Entity

Time filter

Source Type


Yamakawa K.,Tohoku University | Yamakawa K.,Akita | Muraoka H.,Tohoku University | Fudano K.,Tohoku University | And 6 more authors.
IEEE Transactions on Magnetics | Year: 2010

A planar single-pole head generating a large head field was investigated to improve head field distribution with a steep field gradient. Good writing performance is generally difficult for a narrow track writer. This work presents a novel design of the main pole shape, the three-step-slope, to achieve a high gradient and strong write head field for terabit recording. Furthermore, the medium design was investigated to increase head-to-medium interaction for achieving a high field gradient and narrow field distribution. A soft guard band in the recording layer and a cross-track groove on the SUL surface were effective for improving the field gradient. © 2010 IEEE. Source


Iyama T.,Iwate University | Mizuno M.,Iwate University | McKay K.N.,University of Waterloo | Nishikawa N.,Iwate University | Kawakita T.,NEC Software Tohoku Ltd.
International Journal of Production Research | Year: 2011

Electric relay manufacture and assembly is an example of high-precision machining followed by an assembly process. During machining, parts exhibit dimensional variance and manufacturers have several techniques and strategies for how to maximise production and improve efficiency when variance is present. One approach is to measure the variance, select parts appropriately for the best match, perform the minimum amount of adjustment and rework, and then perform the assembly. This approach is difficult in practice because measurement errors also occur and confound the knowledge about each part's dimensions. In this paper, we consider a new matching approach to increase the production rate. We propose a part-combination selection method in which a pair of assembly parts in buffers is optimally selected using a target of an estimated assembly error, and an adjustment machine is then optimally selected using a range of estimated assembly errors. Furthermore, we consider an analytical approach to estimate the optimal control parameters for the proposed method that yield the maximum production rate. The analysis results show that the analytical approach can estimate the nearly optimal control parameters, including the nearly maximum production rate in any buffer capacity. The results also show that by installing a small buffer capacity, the production rate can be increased. © 2011 Taylor & Francis. Source


A memory device sharing system includes M (M represents an integer of 2 or greater) access control apparatus for sharing N (N represents an integer of 2 or greater) memory devices which store data, and a managing apparatus for managing access to the memory devices via the access control apparatus. The managing apparatus checks data stored in the N memory devices, generates data position information representative of the storage positions of data stored in any one of the N memory devices, and sends the data position information to the M access control apparatus. Each of the M access control apparatuses receives the data position information sent from the manager, and accesses the storage position indicated by the data position information if each of the M access control apparatuses receives an access request to access the data from an access request source.


Patent
NEC Software Tohoku Ltd. and NEC Corp | Date: 2012-01-18

A storage system of the present invention includes: a storage device for storing division data configuring a file, and also storing address data referring to the division data or other address data; and a data storage controlling unit for, when the division data or the other address data to be newly stored has a same data content as data already stored in the storage device, controlling to refer to the data already stored in the storage device as the division data or the other address data to be newly stored, by using the address data. The data storage controlling unit includes an address data acquiring unit for acquiring address data of a copy target file based on file specification information, and an address data set-up unit for setting up the acquired address data in the storage device so that the copy target file is stored in a copy destination directory.


Tajima Y.,Tohoku University | Fudano K.,NEC Software Tohoku Ltd. | Ito K.,NEC Software Tohoku Ltd. | Aoki T.,Tohoku University
IEICE Transactions on Information and Systems | Year: 2013

This paper presents a fast and accurate volume correspondence matching method using 3D Phase-Only Correlation (POC). The proposed method employs (i) a coarse-to-fine strategy using multi-scale volume pyramids for correspondence search and (ii) high-accuracy POC-based local block matching for finding dense volume correspondence with subvoxel displacement accuracy. This paper also proposes its GPU implementation to achieve fast and practical computation of volume registration. Experimental evaluation shows that the proposed approach exhibits higher accuracy and lower computational cost compared with conventional method. We also demonstrate that the GPU implementation of the proposed method can align two volume data in several seconds, which is suitable for practical use in the image-guided radiation therapy. Copyright © 2013 The Institute of Electronics, Information and Communication Engineers. Source

Discover hidden collaborations