Back to Search
Start Over
基于机器视觉的板式换热器板片 微裂纹检测方法研究.
- Source :
-
Journal of Mechanical & Electrical Engineering . 2020, Vol. 37 Issue 7, p86-90. 5p. - Publication Year :
- 2020
-
Abstract
- Ａｉｍｉｎｇ ａｔ ｔｈｅ ｐｒｏｂｌｅｍｓ ｏｆ ｈｉｇｈ ｄｅｔｅｃｔｉｏｎ ｅｒｒｏｒ ｒａｔｅ ａｎｄ ｌｏｗ ｒｅａｌ-ｔｉｍｅ ｐｅｎｅｔｒａｔｅ ｄｅｔｅｃｔｉｏｎ ｏｆ ｍｉｃｒｏ-ｃｒａｃｋｓ ｉｎ ｐｌａｔｅ ｈｅａｔ ｅｘｃｈａｎｇｅｒ ｐｌａｔｅｓ, ｔｈｅ ｉｍａｇｅ ａｃｑｕｉｓｉｔｉｏｎ, ｍｉｃｒｏ-ｃｒａｃｋ ｍｏｒｐｈｏｌｏｇｙ, ｒｅｇｉｏｎ ｓｅｇｍｅｎｔａｔｉｏｎ, ｅｄｇｅ ｅｘｔｒａｃｔｉｏｎ ａｎｄ ｆｅａｔｕｒｅ ｑｕａｎｔｉｚａｔｉｏｎ ｗｅｒｅ ｓｔｕｄｉｅｄ. Ａ ｍｉｃｒｏｃｒａｃｋ ｄｅｔｅｃｔｉｏｎ ｍｅｔｈｏｄ ｗａｓ ｂａｓｅｄ ｏｎ ｍａｃｈｉｎｅ ｖｉｓｉｏｎ. Ｔｈｅ ｒｅｇｉｏｎ ｏｆ ｉｎｔｅｒｅｓｔ ｗｈｅｒｅ ｍｉｃｒｏｃｒａｃｋｓ ｍａｙ ｅｘｉｓｔ ｗａｓ ｓｅｇｍｅｎｔｅｄ ｂｙ ｔｈｅ ｔｈｒｅｓｈｏｌｄ ａｌｇｏｒｉｔｈｍ. Ｔｈｅ Ｃａｎｎｙ ｏｐｅｒａｔｏｒ ｗａｓ ｕｓｅｄ ｔｏ ｅｘｔｒａｃｔ ｔｈｅ ｍｉｃｒｏｃｒａｃｋｓ ｉｎ ｔｈｅ ｒｅｇｉｏｎ ｏｆ ｉｎｔｅｒｅｓｔ, ａｎｄ ｔｈｅ ｓｐａｔｉａｌ ｐｏｓｉｔｉｏｎ ｆｅａｔｕｒｅｓ ａｎｄ ｌｏｃａｌ ｇｒａｙ ｆｅａｔｕｒｅｓ ｏｆ ｅａｃｈ ｒｏｕｇｈ ｅｘｔｒａｃｔｉｏｎ ｒｅｓｕｌｔ ｗｅｒｅ ｃａｌｃｕｌａｔｅｄ. Ａｃｃｏｒｄｉｎｇ ｔｏ ｔｈｅ ｄｅｆｉｎｅｄ ｆｅａｔｕｒｅｓ, ｔｈｅ ｃｏｎｓｔｒａｉｎｔｓ ｗｅｒｅ ｕｓｅｄ ｔｏ ｓｃｒｅｅｎ ｔｈｅ ｒｅｓｕｌｔｓ ｏｆ ｅａｃｈ ｒｏｕｇｈ ｅｘｔｒａｃｔｉｏｎ ｔｏ ｏｂｔａｉｎ ｔｒｕｅ ｍｉｃｒｏｃｒａｃｋｓ. Ｔｈｅ ｄｅｔｅｃｔｉｏｎ ｍｅｔｈｏｄ ｗａｓ ｔｅｓｔｅｄ ｂｙ ｔｈｅ ｉｍａｇｅ ｃｏｌｌｅｃｔｅｄ ｂｙ ｔｈｅ ｆａｃｔｏｒｙ. Ｔｈｅ ｒｅｓｕｌｔｓ ｉｎｄｉｃａｔｅ ｔｈａｔ ｔｈｅ ｍｅｔｈｏｄ ｃａｎ ｍｅｅｔ ｔｈｅ ｉｎｄｕｓｔｒｉａｌ ｒｅｑｕｉｒｅｍｅｎｔｓ ｏｆ ｎｏ-ｌｅａｋａｇｅ ｄｅｔｅｃｔｉｏｎ ａｎｄ ｌｏｗ ｆａｌｓｅ ｄｅｔｅｃｔｉｏｎ, ａｎｄ ｈａｓ ｒｅａｌ-ｔｉｍｅ ｐｅｒｆｏｒｍａｎｃｅ. Ｉｔ ｈａｓ ｃｅｒｔａｉｎ ｐｒａｃｔｉｃａｌ ａｐｐｌｉｃａｔｉｏｎ ｖａｌｕｅ ｆｏｒ ｔｈｅ ａｕｔｏｍａｔｉｃ ｄｅｔｅｃｔｉｏｎ ｏｆ ｍｉｃｒｏ-ｃｒａｃｋｅｄ ｏｎ ｔｈｅ ｐｌａｔｅ ｈｅａｔ ｅｘｃｈａｎｇｅｒ ｐｌａｔｅ [ABSTRACT FROM AUTHOR]
Details
- Language :
- Chinese
- ISSN :
- 10014551
- Volume :
- 37
- Issue :
- 7
- Database :
- Academic Search Index
- Journal :
- Journal of Mechanical & Electrical Engineering
- Publication Type :
- Academic Journal
- Accession number :
- 146252421
- Full Text :
- https://doi.org/10.3969/j.issn.1001-4551.2020.07.023