Title:
欠点分類システム
Document Type and Number:
Japanese Patent JP7390085
Kind Code:
B2
Abstract:
[Problem] To provide a defect classification system that is easy to prepare and maintain and that makes it possible to recognize the origin of a defect. [Solution] The present invention is a defect classification system 10 comprising: an imaging means 11 that causes an imaging device G to capture an image of paper P2 that has passed through a dry part D and acquires the captured image data; a detection means 12 for detecting a defect in the paper P2 within the image data; an extraction means 13 for extracting a feature quantity pertaining to the defect; a calculation means 14 that, on the basis of the feature quantity pertaining to the defect, calculates the degree of confidence in a defect origin item for a classification model 20 in which a reference feature quantity is set in advance; and a display means 15 for displaying the degree of confidence. The classification model 20 is trained on the reference feature quantity through machine learning from the relationship between the feature quantities of defects accumulated in advance and a plurality of defect origin items.
Inventors:
Hiroshi Sekiya
Hitomaru Sakata
Hitomaru Sakata
Application Number:
JP2023515256A
Publication Date:
December 01, 2023
Filing Date:
September 06, 2022
Export Citation:
Assignee:
Mentec Co., Ltd.
International Classes:
G01N21/892; D21F7/00; G06T7/00
Domestic Patent References:
JP10302049A | ||||
JP2021025161A | ||||
JP5101164A | ||||
JP2019159499A | ||||
JP2019212073A | ||||
JP2018524804A |
Foreign References:
US20190010005 |
Attorney, Agent or Firm:
Shinji Shirasaki
Abe Satoshi
Toshiharu Katsuki
Shingo Okazaki
Abe Satoshi
Toshiharu Katsuki
Shingo Okazaki
Previous Patent: Air purification equipment and air purification systems
Next Patent: umbrella with cam assembly
Next Patent: umbrella with cam assembly