Eiji Nunohiro

Implementation and Evaluation of OpenMP for Hitachi SR8000

This paper describes the implementation and evaluation of the OpenMP compiler designed for the Hitachi SR8000 Super Technical Server. The compiler performs parallelization for the shared memory multiprocessors within a node of SR8000 using the synchronization mechanism of the hardware to perform high-speed parallel execution. To create an optimized code, the compiler can perform optimizations across inside and outside of a PARALLEL region or can produce a code optimized for a fixed number of processors according to the compile option. For users convenience, it supports combination of OpenMP and automatic parallelization or Hitachi proprietary directive and also supports reporting diagnostic messages which help users parallelization. We evaluate our compiler by parallelizing NPB2.3-serial benchmark with OpenMP. The result shows 5.3 to 8.0 times speedup on 8 processors.

Artificial neural network ensemble-based land-cover classifiers using MODIS data

Terra and Aqua, two satellites launched by the NASA-centered International Earth Observing System project, house MODIS (moderate resolution imaging spectroradiometer) sensors. Moderate-resolution remote sensing allows the quantifying of land-surface type and extent, which can be used to monitor changes in land cover and land use for extended periods of time. In this article, we propose land-surface classification by applying an ensemble technique based on fault masking among individual classifiers in N-version programming. An N-version programming ensemble of artificial neural networks is created, in which the majority vote result is used to predict land-surface cover from MODIS data. It is shown by experiment that an N-version programming ensemble of neural networks greatly improves the classification error rate of land-cover type.

Development and evaluation of satellite image data analysis infrastructure

Tokyo University of Information Sciences (TUIS) receives moderate resolution imaging spectroradiometer (MODIS) data, and provides the processed data to universities and research institutes as part of the academic frontier project. One of the major fields of research using MODIS data is the analysis of changes in the environment. We are currently developing applications to analyze environmental changes. These applications run on our satellite image data analysis system, which is implemented in a parallel distributed system and a database server. When using satellite data, one common problem is the interference of clouds. In order to remove this interference, the standard solution is to create composite data of the same regions during a selected time span, and to patch together data which are not covered by clouds to create a clear image. We introduced a piece-processing algorithm which separates one set of satellite image data into many small pieces of image data, making it quicker and easier to analyze and process the time-series satellite data. In this research, we implemented the pieceprocessing and composite-processing algorithms in order to increase the speed of analysis within the satellite image database. We tested the proposed processing and verified its effectiveness for target applications.

Artificial neural networks paddy-field classifier using spatiotemporal remote sensing data

Monitoring changes in a paddy-field area is important since rice is a staple food and paddy agriculture is a major cropping system in Asia. For monitoring changes in land surface, various applications using different satellites have been researched in the field of remote sensing. However, monitoring a paddy-field area with remote sensing is difficult owing to the temporal changes in the land surface, and the differences in the spatiotemporal characteristics in countries and regions. In this article, we used an artificial neural network to classify paddy-field areas using moderate resolution sensor data that includes spatiotemporal information. Our aim is to automatically generate a paddy-field classifier in order to create localized classifiers for each country and region.

Ensemble of artificial neural network based land cover classifiers using satellite data

Terra and Aqua, 2 satellites launched by the NASA-centered international Earth Observing System project, house MODIS (Moderate Resolution Imaging Spectroradiometer) sensors. Moderate resolution remote sensing allows the quantifying of land surface type and extent, which can be used to monitor changes in land cover and land use for extended periods of time. In this paper, we propose applying an ensemble technique, based on fault masking among individual classifier for N-version programming. We create an N-version programming ensemble of artificial neural networks and use the majority voting result to predict land surface cover from MODIS data. We show that an N-version programming ensemble of neural networks greatly improves the classification error rate of land cover type.

Applying Brightness Information in Satellite Image Data Search using Distributed Genetic Algorithm

Tokyo University of Information Sciences maintains and distributes MODIS (Moderate Resolution Imaging Spectroradiometer) satellite data as part of the research output for Frontier project. An intelligent image search system is being developed as part of the project, in order to retrieve requested images such as matching images patterns or forest and field fires extraction. The intelligent image search system applies genetic algorithm (GA) in the search algorithm. When searching for a target image area within the MODIS image database, it is possible that the search algorithm cannot match the optimal location when the brightness of the search image data and MODIS data image are very different. In order to solve this problem, we extended the search algorithm by implementing brightness information in the GA chromosome, so that brightness is adjusted within the GA image search. Further, we implemented the image search as distributed genetic algorithm search over a PC cluster network, in order to increase the search speed within the satellite image database. We tested the proposed system and verified the effectiveness of distributed genetic algorithm for the distributed MODIS satellite database search process.

A Personal Branding for University Students by Practical Use of Social Media

Social media, such as Twitter and Facebook, has been popularized to our society. In the social media times, people become more active between online and offline. Today, personal branding is very important to harness an individual strong point. We are studying about the personal branding for university students by practical use of social media. Concretely, we gave a lecture about social media to freshmen in our university. We conduct surveys on the actual status of the usage of social media before/after the lecture. As the result, we confirmed that they had improved literacy and understanding of social media. From this survey, we are planning the practical approach for the personal branding of the students using social media.

Land cover classification from MODIS satellite data using probabilistically optimal ensemble of artificial neural networks

Terra and Aqua, 2 satellites launched by the NASA-centered international Earth Observing System project, house MODIS (Moderate Resolution Imaging Spectroradiometer) sensors. Moderate resolution remote sensing allows the quantifying of land surface type and extent, which can be used to monitor changes in land cover and land use for extended periods of time. In this paper, we propose applying a probabilistically optimal ensemble technique, based on fault masking among individual classifier for N-version programming. We create an optimal ensemble of artificial neural networks and use the majority voting result to predict land surface cover from MODIS data. We show that an optimal ensemble of neural networks greatly improves the classification error rate of land cover type.

Techniques for compiling and implementing all NAS parallel benchmarks in HPF

The NAS parallel benchmarks (NPB) are a well‐known benchmark set for high‐performance machines. Much effort has been made to implement them in High‐Performance Fortran (HPF). In previous attempts, however, the HPF versions did not include the complete set of benchmarks, and the performance was not always good. In this study, we implement all eight benchmarks of the NPB in HPF, and parallelize them using an HPF compiler that we have developed. This report describes the implementation techniques and compiler features necessary to achieve good performance. We evaluate the HPF version on the Hitachi SR2201, a distributed‐memory parallel machine. With 16 processors, the execution time of the HPF version is within a factor of 1.5 of the hand‐parallelized version of the NPB 2.3 beta. Copyright

Development of game-based learning features in programming learning support system

In this research, a programming learning support system incorporating game-based learning is proposed. The aim of the game-based learning approach is to stimulate and sustain the motivation of the learner during programming training. In the developed system, a puzzle-solving interface to programming training and a competitive scoring system was implemented to incorporate game-based learning. The proposed system was applied to an actual college programming course to verify the effectiveness of the proposed system. Finally, future works based on the results are discussed.