Deep learning has been used for various processing centered on image recognition. One core algorithms of the deep learning, convolutional neural network is an deep neural network that specialized in image recognition. In this paper, we use a convolutional neural network to classify forest insects and propose an optimization method. Experiments were carried out by combining two weight initialization and six parameter update methods. As a result, the Xavier-SGD method showed the highest performance with an accuracy of 82.53% in the 12 different combinations of experiments. Through this, the latest learning algorithms, which complement the disadvantages of the previous parameter update method, we conclude that it can not lead to higher performance than existing methods in all application environments.

Sound based machine fault diagnosis is the process consisting of detecting automatically the damages that affect the machines by analyzing the sounds they produce during their operating time. The collected sounds being inevitably corrupted by random disturbance, the most important part of the diagnosis consists of discovering the hidden elements inside the data that can reveal the faulty patterns. This paper presents a novel feature extraction methodology that combines various digital signal processing and pattern recognition methods for the analysis of the sounds produced by the drills. Using the Fourier analysis, the magnitude spectrum of the sounds are extracted, converted into two-dimensional vectors and uniformly normalized in such a way that they can be represented as 8-bit grayscale images. Histogram equalization is then performed over the obtained images in order to adjust their very poor contrast. The obtained contrast enhanced images will be used as the features of our diagnosis system. Finally, principal component analysis is performed over the image features for reducing their dimensions and a nonlinear classifier is adopted to produce the final response. Unlike the conventional features, the results demonstrate that the proposed feature extraction method manages to capture the hidden health patterns of the sound.

With the introduction of the e-learning since 2000, the place of the education has not been limited to off-line, but the range of it has become broader in online. The e-learning market has evolved steadily over time. With the advent of the term "Edu-tech", which means a combination of education and technology, various IT technologies have incorporated education. Particularly, the Korean education market collects patterns by computerizing the learning history in classes taught according to curriculums. Because of that environment, various personalized learning services have been developed which maximize the effect of the learning. These services have qualitative differences depending on how many data is accumulated and algorithms are developed for the precise analysis. The purpose of this study is to recognize and data-ize OMR marking by the most suitable method to convert analog data into digital data without harming the Korean education system.

As an important key technology using on electronic devices, face recognition has become one of the hottest technology recently. The traditional BP Neural network has a strong ability of self-learning, adaptive and powerful non-linear mapping but it also has disadvantages such as slow convergence speed, easy to be traversed in the training process and easy to fall into local minimum points. So we come up with an algorithm based on BP neural network but also combined with the PCA algorithm and other methods such as the elastic gradient descent method which can improve the original network to try to improve the whole recognition efficiency and has the advantages of both PCA algorithm and BP neural network.

The choice of the motion features influences the result of the human action recognition method directly. Many factors often influence the single feature differently, such as appearance of the human body, environment and video camera. So the accuracy of action recognition is restricted. On the bases of studying the representation and recognition of human actions, and giving fully consideration to the advantages and disadvantages of different features, the Dense Trajectories(DT) algorithm is a very classic algorithm in the field of behavior recognition feature extraction, but there are some defects in the use of optical flow images. In this paper, we will use the improved Dense Trajectories(iDT) algorithm to optimize and extract the optical flow features in the movement of human action, then we will combined with Support Vector Machine methods to identify human behavior, and use the image in the KTH database for training and testing.

Recently, the numerous authentication and key distribution protocol for NFC payment environment have been proposed using public key cryptosystems. However, these protocol are vulnerable to quantum computing attack because quantum computing can solve factoring and discrete logarithm problem effectively using Grover and Shor's algorithm. For these reason, the secure authentication and key distribution have become a very important security issue in order to prevent quantum computing attacks. Therefore, to ensure user's payment information and privacy, we propose a secure quantum resistant authentication and key distribution protocol for NFC payment environments.

In this paper, we provide the performance analyses of a dual-hop amplify-and-forward(AF) relay transmission composed of asymmetric radio-frequence(RF) and free-space optical(FSO) links. In the mixed RF/FSO system, a relay is equipped with two receive antennas for RF signals and one additional transmit antenna for FSO signals. In order to improve a performance of RF link, a switch-and-stay (SSC) diversity technique is applied at the relay which can provide a proper link performance with a low complexity. Specifically, we offer the performance analyses of the proposed system in terms of outage probability and secrecy outage probability. In numerical examples, we compare the system performances with no diversity and selection combining systems and verify our analytical results via computer-based Monte-Carlo simulations.

The application of analytical techniques for the rational determination of problems arising from management science and corporate management. In a way that is the opposite of the repair plan that can predict accurate results is increasing utilization of the complex-based analysis methodology. In this study, we examined the application of physical space and the methodological utilization of the pedestrian model analysis that applied the simulation to the Health Checkup Center. The conclusions are as follows. First, the spatial analysis and measurement for empirical research has confirmed that the efficiency assessment through the pedestrian model simulation can lead to an objective evaluation. Second, it seems to be able to reduce the queue through a change in the number of services of the low-pressure and the hearing laboratory, the recovery room with a high proportion of male disturbances and relatively long use time. The third, the spatial density analysis and the time required to reduce the density change in comparative analysis, and the spatial layout changes, the increase in the capacity of 80 people, approximately 16 minutes to shorten the process duration.

Container load problems are conventional issues in history. Loading cylindrical or rectangle packages in restricted space is required to be efficient to minimize remaining area. Many people have depended on experience or known simple algorithms for a long time. Recently systematic algorithms are published over the world with developed computer technology. Maximal rectangles algorithm is well known to load arbitrary rectangles in a larger rectangle container with the organized manners. However, this algorithm has significant drawbacks, which are influenced by orientation of input load. We propose post-selective method to complement these orientation effects, which is found out to improve the loading efficiency.

As the speed and extent of the expansion of the retro culture continue to expand, interest in industrial values related to it as well as a cultural phenomenon is also increasing. Under these circumstances it is meaningful that the planning of movies using cultural memory can increase the possibility of continuous development of retro contents. In this article, I will look at the possibilities of retro film using cultural memory and the memory industry. The purpose of this research is to find creative and sustainable content production methods.

Developing a new texture path tool is crucial in 3D industries to automatically manage and set up texture paths for characters and environment modelings. For a better study, I compare the problems, methods and functions of the new tool(SongRepath tool) with other commonly used tools such as File Path Editor, Genie, He Texture Path. Next, I analyze the top down method system for these commonly used tools, and create a new bottom-up algorithm for a better use and efficiency. Finally, analyzing and comparing SongRepath tool with other tools in terms of operation in convenience, steps for results, checking number of Texture files, checking the numbers of new paths, checking the numbers of files not found, checking the numbers of path not changed, errors or problems, and availability of cancelation while proceeding the result. As a result, this study shows that SongRepath tool has less errors and saves more time on the processing works.