• Journal of Broadcast Engineering
• /
• v.26 no.5
• /
• pp.608-620
• /
• 2021

Recent deep learning-based face super-resolution (FSR) works showed significant performances by utilizing facial prior knowledge such as facial landmark and dictionary that reflects structural or semantic characteristics of the human face. However, most of these methods require additional processing time and memory. To solve this issue, this paper propose an efficient FSR models using knowledge distillation techniques. The intermediate features of teacher network which contains dictionary information based on major face regions are transferred to the student through adversarial multi-scale features distillation. Experimental results show that the proposed model is superior to other SR methods, and its effectiveness compare to teacher model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.77-83
• /
• 2019

With the recent development of medical technology, the diagnosis rate of type 1 diabetes is increasing and patients are increasing. However, diabetes education content is not aligned with the interest level of children. As a result of the interviews with experts, It was found that the measures of coping with the change of blood sugar and the behavior therapy require steady and repetitive learning. Therefore, this study proposes Edutainment content which can be repeatedly Learned by 10~11year old children. For effective learning, the contents of the laboratory practice were constructed and the hybrid method was used for the repetitive learning. Usability test showed that this configuration is effective. This study is expected to contribute to the study of diabetic education content that is suitable for the children's level of understanding which will be developed easily in the future.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.2
• /
• pp.132-140
• /
• 2019

In this paper, we propose a classification method for radar signals depending on the type of threat by applying machine learning to parameter data of radar signals. Currently, the army uses a library of mapping relations between the parameters and the types of threat to recognize threat signals. This approach has certain limitations when classifying signals and recognizing new types of threat or types of threat that do not exist in the current libraries. In this paper, we propose an automatic radar signal classification method depending on the type of threat that uses only parameter data without a library. A convolutional neural network is used as the classifier and machine learning is applied to train the classifier. The proposed method does not use a library, and hence, can classify threat signals that are new or do not exist in the current library.

• Journal of Internet Computing and Services
• /
• v.22 no.5
• /
• pp.79-86
• /
• 2021

As the usefulness of big data analysis has been drawing attention, many studies in the business research area begin to use big data to predict firm performance. Previous studies mainly rely on data outside of the firm through news articles and social media platforms. The voices within the firm in the form of employee satisfaction or evaluation of the strength and weakness of the firm can potentially affect firm value. However, there is insufficient evidence that online employee reviews are valid to predict firm value because the data is relatively difficult to obtain. To fill this gap, from 2014 to 2019, we employed 97,216 reviews collected by JobPlanet, an online firm review website in Korea, and developed a machine learning-based predictive model. Among the proposed models, the LSTM-based model showed the highest accuracy at 73.2%, and the MAE showed the lowest error at 0.359. We expect that this study can be a useful case in the field of firm value prediction on domestic companies.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.935-943
• /
• 2020

In this paper, we propose a method using deep learning for high-resolution display of phase holograms. If a general interpolation method is used, the brightness of the reconstruction result is lowered, and noise and afterimages occur. To solve this problem, a hologram was trained with a neural network structure that showed good performance in the single-image super resolution (SISR). As a result, it was possible to improve the problem that occurred in the reconstruction result and increase the resolution. In addition, by adjusting the number of channels to increase performance, the result increased by more than 0.3dB in same training.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.576-583
• /
• 2020

This paper proposes a technique to determine the spam comments on YouTube, which have recently seen tremendous growth. On YouTube, the spammers appeared to promote their channels or videos in popular videos or leave comments unrelated to the video, as it is possible to monetize through advertising. YouTube is running and operating its own spam blocking system, but still has failed to block them properly and efficiently. Therefore, we examined related studies on YouTube spam comment screening and conducted classification experiments with six different machine learning techniques (Decision tree, Logistic regression, Bernoulli Naive Bayes, Random Forest, Support vector machine with linear kernel, Support vector machine with Gaussian kernel) and ensemble model combining these techniques in the comment data from popular music videos - Psy, Katy Perry, LMFAO, Eminem and Shakira.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.487-495
• /
• 2021

The purpose of this study is to make it easier, more fun and more convenient to learn a foreign language through the development of an efficient audio contents platform that utilizes each person's native language ability. In order to achieve the goal to produce audio contents centering on the native language used in real life. Contents that are created without much effort in daily life could be used as precious contents for foreign language learners to learn the natural use of the language. Currently, most of the online foreign language learning platforms have problems with the contents depletion and the low practicality of contents. Accordingly, I am expecting this platform improves the existing shortcomings, giving foreign language learners the opportunity to learn a foreign language more realistically and at the same time giving native speakers an opportunity to generate additional revenue by utilizing their spare time.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.283-294
• /
• 2022

The feature map used in the network for deep learning generally has larger data than the image and a higher compression rate than the image compression rate is required to transmit the feature map. This paper proposes a method for transmitting a pyramid feature map with high compression rate, which is used in a network with an FPN structure that has robustness to object size in deep learning-based image processing. In order to efficiently compress the pyramid feature map, this paper proposes a structure that predicts a pyramid feature map of a level that is not transmitted with pyramid feature map of some levels that transmitted through the proposed prediction network to efficiently compress the pyramid feature map and restores compression damage through the proposed reconstruction network. Suggested mAP, the performance of object detection for the COCO data set 2017 Train images of the proposed method, showed a performance improvement of 31.25% in BD-rate compared to the result of compressing the feature map through VTM12.0 in the rate-precision graph, and compared to the method of performing compression through PCA and DeepCABAC, the BD-rate improved by 57.79%.

• Journal of Industrial Convergence
• /
• v.20 no.7
• /
• pp.37-50
• /
• 2022

This study intends to analyze the learning effect of asynchronous online classes used in education fields around the world after the COVID-19 pandemic. To this end, we compared and analyzed the lecture evaluation and final exam questions of the HTML5 web programming course, which was conducted offline in 2019 and asynchronously online in 2020 due to COVID-19. As a result of the analysis, no significant difference was drawn between the two teaching methods in the lecture evaluation score and final exam score. However, contrary to concerns about the application of online classes to the entire curriculum, the lecture evaluation and final exam scores of the video-based online classes were high, suggesting the possibility that online classes could be more effective than offline classes if well organized and managed in the future.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.174-184
• /
• 2022

3D human pose estimation models can be classified into a multi-view model and a single-view model. In general, the multi-view model shows superior pose estimation performance compared to the single-view model. In the case of the single-view model, the improvement of the 3D pose estimation performance requires a large amount of training data. However, it is not easy to obtain annotations for training 3D pose estimation models. To address this problem, we propose a method to generate pseudo ground-truths of multi-view human pose data from a multi-view model and exploit the resultant pseudo ground-truths to train a single-view model. In addition, we propose a multi-view consistency loss function that considers the consistency of poses estimated from multi-view images, showing that the proposed loss helps the effective training of single-view models. Experiments using Human3.6M and MPI-INF-3DHP datasets show that the proposed method is effective for training single-view 3D human pose estimation models.