Screening with low-dose spiral computed tomography (LDCT) has been shown to reduce lung cancer mortality by about 20% when compared to standard chest radiography. One of the problems arising from screening programs is that large amounts of CT image data must be interpreted by radiologists. To solve this problem, automated detection of pulmonary nodules is necessary; however, this is a challenging task because of the high number of false positive results. Here we demonstrate detection of pulmonary nodules using six off-the-shelf convolutional neural network (CNN) models after modification of the input/output layers and end-to-end training based on publicly databases for comparative evaluation. We used the well-known CNN models, LeNet-5, VGG-16, GoogLeNet Inception V3, ResNet-152, DensNet-201, and NASNet. Most of the CNN models provided superior results to those of obtained using customized CNN models. It is more desirable to modify the proven off-the-shelf network model than to customize the network model to detect the pulmonary nodules.

In this paper, we propose an indirect PIN entry method that provides enhanced security against smudge, recording, and thermal attacks. Conventional mobile PIN entry methods use on-screen numeric keypad for both use of display and entry. Thus These methods are vulnerable to aforementioned attacks. In our method, passcode is same as that of the conventional PIN entry methods, and that is user-friendly way for mobile device users. Therefore, our method does not reduce user convenience which is one of the advantages of the conventional methods. In addition, our method is not a method of directly touching the on-screen numeric keypad for entering passcode like the conventional PIN methods. Unlike the conventional methods, our method uses an indirect passcode entry method that applied a passcode indicating key. According to the performance comparison result, proposed method provides user convenience similar to the conventional methods, and also provides a higher level of security and safety against recording, smudge, and thermal attacks than the conventional methods.

In this paper, we deal with outlier data problems that occur when constructing a PM_2.5 fine dust forecasting system using a neural network. In general, when learning a neural network, some of the data are not helpful for learning, but rather disturbing. Those are called outlier data. When they are included in the training data, various problems such as overfitting occur. In building a PM_2.5 fine dust concentration forecasting system using neural network, we have found several outlier data in the training data. We, therefore, remove them, and then make learning 3 ways. Over_outlier model removes outlier data that target concentration is low, but the model forecast is high. Under_outlier model removes outliers data that target concentration is high, but the model forecast is low. All_outlier model removes both Over_outlier and Under_outlier data. We compare 3 models with a conventional outlier removal model and non-removal model. Our outlier removal model shows better performance than the others.

A typical algorithm used for image analysis is the Convolutional Neural Network(CNN). R-CNN, Fast R-CNN, Faster R-CNN, etc. have been studied to improve the performance of the CNN, but they essentially require large amounts of data and high algorithmic complexity., making them inappropriate for small and medium-sized services. Therefore, in this paper, the image label prediction algorithm based on CNN with collaborative layer with low complexity, high accuracy, and small amount of data was proposed. The proposed algorithm was designed to replace the part of the neural network that is performed to predict the final label in the existing deep learning algorithm by implementing collaborative filtering as a layer. It is expected that the proposed algorithm can contribute greatly to small and medium-sized content services that is unsuitable to apply the existing deep learning algorithm with high complexity and high server cost.

To enable AI(artificial intelligence) to realize visual emotions, it attempts to create music centered on color, an element that causes emotions in paintings. Traditional image-based music production studies have a limitation in playing notes that are unrelated to the picture because of the absence of musical elements. In this paper, we propose a new algorithm to set the group of music through the average color of the picture, and to produce music after adding diatonic code progression and deleting sound using median value. And the results obtained through the proposed algorithm were analyzed.

Tsunami is a frightful natural disaster that causes severe damages worldwide. To minimize the damage, South Korea has built a tsunami warning system and designated evacuation sites in the east and south coasts. However, such countermeasures have not been verified whether they are adequate to minimize casualties since tsunami rarely occurs in South Korea. Recently, due to increasing earthquakes in the west coast of Japan, the likelihood of South Korea entering the damage area of tsunami rises; thus, in this paper, we develops a simulator based on Unity game engine to simulate the evacuation from tsunami. In order to increase the fidelity of the simulation results, the simulator applies a tsunami simulation model that analyzes coastal inundation based on cellular automata. In addition, the objects included in tsunami evacuation, such as humans, are modeled as an agent model that determines the situation and acts itself, based on the discrete-event system specification (DEVS), a mathematical formalism for describing a discrete event system. The tsunami simulation model and agent models are integrated and visualized in the simulator using Unity game engine. As an example of the use of this simulator, we verify the existing tsunami evacuation site in Gwangalli Beach in Busan and suggest the optimal alternative site minimizing casualties.

A site-specific performance is a genre and a phenomenon that has recently emerged as a recent paradigm of performing arts: a communality. It involves the performativity of the audience and has various phenomena of contemporary performing art. This study aims to figure out methods of performativity. "Lost Missing and Forgotten," which is a collaboration from Finland and Korea created in 2012 and "Trailer," which is held in Copenhagen Denmark in 2010 by Kitt Johnson. Three principles of performativity which have borrowed as critical tools for this study are role reversal, community building, and contact, which based on "The Aesthetics of Performativity," written by Erika Fischer-Lichte in 2017. The study results show that performativity in the vertical structure and horizontal structure of site-specific performance can act as an environmental factor through the audience's body. Additionally, it turned out that role reversal principle was transformation audience to a creator; community building principle led to an interest in neighbors. Lastly, the contact principle was one of the methods to watch the performance.

Today, the media environment has changed dramatically with the development of Internet transmission technology, computer performance, and various media production technologies. In particular, video content is widely used every day through mobile device or computer connected to the Internet. In addition, the ability to produce video content today is more important than ever, as the use of video content is also increasing in the area of information delivery and communication over the Internet. In this study, it was developed a liberal education curriculum in universities for the production of video content. The developed curriculum was composed of a combination of an online curriculum for theory and an offline curriculum for practicum. In the online curriculum, a 54 DOOR contents were developed to allow students to learn the theoretical knowledge and basic making techniques related to the production of video content. In an offline 15-week practicum curriculum, it covered the practical training contents of video content production that should be acquired every week. This developed liberal education curriculum aims to give non-major students in universities the opportunity to learn the technology of producing video content and to foster the ability to communicate in the media environment by using the video content produced.