in this paper and they are extracted from "Naver" and "Yahoo! Korea". The final result shows that most services are just a simple shopping mall information service in growing stage, product information services with data warehousing are lack in product ontology in middle stage, and that product information services with shopping agent are in early stage.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.14 no.4
• /
• pp.86-93
• /
• 2010

The wear of railway track affects loss of rough ride, noise or vibration of train and traveling safety. Moreover as the track is worn away, this promotes destruction of structural mechanism of rail track which can bring about increasing of rail track maintenance cost drastically. For this reason, it is very important and interested research subject to design railway track structure and to analyse train movement mechanism based on systematic analysis of the reasons causing rail wear possible in real field. In this research, for the efficient maintenance, Life Cycle Performance of rail track and maintenance characteristics are computed considering some track components such as track type, contracting type, sleeper type and roadbed type. Time - Wear probabilistic distribution relationship as well as multiple regression analysis based on time, curvature and wear data are computed to predict the service life remainder of railway track and to be adapted to safety assessment.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.12C
• /
• pp.1245-1250
• /
• 2002

For recent years, the environment of use on internet has been changed from the wired internet access to the wireless internet access increasingly, and the use of real time application data has been increased. Therefore the research and evolution of wireless network for providing of user's mobility and QoS at the same time must be essential. IETF already proposed Mobile IPv6 for providing mobility and RSVP for QoS through resource reservation. These future are expected to be used not only wired network but also wireless network. But there are not yet proposed the integration and efficient interworking of two mechanism. For the solution of problem on efficient interworking issue, for instance signaling overhead and reservation delay increasing due to often handoff, this paper first proposed to use of IETF Cellular IP based on Mobile IPv6 for wireless network model with fast mobility and a improved RSVP mechanism using the flow label field in IPv6 header.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.8
• /
• pp.11-19
• /
• 2012

In this paper, we propose a power-efficient scheduling scheme for lightly loaded multicore processors which contain more processing cores than running tasks. The proposed scheme activates a portion of available cores and inactivates the other unused cores in order to save power consumption. The tasks are assigned to the activated cores based on a heuristic mechanism for fast task assignment. Each activated core executes its assigned tasks with the optimal clock frequency which minimizes the power consumption of the tasks while meeting their deadlines. Evaluation shows that the proposed scheme saves up to 78% power consumption of the previous method which activates as many processing cores as possible for the execution of the given tasks.

• Journal of Digital Convergence
• /
• v.13 no.8
• /
• pp.503-514
• /
• 2015

The effects of Flavonol contents from Ginkgo biloba leaf on anti-atopy activity have not rarely been verified. This study is to investigate the effects of flavonol on Th2 cytokine production in MC/9 mast cells. For this, flavonol was analyzed by ELISA and Real-time PCR. Analysis results showed that flavonol significantly suppressed production of Th2 cytokines(IL-13, MIP-1a) in a dose dependent manner. The mRNA expression of IL-4, IL-5, IL-13, TNF-a were effectively restrained by Flavonol at the concentration 25,50,$100{\mu}g/m{\ell}$. And decrease of expression of NFAT-1, c-jun protein was confirmed by western blot analysis. These results indicate that flavonol has effects of decreasing the Th2 cytokine production in the MC/9 mast cell causing inhibition of transcription factors such as NFAT-1, c-jun. Thus, we would like to brief that flavonol may have the applicability as therapeutic agent for atopic dermatitis.

• Mass Spectrometry Letters
• /
• v.6 no.1
• /
• pp.1-6
• /
• 2015

Direct analysis in real time mass spectrometry (DART-MS) is one of the variants of ambient mass spectrometry. The ionization process of DART-MS is in open environment and only takes few seconds, so it is suitable for fast analysis. Actually, since its introduction in 2005, more and more attentions have been drawn to its various applications due to its excellent properties, e.g., fast analysis, and no or less sample preparation, high salt tolerance and so on. This review summarized the promising features of DART-MS, including its ionization mechanism, equipment modification, wide applications, coupling techniques and extraction strategies before analysis.