• Journal of Korea Entertainment Industry Association
• /
• 제15권4호
• /
• pp.267-277
• /
• 2021

Both teachers and learners should participate in non-face-to-face class due to COVID-19. The non-face-to-face class has brought about many problems, where they made adequate preparations for such abrupt situation. This study attempted to understand and improve problems occurring during non-face-to-face midi class. The findings are as follows: First, there were differences in equipment available to contact and non-face-to-face class. Such a problem could be improved by using Reaper, DAW which can be installed and freely utilized without any functional limits, regardless of the types of operating systems. Second, latency could not be reduced, when the screen share function of Zoom was used, since it was impossible to select audio interface's drivers in DAW. This problem was improved by again receiving audio output as input and sending it, from the perspectives of teachers. In addition, learners who used the operating system of Windows and have no audio interfaces usually suffer from latency during practices. The latency can be reduced by installing Asio4all. Third, image degradation and screen disconnection phenomena occurred due to the lack of resource. Two computers were connected by using a capture board and the screen disconnection phenomena could be improved by distributing resources and maintaining high-resolution. The system for allowing non-face-to-face midi class could be successfully established, as one more computer was connected by using Vienna Ensemble Pro and more plug-ins were used by securing additional resources. Consequently, the problems of non-face-to-face midi class could be understood and improved.

• Korean Journal of Remote Sensing
• /
• 제38권6_4호
• /
• pp.1889-1900
• /
• 2022

Roads are an essential factor in the physical functioning of modern society. The spatial information of the road has much longer update cycle than the traffic situation information, and it is necessary to generate the information faster and more accurately than now. In this study, as a way to achieve that goal, the Pan-sharpening technique was applied to satellite images of Kompsat 3 and 3A to improve spatial resolution. Then, the data were used for road extraction using the semantic segmentation technique, which has been actively researched recently. The acquired Kompsat 3/3A pan-sharpened images were trained by putting it into a U-Net based segmentation model along with Massachusetts road data, and the applicability of the images were evaluated. As a result of training and verification, it was found that the model prediction performance was maintained as long as certain conditions were maintained for the input image. Therefore, it is expected that the possibility of utilizing satellite images such as Kompsat satellite will be even higher if rich training data are constructed by applying a method that minimizes the impact of surrounding environmental conditions affecting models such as shadows and surface conditions.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• 제40권5호
• /
• pp.381-391
• /
• 2022

This paper is intended to find one of the prevailing deep learning models that are a type of AI (Artificial Intelligence) that helps rapidly detect damaged buildings where disasters occur. The models selected are SSD-512, RetinaNet, and YOLOv3 which are widely used in object detection in recent years. These models are based on one-stage detector networks that are suitable for rapid object detection. These are often used for object detection due to their advantages in structure and high speed but not for damaged building detection in disaster management. In this study, we first trained each of the algorithms on xBD dataset that provides the post-disaster imagery with damage classification labels. Next, the three models are quantitatively evaluated with the mAP(mean Average Precision) and the FPS (Frames Per Second). The mAP of YOLOv3 is recorded at 34.39%, and the FPS reached 46. The mAP of RetinaNet recorded 36.06%, which is 1.67% higher than YOLOv3, but the FPS is one-third of YOLOv3. SSD-512 received significantly lower values than the results of YOLOv3 on two quantitative indicators. In a disaster situation, a rapid and precise investigation of damaged buildings is essential for effective disaster response. Accordingly, it is expected that the results obtained through this study can be effectively used for the rapid response in disaster management.

• The Journal of the Convergence on Culture Technology
• /
• 제7권2호
• /
• pp.85-95
• /
• 2021

Currently, we stand between computers as creative tools and computers as creators. A new genre of movies, which can be called a post-cinema situation, is emerging. This paper aims to diagnose the possibility of the emergence of AI cinema. To confirm the possibility of AI cinema, it was examined through a case study whether the creation of a story, narrative, image, and sound, which are necessary conditions for film creation, is possible by artificial intelligence. First, we checked the visual creation of AI painting algorithms Obvious, GAN, and CAN. Second, AI music has already entered the distribution stage in the market in cooperation with humans. Third, AI can already complete drama scripts, and automatic scenario creation programs using big data are also gaining popularity. That said, we confirmed that the filmmaking requirements could be met with AI algorithms. From the perspective of Manovich's 'AI Genre Convention', web documentaries and desktop documentaries, typical trends post-cinema, can be said to be representative genres that can be expected as AI cinemas. The conditions for AI, web documentaries and desktop documentaries to exist are the same. This article suggests a new path for the media of the 4th Industrial Revolution era through research on AI as a creator of post-cinema.

• Journal of the Korean Society of Radiology
• /
• 제16권7호
• /
• pp.961-968
• /
• 2022

Rotator cuff tears are a leading cause of shoulder pain in adults. Due to the increase in social activities, the number of patients complaining of shoulder pain is increasing, and interest in shoulder diseases is also increasing. With the development of ultrasound equipment, the sensitivity and specificity of diagnosis are high, and it is used to diagnose rotator cuff tears in musculoskeletal disease. Ultrasound is recognized as a complementary method to MRI examination in rotator cuff tears. Therefore, this study aimed to find out the diagnostic usefulness of ultrasound and MRI examinations in the diagnosis of shoulder rotator cuff tears.A retrospective analysis was performed on 262 patients who were diagnosed with final rotator cuff damage by arthroscopy after completing ultrasound and MRI examinations. Sensitivity, feature, positive predictive value, image predictive value, and touch were disassembled for the test results. In addition, the degree of clavicular tear was scored and recorded in 5 stages. Ultrasound examination was similar to MRI examination results for both full-thickness and partial tears, and there was no statistically significant difference. Partial tear test results showed higher positive predictive value and accuracy than MRI test. In conclusion, ultrasound can be fully utilized as a screening test for rotator cuff disease, and it is thought that it will be selected and used clinically according to the patient's constitution and situation.

