Advanced of internal defect detection technology for metal components

2020 Fiscal Year Final Research Report

• Project/Area Number: 19K14872
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 18030:Design engineering-related
• Research Institution: Tokyo Metropolitan Industrial Technology Research Institute
• Principal Investigator: TOMIYAMA Shinichi 地方独立行政法人東京都立産業技術研究センター, 事業化支援本部地域技術支援部城南支所, 副主任研究員 (40614524)
• Project Period (FY): 2019-04-01 – 2021-03-31
• Keywords: X線CT / 鋳巣検出 / 画像解析 / ボイド検出 / 異物検出

Outline of Final Research Achievements

Precision instruments makers want higher accuracy detection of internal defects. The detection of interior defects are detected by image processing of CT images. The conventional method has false detection and undetection of defects, because noise in the CT image or a discrimination method based on a manually set threshold value. In this research, we develop an internal defect detection method that reduces false detection and undetection of defects internal defects. We proposed a new defect detection method that after narrowing down only the defects that do not have similar features, the degree of separation of the defects is calculated, and only the defects that are not isolated are narrowed down. The proposed method enabled automatic detection of internal defects with less false detection and undetection of defects than the conventional method.

Free Research Field

画像処理

Academic Significance and Societal Importance of the Research Achievements

提案法により、内部欠陥の誤検出と未検出を低減した内部欠陥自動検出が可能になる。また、内部の異物検出も設定を反転することで高精度な検出が可能になる。高精度な内部欠陥や内部の異物検出技術は、肉厚検査や現物融合型エンジニアリングへの応用が見込める。これは製品設計、信頼性設計、保全性設計、リバースエンジニアリング、安全設計の高付加価値化が見込めるため、非常に有意義なことである。