• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.11
• /
• pp.529-536
• /
• 2015

When rendering both three-dimensional objects and photo images together, the non-photorealistic rendering results are in visual discord since the two contents have their own independent color distributions. This paper proposes a non-photorealistic rendering technique which renders both three-dimensional objects and photo images such as cartoons and sketches. The proposed technique computes the color distribution property of the photo images and reduces the number of colors of both photo images and 3D objects. NPR is performed based on the reduced colormaps and edge features. To enhance the natural scene presentation, the image region segmentation process is preferred when extracting and applying colormaps. However, the image segmentation technique needs a lot of computational operations. It takes a long time for non-photorealistic rendering for large size frames. To speed up the time-consuming segmentation procedure, we use GPGPU for the parallel computing using the GPU. As a result, we significantly improve the execution speed of the algorithm.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.3
• /
• pp.316-323
• /
• 2010

We propose a method to find flows of transit users in the subway transportation network of the metropolitan Seoul and analyze the passenger flows on some central links of the network. The transportation network consists of vertices for subway stops, edges for links between two adjacent subway stops, and flows on the edges' Each subway transit user makes a passenger flow along edges of the shortest path from the origin stop to the destination stop in his trip. In this paper, we have developed a new algorithm to find the passenger flow of each link in the subway network from a large trip-transaction database of subway transit users. We have applied the algorithm to find the passenger flows from one day database of about 5 million transactions by the subway transit users. As results of the experiments, the travel behavior on 4 central subway links is analyzed in passenger flows and top 10 flows among all subway links are explained in a table.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.9
• /
• pp.889-897
• /
• 2010

Recently, a cutting-edge large-scale image database system has demanded these attributes: search with alarming speed, performs with high accuracy, archives efficiently and much more. An image identifier (descriptor) is for measuring the similarity of two images which plays an important role in this system. The extraction method of an image identifier can be roughly classified into two methods: a local and global method. In this paper, the proposed image identifier, LFH(Local Feature's Histogram), is obtained by a histogram of robust and distinctive local descriptors (features) constrained by a district sub-division of a local region. Furthermore, LFH has not only the properties of a local and global descriptor, but also can perform calculations at a magnificent clip to determine distance with pinpoint accuracy. Additionally, we suggested a way to extract LFH via GPU (OpenGL and GLSL). In this experiment, we have compared the LFH with SIFT (local method) and EHD (global method) via storage capacity, extraction and retrieval time along with accuracy.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.11
• /
• pp.841-845
• /
• 2009

The independent cascade model has been widely used to analyze information diffusion in the blog world. In this paper, we propose a new method to construct a blog network for applying the independent cascade model to analyzing of information diffusion in a blog world. To construct a blog network, the proposed method establishes the edge between two users and calculates diffusion probabilities between them by analyzing the activities happened between two users. To calculate diffusion probabilities, the method exploits the ratio of the number of documents actually diffused to a specific user to that of documents written for the purpose of being diffused to other blogs. The experimental result using a real world blog data demonstrates that our method reflects actual information diffusion in a blog world better than existing ones.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.1
• /
• pp.110-114
• /
• 2010

We describe a new method for text categorization using TextRank algorithm. Text categorization is a problem that over one pre-defined categories are assigned to a text document. TextRank algorithm is a graph-based ranking algorithm. If we consider that each word is a vertex, and co-occurrence of two adjacent words is a edge, we can get a graph from a document. After that, we find important words using TextRank algorithm from the graph and make feature which are pairs of words which are each important word and a word adjacent to the important word. We use classifiers: SVM, Na$\ddot{i}$ve Bayesian classifier, Maximum Entropy Model, and k-NN classifier. We use non-cross-posted version of 20 Newsgroups data set. In consequence, we had an improved performance in whole classifiers, and the result tells that is a possibility of TextRank algorithm in text categorization.

