A VLSI Brain System Integrating Massively-Parallel Subconscious Processing With Sequential Conscious Processing in the Mind
Project/Area Number |
20246056
|
Research Category |
Grant-in-Aid for Scientific Research (A)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Electronic materials/Electric materials
|
Research Institution | The University of Tokyo |
Principal Investigator |
SHIBATA Tadashi 東京大学, 大学院・工学系研究科, 教授 (00187402)
|
Project Period (FY) |
2008 – 2010
|
Project Status |
Completed (Fiscal Year 2010)
|
Budget Amount *help |
¥43,810,000 (Direct Cost: ¥33,700,000、Indirect Cost: ¥10,110,000)
Fiscal Year 2010: ¥10,920,000 (Direct Cost: ¥8,400,000、Indirect Cost: ¥2,520,000)
Fiscal Year 2009: ¥15,340,000 (Direct Cost: ¥11,800,000、Indirect Cost: ¥3,540,000)
Fiscal Year 2008: ¥17,550,000 (Direct Cost: ¥13,500,000、Indirect Cost: ¥4,050,000)
|
Keywords | VLSI / 知能システム / ブレインコンピューティング / 画像認識動作認識 / 分類器 / 学習アルゴリズム / 学習機能搭載プロセッサ / ブレイン・コンピューティング / K-Means Processor / Support Vector Machine / Nearest Neighbor Classifier / 動作認識 / 風景認識 / 物体抽出 / 情景認識 / K-means学習プロセッサ / 二分木階層アーキテクチャ / 画像特徴ベクトル生成プロセッサ / 実時間学習 / 物体追跡 / CMOSイメージセンサー / VLSI脳システム / object tracking / エッジ検出 / 画像認識 / digital pixel sensor / multiple candidate生成 / VLSIシステム / CMOS VLSI / 並列演算処理 |
Research Abstract |
The research aims to develop a human-like intelligent system that can flexibly understand the situation, make a quick decision and take an immediate action against a happening. Several massively-parallel-architecture processors each dedicated to specific subconscious processing in the mind such as perception of images and their motions have been developed. By integrating such primitive information provided from these chips by a conventional sequential processor, we have successfully built a human-like intelligent system that can track moving objects and recognize what they are and, if they are humans, what actions they are taking.
|
Report
(4 results)
Research Products
(93 results)