• Proceedings of the Korea Contents Association Conference
• /
• 2012.05a
• /
• pp.193-194
• /
• 2012

Most of the Historic Preservation Areas are very weak to fire disasters. The aim of this study is to build a proposal-scheme for evacuation routes planning at a large-scale disaster in a historic preservation area. The study area is a preservation area called "Hizen-Hama-Shuku" which has many straw-roofed wooden houses in Kashima city, Japan. To achieve this aim, outline of the scheme is discussed. The proposal-scheme of evacuation routes consists of objectives to propose, target time to evacuate for the refuge place, evaluation method, and procedures. The existing condition and the measures are evaluated using the software "SimTread".

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.259-260
• /
• 2016

Our research group, which was organized by the South Korean and Japanese researchers, have carried out research about natural disaster in our regional heritage villages derived from the concept of ICT(information communication technology)-based DPD(disaster prevention design). In this research, we performed simulation analysis on the fire safety diagnosis in Asan-Oeam folk and JeonjuHanok village for developing our research of ICT-based PBD. In order to do this, we used fire simulation program which was developed by BRI(Building Research Institute) of Japan. Based on the results of fire simulation of Asan-Oeam folk and JeonjuHanokvillage, we can demonstrated that the fire which broken out inregional heritage folk village will be easily expanded to adjacent houses, because the house which are built with wood structure.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.261-262
• /
• 2016

In this article, we present some derived subjects from the concept of ICT-based DPD concept for the safety of folk villages in both Korea and Japan. First, our deduced topic would rather be a monitoring system design of structures in folk villages. We, therefore, offer an integrated model of maintenance and management monitoring scheme. As another research subject, we submit safety sign or sign system installed in traditional towns and their standardization. We have draw up a plan to make signs upgrade applied to folk villages in Korea and Japan. According to our investigations, we should suggest and focus on flood in the area of traditional town in Korea. We present a water-level expectation model using deep learning simulation. We have applied this method to the area of 'Andong Hahoe' village which had been registered on World Cultural Heritage of UNESCO. The final goal of our research is to propose and realize an integrated disaster prevention and/or safety system based on big data concepts for both Korea and Japan.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.237-238
• /
• 2016

This paper describes Web-based information sharing mechanism in our hazard map creation support system. This system aims at collecting the unique information received from residents and raising resident's consciousness to disasters by recording hazardous locations where residents feel danger in case of disasters. We have implemented a mechanism to share and integrate data of each terminal through a Web server. We expect that this information sharing effects scalability and usefulness of our system by utilizing collected local hazard information of each district.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.243-244
• /
• 2016

From the viewpoint of ICT-based disaster prevention design, evacuation route planning is one of the hot topics to be solved. Since we have quite a lot of natural disasters these days, its role has become important more and more than before. In order to deal with this tough problem, a genetic algorithm (GA) is adopted, and a revised selection procedure for preserving variety of individuals in each generation is introduced. As a result of some computer simulations applied to Hizen-Hamashuku, one of the model areas in this research project, it is found that the proposed method is effective.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.247-248
• /
• 2016

To investigate activities in the evacuated situation of people, the measurement system is newly constructed, that composed of a wearable sensor device of heart beats rates and mobile devices like an Android smartphone with a bluetooth low energy (BLE) connection. Smartphone not only displays the heart beats variation (HBR) and the current location of evacuation person by Global Positioning System (GPS), but also exports the CSV formatted file that would be used for further analyzing the activity of person in detail. As an example of the application of this system, we show the case of evacuation routes for elderly person in Hizen-Hamashuku Area, Saga Prefecture. Using the proposed measuring system, the activities of evacuates can be clearly shown on the map of Geospatial Information System (GIS).

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.274-277
• /
• 2000

Linear discriminant analysis (LDA) is a basic tool of pattern recognition, and it is used in extensive fields, e.g. face identification. However, LDA is poor at adaptability since it is a batch type algorithm. To overcome this, a new algorithm of online LDA is proposed in the present paper. It is experimentally shown that the new algorithm is about two times faster than the previously proposed algorithm.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.298-301
• /
• 2000

We discuss the iterative methods for linear systems on a single processing node of the HITACHI SR8000. Each processing node of the SR8000 is a shared memory parallel computer which is composed of eight RISC processors with a pseudo-vector facility. We implement highly optimized codes for basic linear operations including a matrix-vector product and apply them to the conjugate gradient (CG) and the conjugate residual (CR) methods for linear systems. Our tuned codes for both method score nearly 50% of the theoretical peak performance, which is the best in the sense that it corresponds to an asymptotic performance of the inner product.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.333-336
• /
• 2000

We discuss the direct methods (Gauss-Jordan and Gaussian eliminations) to solve linear systems on distributed memory parallel computers. It will be shown that the so-called row-cyclic storage gives rise to the best performance among the standard three (row-cyclic, column-cyclic and cyclic-cyclic) data storages. We also show that Gauss-Jordan elimination, rather than Gaussian elimination, is highly efficient for the direct solution of linear systems in parallel processing, though Gauss-Jordan elimination requires a larger number of arithmetic operations than Gaussian elimination. Numerical experiment is performed on HITACHI SR12201 with the standard libraries MPI and BLAS.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1670-1673
• /
• 2002

For pattern recognition on high-dimensional data, such as images, the dimensionality reduction as a preprocessing is effective. By dimensionality reduction, we can (1) reduce storage capacity or amount of calculation, and (2) avoid "the curse of dimensionality" and improve classification performance. Popular tools for dimensionality reduction are Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Independent Component Analysis (ICA) recently. Among them, only LDA takes the class labels into consideration. Nevertheless, it, has been reported that, the classification performance with ICA is better than that with LDA because LDA has restriction on the number of dimensions after reduction. To overcome this dilemma, we propose a new dimensionality reduction technique based on an information theoretic measure for difference of distribution. It takes the class labels into consideration and still it does not, have restriction on number of dimensions after reduction. Improvement of classification performance has been confirmed experimentally.