A Recognition System for Cross-section-analysis of Tunnels Using a Visual Cortex Model
| Project/Area Number |
17K18897
|
|---|
| Research Category |
Grant-in-Aid for Challenging Research (Exploratory)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Civil engineering and related fields
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-06-30 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥5,850,000 (Direct Cost: ¥4,500,000、Indirect Cost: ¥1,350,000)
Fiscal Year 2019: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2018: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2017: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 自己組織化マップ / 大脳視覚野 / トンネル / 切羽 / コンクリート / 打音検査 / パターン認識 |
|---|
| Outline of Final Research Achievements |
The Self-organizing map (SOM), which mimics the visual cortex in brain, was applied to classification tasks in cross-section analysis of tunnels and impact-echo testing of concrete structures, and it was demonstrated that the SOM was effective for the information technologies for the civil construction (I-Construction) in the coming future.
On the other hand, SOM needs very long computation time in its learning. In order to overcome this difficulty, we proposed a novel specialized hardware for the SOM, and showed it was able to accelerate the computation speed several to 40 times faster than the software computation, using a hardware prototype.
|
| Academic Significance and Societal Importance of the Research Achievements |
トンネルやダム，橋梁などの社会インフラは，その老朽化が急速に進んでいる．一方で，熟練の検査者は減少しており，今後は人工知能をこれらの検査・解析に応用することが望まれている．本研究では，大脳の視覚野をモデルとした人工知能である自己組織化マップをこれら社会インフラの検査・解析に適用した．自己組織化マップは，“学習結果の可視化”というディープラーニング等の人工知能技術にはない特徴を有する．本研究では，この特徴が検査者をアシストする新たな検査システム開発に有効であることを示した．さらに，本システムを専用ハードウェア化することで，大規模な自己組織化マップを高速に実行できることを試作・評価により示した．
|
Report
(4 results)
Research Products
(20 results)
