• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.479-487
• /
• 2021

In this paper, we propose an Sound Event Detection (SED) model using self-training based on a noisy student model. The proposed SED model consists of two stages. In the first stage, a mean-teacher model based on an Residual Convolutional Recurrent Neural Network (RCRNN) is constructed to provide target labels regarding weakly labeled or unlabeled data. In the second stage, a self-training-based noisy student model is constructed by applying different noise types. That is, feature noises, such as time-frequency shift, mixup, SpecAugment, and dropout-based model noise are used here. In addition, a semi-supervised loss function is applied to train the noisy student model, which acts as label noise injection. The performance of the proposed SED model is evaluated on the validation set of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2020 Challenge Task 4. The experiments show that the single model and ensemble model of the proposed SED based on the noisy student model improve F1-score by 4.6 % and 3.4 % compared to the top-ranked model in DCASE 2020 challenge Task 4, respectively.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.32 no.3
• /
• pp.201-206
• /
• 2019

This report constitutes the first demonstration in Korea of single-crystal lateral gallium oxide ($Ga_2O_3$) as a metal-oxide-semiconductor field-effect-transistor (MOSFET), with a breakdown voltage in excess of 480 V. A Si-doped channel layer was grown on a Fe-doped semi-insulating ${\beta}-Ga_2O_3$ (010) substrate by molecular beam epitaxy. The single-crystal substrate was grown by the edge-defined film-fed growth method and wafered to a size of $10{\times}15mm^2$. Although we fabricated several types of power devices using the same process, we only report the characterization of a finger-type MOSFET with a gate length ($L_g$) of $2{\mu}m$ and a gate-drain spacing ($L_{gd}$) of $5{\mu}m$. The MOSFET showed a favorable drain current modulation according to the gate voltage swing. A complete drain current pinch-off feature was also obtained for $V_{gs}<-6V$, and the three-terminal off-state breakdown voltage was over 482 V in a $L_{gd}=5{\mu}m$ device measured in Fluorinert ambient at $V_{gs}=-10V$. A low drain leakage current of 4.7 nA at the off-state led to a high on/off drain current ratio of approximately $5.3{\times}10^5$. These device characteristics indicate the promising potential of $Ga_2O_3$-based electrical devices for next-generation high-power device applications, such as electrical autonomous vehicles, railroads, photovoltaics, renewable energy, and industry.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.6
• /
• pp.839-850
• /
• 2018

The flood damage in urban areas due to torrential rain is increasing with urbanization. For this reason, accurate and rapid flooding forecasting and expected inundation maps are needed. Predicting the extent of flooding for certain rainfalls is a very important issue in preparing flood in advance. Recently, government agencies are trying to provide expected inundation maps to the public. However, there is a lack of quantifying the extent of inundation caused by a particular rainfall scenario and the real-time prediction method for flood extent within a short time. Therefore the real-time prediction of flood extent is needed based on rainfall-runoff-inundation analysis. One/two dimensional model are continued to analyize drainage network, manhole overflow and inundation propagation by rainfall condition. By applying the various rainfall scenarios considering rainfall duration/distribution and return periods, the inundation volume and depth can be estimated and stored on a database. The Rainfall-Duration-Flooding Quantity (RDF) relationship curve based on the hydraulic analysis results and the Self-Organizing Map (SOM) that conducts unsupervised learning are applied to predict flooded area with particular rainfall condition. The validity of the proposed methodology was examined by comparing the results of the expected flood map with the 2-dimensional hydraulic model. Based on the result of the study, it is judged that this methodology will be useful to provide an unknown flood map according to medium-sized rainfall or frequency scenario. Furthermore, it will be used as a fundamental data for flood forecast by establishing the RDF curve which the relationship of rainfall-outflow-flood is considered and the database of expected inundation maps.

• Journal of Navigation and Port Research
• /
• v.42 no.6
• /
• pp.491-500
• /
• 2018

Due to changes in the distribution industry, inner harbors have been on a sharp decline in the recent past. However, through the application of the right development plans, such harbors can be revitalized into vibrant urban areas again. The importance of inner harbors has been recognized by the relevant authorities in Korea which are now pushing forward with redevelopment plans for its inner harbors. This study proposes a new approach to redevelopment plans based on the recognition that inner harbors have unique characteristics involving both inland areas and the ocean. In the study, representative inner harbors were selected and analyzed comparatively according to two distinct concepts of location: the Gateway Concept and the Central Place Concept. Based on these concepts, the conditions of the inner harbors were examined. Their location can be typed, and development directions were proposed according to their types and conditions. However, difficult points such as isolation and separation between an urban district and the harbor area, can be obstructions to their potential revival. An inner harbor needs to be considered as an intermediary, connecting place between the ocean and the city, not as another ordinary development area. In addition, a redevelopment plan should be accompanied by a strategic viewpoint to make the most of this feature.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.309-311
• /
• 2014

