• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.645-648
• /
• 2018

기계 독해는 기계가 주어진 본문을 이해하고 질문에 대한 정답을 본문 내에서 찾아내는 문제이다. 본 논문은 질문 유형을 추가하여 정답 선택에 도움을 주도록 설계하였다. 우리는 Person, Location, Date, Number, Why, How, What, Others와 같이 8개의 질문 유형을 나누고 이들이 본문의 중요 자질들과 Attention이 일어나도록 설계하였다. 제안 방법의 평가를 위해 SQuAD의 한국어 번역 데이터와 한국어 Wikipedia로 구축한 K-QuAD 데이터 셋으로 실험을 진행하였다. 제안한 모델의 실험 결과 부분 일치를 인정하여, EM 84.650%, F1 86.208%로 K-QuAD 제안 논문 실험인 BiDAF 모델보다 더 나은 성능을 얻었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.1981-1995
• /
• 2021

When a person tries to conceal emotions, real emotions will manifest themselves in the form of micro-expressions. Research on facial micro-expression recognition is still extremely challenging in the field of pattern recognition. This is because it is difficult to implement the best feature extraction method to cope with micro-expressions with small changes and short duration. Most methods are based on hand-crafted features to extract subtle facial movements. In this study, we introduce a method that incorporates optical flow and deep learning. First, we take out the onset frame and the apex frame from each video sequence. Then, the motion features between these two frames are extracted using the optical flow method. Finally, the features are inputted into an improved MobileNetV2 model, where SVM is applied to classify expressions. In order to evaluate the effectiveness of the method, we conduct experiments on the public spontaneous micro-expression database CASME II. Under the condition of applying the leave-one-subject-out cross-validation method, the recognition accuracy rate reaches 53.01%, and the F-score reaches 0.5231. The results show that the proposed method can significantly improve the micro-expression recognition performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.413-415
• /
• 2019

In this paper, we develop IoT-HiAS(IoT-Human Interactive Advertising Service) for sharing economy. The HiAS device shooting the front of the device via a webcam and recognizes the person in real time through the SSD model using MobileNet. If the number of persons above the set threshold is recognized by counting the recognized person, the advertisement is reproduced on the idle resource through the beam projector. At the same time as the advertisement is reproduced, the captured image of the advertisement start time in the front of the device and the number of recognized persons are sends the IoT server of an oneM2M-compliant HiAS server using the IoT client. When the advertisement is finish, the webcam is shooting the front of the device and the image is sends the IoT server. We implemented the IoT-based human interactive advertising service by transmitting the received data to the advertiser and the advertising producers through the SNS(Social Network Service) agent of the HiAS server.

• Journal of Digital Convergence
• /
• v.16 no.12
• /
• pp.635-640
• /
• 2018

The study conducted a Focus Group Interview (FGI) on college students to identify single-person households and Solitary Deaths of single-person households recognized by college students. This can be provided as basic data to address problems in single-person households and social problems such as future youth solitary death and will contribute to building a social safety net. This study conducted FGI to analyze data for five fourth graders majoring in social welfare. In the case of involuntary independent living, the high poverty and unemployment rate of single-person households was cited as the cause of economic instability, housing problems and emotional relationship formation. He said that he thinks about young loneliness because he has vague fears about what happens in the media and what can happen to them. As the number of young single-person households will inevitably increase in the coming months and economic difficulties are the biggest problem and the biggest cause of young solitude, institutional support is needed first, especially for housing costs.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.4
• /
• pp.519-525
• /
• 2022

The face detection task of detecting a person's face in an image is used as a preprocess or core process in various image processing-based applications. The neural network models, which have recently been performing well with the development of deep learning, are dependent on 2D images, so if noise occurs in the image, such as poor camera quality or pool focus of the face, the face may not be detected properly. In this paper, we propose a face detection method that uses depth information together to reduce the dependence of 2D images. The proposed model was trained after generating and preprocessing depth information in advance using face detection dataset, and as a result, it was confirmed that the FRN model was 89.16%, which was about 1.2% better than the RetinaNet model, which showed 87.95%.

• Korean Journal of Artificial Intelligence
• /
• v.9 no.2
• /
• pp.23-28
• /
• 2021

In this paper, we studied a system that can efficiently build security management for single-person households using Arduino, ESP32-CAM and PIR sensors, and proposed an Android app with an internet connection. The ESP32-CAM is an Arduino compatible board that supports both Wi-Fi, Bluetooth, and cameras using an ESP32-based processor. The PCB on-board antenna may be used independently, and the sensitivity may be expanded by separately connecting the external antenna. This system has implemented an Arduino-based Unauthorized intrusion system that can significantly help prevent crimes in single-person households using the combination of PIR sensors, Arduino devices, and smartphones. unauthorized intrusion system, showing the connection between Arduino Uno and ESP32-CAM and with smartphone applications. Recently, if daily quarantine is underway around us and it is necessary to verify the identity of visitors, it is expected that it will help maintain a safety net if this system is applied for the purpose of facial recognition and restricting some access. This technology is widely used to verify that the characters in the two images entered into the system are the same or to determine who the characters in the images are most similar to among those previously stored in the internal database. There is an advantage that it may be implemented in a low-power, low-cost environment through image recognition, comparison, feature point extraction, and comparison.

