Study to Design Discrete-Time Oscillator Network for Computer-Aided Image Diagnosis
Project/Area Number |
25330288
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Soft computing
|
Research Institution | Kagawa University (2015) The University of Tokushima (2013-2014) |
Principal Investigator |
|
Project Period (FY) |
2013-04-01 – 2016-03-31
|
Project Status |
Completed (Fiscal Year 2015)
|
Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2015: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2014: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2013: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
Keywords | 離散時間振動子ネットワーク / 動的画像領域分割 / 振動応答 / カオス / フラグメント / 離散時間振動子結合系 / カオス振動応答 / 最大リアプノフ指数 / 分岐制御 |
Outline of Final Research Achievements |
The purpose of this study is developing an image segmentation system to be applicable to computer-aided diagnosis system for medical images. We studied the functional design of a discrete-time oscillator network for dynamic image segmentation and obtained the following results. Our detailed analysis of the proposed system gave the parameter values so that the proposed system can generate fragments that are candidates of segmented image regions. We also proposed a method to find such values of system parameters on the basis of an optimization problem on the maximum local Lyapunov exponent. We demonstrated that the proposed system with the parameter values works for three-dimensional medical images well through our experiments. Moreover, from a practical application perspective, we simplified the structure of discrete-time oscillators and implemented the simplified oscillator network in a graphics processing unit.
|
Report
(4 results)
Research Products
(13 results)