Title:
単板判定システムおよび単板の判定方法
Document Type and Number:
Japanese Patent JP7318907
Kind Code:
B2
Abstract:
To provide a veneer determination system and a venner determination method that can more accurately extract an area where a defect of a veneer appears, improve classification accuracy of a defect type, and perform more appropriate grade determination.SOLUTION: A veneer determination system photographs a veneer B with a camera 12, and obtains a line image 51. It generates a color image 52 from the line image 51 and calculates a local feature quantity of the color image 52. It associates feature points P corresponding to the local feature quantity corresponding to the strength of the local feature quantity and sets a region including all the associated feature points P as a defect region 55. It classifies a type of defect by using AI regarding the defect region 55. It determines the grade of the veneer B using AI for each defect classification.SELECTED DRAWING: Figure 1
Inventors:
Tomohito Kamiya
Ryosuke Kamiya
Masahiro Funase
Kazuhisa Matsuda
Yuichi Hamasuna
Ryosuke Kamiya
Masahiro Funase
Kazuhisa Matsuda
Yuichi Hamasuna
Application Number:
JP2019042135A
Publication Date:
August 01, 2023
Filing Date:
March 08, 2019
Export Citation:
Assignee:
Hashimoto Electric Industry Co., Ltd.
International Classes:
G06T7/00; G01N21/88; G06T1/00
Domestic Patent References:
| JP2014215233A | ||||
| JP2016110290A | ||||
| JP2009074967A | ||||
| JP2011095109A |
Other References:
LI,Meng-Xin,et al.,Automatic visual inspection and classification based on rough sets and neural network,Proceedings of the 2003 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.03EX693),米国,IEEE,2003年11月05日,pp.3095-3099,https://ieeexplore.ieee.org/document/1260110
Attorney, Agent or Firm:
Tasuku Mizuno
Etsu Tamai
Etsu Tamai