Title:
LEARNED MODEL GENERATION METHOD, LEARNED MODEL GENERATION DEVICE, PRODUCT DISCRIMINATION METHOD, PRODUCT DISCRIMINATION DEVICE, PRODUCT DISCRIMINATION SYSTEM, AND WEIGHING DEVICE
Document Type and Number:
Japanese Patent JP2020161077
Kind Code:
A
Abstract:
To provide a learned model generation method, learned model generation device, product discrimination method, product discrimination device, and weighing device that can improve the discrimination accuracy of products.SOLUTION: A learned model generation method includes: an acquisition step S21 of acquiring a product image and product information related to a type of product; a division step S22 of dividing the product image into a plurality of areas to acquire a plurality of division images; an extraction step S23 of extracting a predetermined division image based on a predetermined condition regarding an amount of a shown product among the plurality of division images; and a generation step S24 of performing machine learning based on the plurality of division images extracted in the extraction step and generating a discrimination model.SELECTED DRAWING: Figure 12
Inventors:
TSUTSUMI HIRONORI
FUCHUYA KOSUKE
FUCHUYA KOSUKE
Application Number:
JP2019062949A
Publication Date:
October 01, 2020
Filing Date:
March 28, 2019
Export Citation:
Assignee:
ISHIDA SEISAKUSHO
International Classes:
G06T7/00; G06V10/774; G07G1/00
Domestic Patent References:
JP2017117019A | 2017-06-29 | |||
JP2017182653A | 2017-10-05 | |||
JP2018169752A | 2018-11-01 | |||
JP2013117861A | 2013-06-13 | |||
JP2002329188A | 2002-11-15 | |||
JPH0942939A | 1997-02-14 | |||
JP2017016268A | 2017-01-19 |
Attorney, Agent or Firm:
Yoshiki Hasegawa
Yoshiki Kuroki
Kenichi Shibayama
Yoshiki Kuroki
Kenichi Shibayama