• Research in Community and Public Health Nursing
• /
• v.22 no.4
• /
• pp.438-450
• /
• 2011

Purpose: This study was to investigate the effect of home visiting care service and to evaluate the effect from the cost-benefit perspective. Methods: Target participants were enrolled in 2007~2008 for home visiting care and provided with a home visiting nursing service for more than 18 months in J Ward of S City. Of 391 participants, 244 who satisfied the inclusion criteria were used in the final analysis. Cost-benefit analysis was done using the net benefit and benefit/cost ratio. Results: After providing the home care nursing service, the blood pressure control rate increased from 50.8% to 75.4%. Of the subjects, 39.8% maintained their blood pressure level within the target range. As a whole, the net benefit of home visiting care per person ranged from 434,964.86 to 447,112.43 won and the benefit/cost ratio ranged from 2.82 to 2.84. Conclusion: Home visiting care for vulnerable populations with hypertension was effective in both maintaining blood pressure and reducing blood pressure to the target range. Therefore these results are especially useful for establishing the value of home visiting services for policy makers as well as for prioritizing vulnerable populations.

• Korean Journal of Social Welfare
• /
• v.61 no.1
• /
• pp.239-262
• /
• 2009

The purpose of this dissertation is to positively analyze the characteristics of job center consultants' interorganizational relations and factors that influence this linkage in order to better understand the employment service of South Korea. The material utilized for analysis was collected by surveying consultants and managers of 38 job centers and 364 middle managers while applying the method of hierarchical regression. The following is the summary of the research results. First, there is a difference in the influence factor between the number of consultant's organization linkage and frequency of contact. As seen from this result, if the number of organizations a consultant links is affected by the organization's objective working conditions such as the organization structure, position, and primary responsibilities, then the frequency of contact which indicates intimacy by frequent contact depends on the professional ability of the person in charge rather than objective conditions. Second, the analysis result of influence factors according to the type of organization also displayed different influence factors based on the type. Public organizations and economics related organizations are existent in all areas in the form of municipal offices, public corporation, the Chamber of Commerce and Industry, and labor organizations. Therefore, rather than the personal ability of the person in charge or the influence of other work environments, information of the organization and the actual activities of the head are more influential. On the other hand, private organizations embrace service organizations, employment target related organizations, corporations; they are differentially distributed, which is why they are more influential in organizational reform, decentralization, communication, and the professional ability of the person in charge is significantly influential.

• Imaging Science in Dentistry
• /
• v.52 no.3
• /
• pp.239-244
• /
• 2022

Purpose: Despite the proliferation of numerous morphometric and anthropometric methods for sex identification based on linear, angular, and regional measurements of various parts of the body, these methods are subject to error due to the observer's knowledge and expertise. This study aimed to explore the possibility of automated sex determination using convolutional neural networks(CNNs) based on lateral cephalometric radiographs. Materials and Methods: Lateral cephalometric radiographs of 1,476 Iranian subjects (794 women and 682 men) from 18 to 49 years of age were included. Lateral cephalometric radiographs were considered as a network input and output layer including 2 classes(male and female). Eighty percent of the data was used as a training set and the rest as a test set. Hyperparameter tuning of each network was done after preprocessing and data augmentation steps. The predictive performance of different architectures (DenseNet, ResNet, and VGG) was evaluated based on their accuracy in test sets. Results: The CNN based on the DenseNet121 architecture, with an overall accuracy of 90%, had the best predictive power in sex determination. The prediction accuracy of this model was almost equal for men and women. Furthermore, with all architectures, the use of transfer learning improved predictive performance. Conclusion: The results confirmed that a CNN could predict a person's sex with high accuracy. This prediction was independent of human bias because feature extraction was done automatically. However, for more accurate sex determination on a wider scale, further studies with larger sample sizes are desirable.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1061-1066
• /
• 2004

It is certain that the future manufacturing environment will be network-centric and spatially distributed based on Internet. Today, wide variety of distributed computing and communication technologies are available for implementing a system for product data exchange and sharing. One of the technologies that have been received most attentions for product data exchange and sharing is Product Data Management (PDM). PDM tries to integrate and manage process of data and technical documents that are connected to physical product components. In accordance to previous researches about PDM, it can be regarded as an integration tool of many different areas, which ensures that the right information is available to the right person at the right time and in the right form throughout the enterprise. PDM with Web-enabled CAD system is proposed in this paper in order to acknowledge the usefulness of the system mentioned. The system will use Web service on Visual Studio C#.Net to invoke the web application system.