• The Journal of Korean Association of Computer Education
• /
• v.21 no.6
• /
• pp.39-47
• /
• 2018

This paper dealt with investigating the number of minimal spanning ladders originated from ladder game and their properties as well as the related computational thinking aspects. The author modified the filtering techniques to enhance Mathematica project where a new type of graph was generated based on the algorithm using a generator of firstly found minimal spanning graph by repeatedly applying independent ladder operator to a subsequence of ladder sequence. The newly produced YC graphs had recursive and hierarchical graph structures and showed the properties of edge-symmetric. As the computational complexity increased the author divided the whole search space into the each floor of the newly generated minimal spanning graphs for the (5, 10) YC graph and the higher (6, 15) YC graph. It turned out that the computational thinking capabilities such as data visualization, abstraction, and parallel computing with Mathematica contributed to enumerating the new YC graphs in order to investigate their structures and properties.

• Smart Media Journal
• /
• v.8 no.3
• /
• pp.95-103
• /
• 2019

The global manufacturing industry has reached the limit to growth due to a long-term recession, the rise of labor cost and raw material. As a solution to these difficulties, we promote the 4th Industry Revolution based on ICT and sensor technology. Following this trend, this paper proposes the design of a model using manufacturing data in the protection film process for smart manufacturing innovation. In the protective film process, the manufacturing data of temperature, pressure, humidity, and motion and thermal image are acquired by various sensors for the raw material blending, stirring, extrusion, and inspection processes. While the acquired manufacturing data is stored in mass storage, A.I. platform provides time-series image analysis and its visualization.

• Advances in nano research
• /
• v.10 no.2
• /
• pp.151-163
• /
• 2021

The main target of this study is to investigate nonlinear transient responses of moving polymer nano-size plates fortified by means of Graphene Platelets (GPLs) and resting on a Winkler-Pasternak foundation under a transverse pressure force and a temperature variation. Two graphene spreading forms dispersed through the plate thickness are studied, and the Halpin-Tsai micro-mechanics model is used to obtain the effective Young's modulus. Furthermore, the rule of mixture is employed to calculate the effective mass density and Poisson's ratio. In accordance with the first order shear deformation and von Karman theory for nonlinear systems, the kinematic equations are derived, and then nonlocal strain gradient scheme is used to reflect the effects of nonlocal and strain gradient parameters on small-size objects. Afterwards, a combined approach, kinetic dynamic relaxation method accompanied by Newmark technique, is hired for solving the time-varying equation sets, and Fortran program is developed to generate the numerical results. The accuracy of the current model is verified by comparative studies with available results in the literature. Finally, a parametric study is carried out to explore the effects of GPL's weight fractions and dispersion patterns, edge conditions, softening and hardening factors, the temperature change, the velocity of moving nanoplate and elastic foundation stiffness on the dynamic response of the structure. The result illustrates that the effects of nonlocality and strain gradient parameters are more remarkable in the higher magnitudes of the nanoplate speed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1500-1505
• /
• 2019

Currently, as part of the ICT Smart City, the company is building C-ITS(Cooperative-Intelligent Transport Systems) for solving urban traffic problems. In order to realize autonomous driving service with C-ITS, the role of advanced road infrastructure is important. In addition to the study of mid- to long-term C-ITS and autonomous driving services, it is necessary to present more realistic solutions for road traffic safety in the short term. Therefore, in this paper, we propose a highway accident detection alarm system that can detect and analyze traffic flow and risk information, which are essential information of C-ITS, based on IoT guard rail and provide immediate alarm and remote control. Intelligent IoT guard rail is expected to be used as an intelligent advanced road infrastructure that provides data at actual road sites that are required by C-ITS and self-driving services in the long term.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.187-192
• /
• 2021

This study proposes a system for take-off in a forest or similar complex environment using an object detector. In the simulator, a raspberry pi is mounted on a quadcopter with a length of 550mm between motors on a diagonal line, and the experiment is conducted based on edge computing. As for the images to be used for learning, about 150 images of 640⁎480 size were obtained by selecting three points inside Kunsan University, and then converting them to black and white, and pre-processing the binarization by placing a boundary value of 127. After that, we trained the SSD_Inception model. In the simulation, as a result of the experiment of taking off the drone through the model trained with the verification image as an input, a trajectory similar to the takeoff was drawn using the label.