Despite increasing demands for big data application based on the analysis of scattered unstructured data, few relevant studies have been reported. Accordingly, the present study suggests a technique enabling a sentence-based semantic analysis by extracting objects from collected web information and automatically analyzing the relationships between such objects with collective intelligence and language processing technology. To be specific, collected information is stored in DBMS in a structured form, and then morpheme and feature information is analyzed. Obtained morphemes are classified into objects of interest, marginal objects and objects of non-interest. Then, with an inter-object attribute recognition technique, the relationships between objects are analyzed in terms of the degree, scope and nature of such relationships. As a result, the analysis of relevance between the information was based on certain keywords and used an inter-object relationship extraction technique that can determine positivity and negativity. Also, the present study suggested a method to design a system fit for real-time large-capacity processing and applicable to high value-added services.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.4
• /
• pp.77-83
• /
• 2019

With the increase in population, the number of such senior citizens is increasing day by day. These senior citizens have a variety of care needs, but there are not enough health workers to look after them. Ambient Assisted Living (AAL) aims at ensuring the safety and health quality of the older adults and extending the number of years the senior citizens can live independently in an environment of their own preference. AAL provides a system comprising of smart devices, medical sensors, wireless networks, computer and software applications for healthcare monitoring. AAL can be used for various purposes like preventing, curing, and improving wellness and health conditions of older adults. While information security and privacy are critical to providing assurance that users of AAL systems are protected, few studies take into account this feature. In this paper, we propose a secure and lightweight authentication scheme for the AAL systems. The proposed authentication scheme not only supports several important security requirements needed by the AAL systems, but can also withstand various types of attacks. Also, the security analysis results are presented to show the proposed authentication scheme is more secure and efficient rather than existing authentication schemes.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.495-505
• /
• 2019

Typical algorithms of deep learning include CNN(Convolutional Neural Networks), which are mainly used for image recognition, and RNN(Recurrent Neural Networks), which are used mainly for speech recognition and natural language processing. Among them, CNN is able to learn from filters that generate feature maps with algorithms that automatically learn features from data, making it mainstream with excellent performance in image recognition. Since then, various algorithms such as R-CNN and others have appeared in object detection to improve performance of CNN, and algorithms such as YOLO(You Only Look Once) and SSD(Single Shot Multi-box Detector) have been proposed recently. However, since these deep learning-based detection algorithms determine the success of the detection in the still images, stable object tracking and detection in the video requires separate tracking capabilities. Therefore, this paper proposes a method of combining Kalman filters into deep learning-based detection networks for improved object tracking and detection performance in the video. The detection network used YOLO v2, which is capable of real-time processing, and the proposed method resulted in 7.7% IoU performance improvement over the existing YOLO v2 network and 20 fps processing speed in FHD images.

• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.1-8
• /
• 2019

In this paper, we propose an improved facial skin color extraction method to solve the problem that facial surface is lost due to shadow or illumination in skin color extraction process and skin color extraction is not possible. In the conventional HSV method, when facial surface is brightly illuminated by light, the skin color component is lost in the skin color extraction process, so that a loss area appears on the face surface. In order to solve these problems, we extract the skin color, determine the elements in the H channel value range of the skin color in the HSV color space among the lost skin elements, and combine the coordinates of the lost part with the coordinates of the original image, To minimize the number of In the face detection process, the face was detected using the LBP Cascade Classifier, which represents texture feature information in the extracted skin color image. Experimental results show that the proposed method improves the detection rate and accuracy by 5.8% and 9.6%, respectively, compared with conventional RGB and HSV skin color extraction and face detection using the LBP cascade classifier method.

• Journal of Korea Water Resources Association
• /
• v.54 no.5
• /
• pp.289-299
• /
• 2021

This study analyzes bifurcation characteristics of the Seolmacheon experimental catchment by extracting the shape variation of channel network due to variable scale of source basin or threshold area. As the area of source basin decreases, a bifurcation process of channel network occurs within the basin of interest, resulting in the elongation of channel network (increase of total channel length) as well as the expansion of channel network (increase of the source number). In the former case, the elongation of channel reaches overwhelms the generation of sources, whereas, in the latter case, the drainage path network tends to fulfill the inner space of the basin of interest reflecting the opposite trend. Therefore, scale invariance of natural channel network could be expressed to be a balanced geomorphologic feature between the elongation of channel network and the expansion of channel network due to decrease of source basin scale. The bifurcation structure of the Seolmacheon experimental catchment can be characterized by the coexistence of the elongation and scale invariance of channel network, and thus a further study is required to find out which factor is more crucial to rainfall transformation into runoff.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.139-147
• /
• 2021

This paper introduces and implements a Sound Event Detection (SED) system based on weakly-supervised learning where only part of the data is labeled, and analyzes the effect of parameters. The SED system estimates the classes and onset/offset times of events in the acoustic signal. In order to train the model, all information on the event class and onset/offset times must be provided. Unfortunately, the onset/offset times are hard to be labeled exactly. Therefore, in the weakly-supervised task, the SED model is trained by "strongly labeled data" including the event class and activations, "weakly labeled data" including the event class, and "unlabeled data" without any label. Recently, the SED systems using the mean-teacher model are widely used for the task with several parameters. These parameters should be chosen carefully because they may affect the performance. In this paper, performance analysis was performed on parameters, such as the feature, moving average parameter, weight of the consistency cost function, ramp-up length, and maximum learning rate, using the data of DCASE 2020 Task 4. Effects and the optimal values of the parameters were discussed.