Apparatus and manual providing apparatus

Patent metadata

Field	Value
Publication number	US-12307677-B2
Application number	US-202217663498-A
Country	US
Kind code	B2
Filing date	May 16, 2022
Priority date	May 18, 2021
Publication date	May 20, 2025
Grant date	May 20, 2025

Abstract

An apparatus includes circuitry; and a memory storing computer-executable instructions that cause the circuitry to execute acquiring training data, in which image data in a manual and text data in the manual are input data, and in which work procedure information is output data, the work procedure information being supplemented by adding, to the text data in the manual, text data that is generated based on the image data; performing machine learning by using the training data; and generating a machine learning model that outputs the work procedure information in response to receiving input of the image data in the manual and the text data in the manual.

First claim

What is claimed is:

1. An apparatus comprising:
an in-manual image data storage unit configured to store image data included in a manual;
an in-manual text data storage unit configured to store text data included in the manual;
circuitry; and
a memory storing computer-executable instructions that cause the circuitry to execute:
acquiring, from a user terminal, information for identifying a target;
identifying image data in the manual and text data in the manual corresponding to the information for identifying the target;
inputting, to a machine learning model, the image data in the manual and the text data in the manual, to cause the machine learning model to output work procedure information; and
outputting the work procedure information to the user terminal, wherein the instructions further cause the circuitry to execute:
training the machine learning model based on training data including the image data that is stored in the in-manual image data storage unit and the text data that is stored in the in-manual text data storage unit as input data, and supplemented text data, as output data, that is supplemented by text data generated based on the image data in the manual;
wherein the work procedure information is generated by supplementing the text data related to the target with first text indicating a position of the target and second text indicating a direction to move the target, said first and second texts being generated based on an image indicating the position of the target included in the image data and an image indicating the direction to move the target, and
wherein the work procedure information is output by voice sound.

Assignees

• Yamaya Yuriko
• Kawamura Shintaro
• Harashima Seigo
• Iguchi Shinya
• Ricoh Co Ltd

Inventors

• Yamaya Yuriko
• Kawamura Shintaro
• Harashima Seigo
• Iguchi Shinya

Classifications

• G06T7/73

  using feature-based methods · CPC title

• G06T7/136

  involving thresholding · CPC title

• G06V30/1801

  Detecting partial patterns, e.g. edges or contours, or configurations, e.g. loops, corners, strokes or intersections (extracting features by contour coding G06V30/182) · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06V30/42

  based on the type of document · CPC title