• The Journal of the Korea institute of electronic communication sciences
• /
• 제17권6호
• /
• pp.1137-1144
• /
• 2022

In this study, an artificial intelligence(AI) was developed to help with facial expression practice in order to express emotions. The developed AI used multimodal inputs consisting of sentences and facial images for deep neural networks (DNNs). The DNNs calculated similarities between the emotions predicted by the sentences and the emotions predicted by facial images. The user practiced facial expressions based on the situation given by sentences, and the AI provided the user with numerical feedback based on the similarity between the emotion predicted by sentence and the emotion predicted by facial expression. ResNet34 structure was trained on FER2013 public data to predict emotions from facial images. To predict emotions in sentences, KoBERT model was trained in transfer learning manner using the conversational speech dataset for emotion classification opened to the public by AIHub. The DNN that predicts emotions from the facial images demonstrated 65% accuracy, which is comparable to human emotional classification ability. The DNN that predicts emotions from the sentences achieved 90% accuracy. The performance of the developed AI was evaluated through experiments with changing facial expressions in which an ordinary person was participated.

• Journal of Korea Society of Industrial Information Systems
• /
• 제27권4호
• /
• pp.19-27
• /
• 2022

In general, the performance of ML(Machine Learning) application is determined by various factors such as the type of ML model, the size of model (number of parameters), hyperparameters setting during the training, and training data. In particular, the recognition accuracy of ML may be deteriorated or experienced overfitting problem if the amount of dada used for training is insufficient. Existing studies focusing on image recognition have widely used open datasets for training and evaluating the proposed ML models. However, for specific applications where the sensor used, the target of recognition, and the recognition situation are different, it is necessary to build the dataset manually. In this case, the performance of ML largely depends on the quantity and quality of the data. In this paper, training data used for motion recognition application is augmented using the kernel density estimation algorithm which is a type of non-parametric estimation method. We then compare and analyze the recognition accuracy of a ML application by varying the number of original data, kernel types and augmentation rate used for data augmentation. Finally experimental results show that the recognition accuracy is improved by up to 14.31% when using the narrow bandwidth Tophat kernel.

• Journal of the Korea Computer Graphics Society
• /
• 제28권2호
• /
• pp.1-9
• /
• 2022

In this paper, we propose a vector-based augmentation technique that can generate data required for crack detection and a ConvNet(Convolutional Neural Network) technique that can learn it. Detecting cracks quickly and accurately is an important technology to prevent building collapse and fall accidents in advance. In order to solve this problem with artificial intelligence, it is essential to obtain a large amount of data, but it is difficult to obtain a large amount of crack data because the situation for obtaining an actual crack image is mostly dangerous. This problem of database construction can be alleviated with elastic distortion, which increases the amount of data by applying deformation to a specific artificial part. In this paper, the improved crack pattern results are modeled using ConvNet. Rather than elastic distortion, our method can obtain results similar to the actual crack pattern. By designing the crack data augmentation based on a vector, rather than the pixel unit used in general data augmentation, excellent results can be obtained in terms of the amount of crack change. As a result, in this paper, even though a small number of crack data were used as input, a crack database can be efficiently constructed by generating various crack directions and patterns.

• The Mathematical Education
• /
• 제61권3호
• /
• pp.419-439
• /
• 2022

This study is based on Freudenthal's mathmatising process and the didactical phenomenology of linear function concept, I have described and examined the process in which students represent the constant rate of change into tables, graphs and equations and, in this way, how they construct mental objects and essence of the linear function concept. The students used the proportionality as composite units, when they represented the phenomenon with constant rate of change into tables. When representing in graphs, all but one student represented it into a line. There were differences among the students in the level they were using the given conditions, co-variation perspective, and corresponding rules when formulating equations. The students compared the relationship between two variables in a multiplicative way, and under the guidance of teachers they reached to the understanding that its relationship becomes a constant. Moreover, they could construct mental objects of a constant rate of change, understanding the situation where the relationship between time difference and distance difference becomes one value, namely speed. The students had difficulties in connecting the rate of change with the inclination of a line. The students constructed the essence (concept) of linear functions, after building and organizing the image that the rate of change is constant, the graph is linear, and the equation is formulated as y=ax+b (a: inclination, b: intercept).

• The Korean Journal of Applied Statistics
• /
• 제36권2호
• /
• pp.101-114
• /
• 2023

The data augmentation technique is effectively used to solve the problem of overfitting the model by allowing the training dataset to be viewed from various perspectives. In addition to image augmentation techniques such as rotation, cropping, horizontal flip, and vertical flip, occlusion-based data augmentation methods such as Cutmix and Cutout have been proposed. For models based on speech data, it is possible to use an occlusion-based data-based augmentation technique after converting a 1D speech signal into a 2D spectrogram. In particular, SpecAugment is an occlusion-based augmentation technique for speech spectrograms. In this study, we intend to compare and study data augmentation techniques that can be used in the problem of false-voice detection. Using data from the ASVspoof2017 and ASVspoof2019 competitions held to detect fake audio, a dataset applied with Cutout, Cutmix, and SpecAugment, an occlusion-based data augmentation method, was trained through an LCNN model. All three augmentation techniques, Cutout, Cutmix, and SpecAugment, generally improved the performance of the model. In ASVspoof2017, Cutmix, in ASVspoof2019 LA, Mixup, and in ASVspoof2019 PA, SpecAugment showed the best performance. In addition, increasing the number of masks for SpecAugment helps to improve performance. In conclusion, it is understood that the appropriate augmentation technique differs depending on the situation and data.