Back to results
US 12,560,508 B2Granted
Information processing apparatus and information processing method
AssigneeCanon Kabushiki Kaisha
InventorAtsushi Nogami
Life of this patent
Estimating the term…
Abstract
Based on a defective state of a structure, a degree of soundness that indicates how much the structure is sound is determined. Information about a defective state of the structure that provides grounds for this determination is outputted.
Claims12
An information processing apparatus comprising:
- one or more memories storing instructions and storing a trained machine-learning model;
- and
- one or more processors executing the instructions to:
- acquire an image that includes a structure and store the image in the one or more memories;
- receive at least one user input via an input device;
- set, based on the at least one user input, at least one of a first condition or a second condition, the first condition being a condition of using a first portion of a damage that exists across a border of a particular area in the image for a determination and not using a second portion of the damage for the determination, the first portion being included in the particular area, the second portion being not included in the particular area, and the second condition being a condition of using both the first portion and the second portion of the damage for the determination;
- in response to receiving the at least one user input, in a case where the at least one user input sets the first condition,
- generate a first modified image based on the image, wherein generating the first modified image includes highlighting the first portion in the image without highlighting the second portion in the image,
- determine, using the trained machine-learning model, how much the structure is damaged using the first portion and without using the second portion, and
- generate a first display, wherein the first display includes the first modified image, and wherein the first display includes an indicator of how much the structure is damaged as determined using the first portion and without using the second portion;
- and
- in response to receiving the at least one user input, in a case where the at least one user input sets the second condition,
- generate a second modified image based on the image, wherein generating the second modified image includes highlighting the first portion and highlighting the second portion in the image,
- determine, using the trained machine-learning model, how much the structure is damaged using the first portion and using the second portion, and
- generate a second display, wherein the second display includes the second modified image, and wherein the second display includes an indicator of how much the structure is damaged as determined using the first portion and using the second portion.
12 more elements
An information processing method comprising:
- acquiring an image that includes a structure;
- receiving at least one user input via an input device;
- setting, based on the at least one user input, at least one of a first condition or a second condition, the first condition being a condition of using a first portion of a damage that exists across a border of a particular area in the image for a determination and not using a second portion of the damage for the determination, the first portion being included in the particular area, the second portion being not included in the particular area, and the second condition being a condition of using both the first portion and the second portion of the damage for the determination;
- in response to receiving the at least one user input, in a case where the at least one user input sets the first condition,
- generating a first modified image based on the image, wherein generating the first modified image includes highlighting the first portion in the image without highlighting the second portion in the image,
- determining, using a trained machine-learning model, how much the structure is damaged using the first portion and without using the second portion, and
- generating a first display, wherein the first display includes the first modified image, and wherein the first display includes an indicator of how much the structure is damaged as determined using the first portion and without using the second portion;
- and
- in response to receiving the at least one user input, in a case where the at least one user input sets the second condition,
- generating a second modified image based on the image, wherein generating the second modified image includes highlighting the first portion and highlighting the second portion in the image,
- determining, using the trained machine-learning model, how much the structure is damaged using the first portion and using the second portion, and
- generating a second display, wherein the second display includes the second modified image, and wherein the second display includes an indicator of how much the structure is damaged as determined using the first portion and using the second portion.
9 more elements
A non-transitory storage medium that stores a program for causing a computer to execute an information processing method, the information processing method comprising:
- acquiring an image that includes a structure;
- receiving at least one user input via an input device;
- setting, based on the at least one user input, at least one of a first condition or a second condition, the first condition being a condition of using a first portion of a damage that exists across a border of a particular area in the image for a determination and not using a second portion of the damage for the determination, the first portion being included in the particular area, the second portion being not included in the particular area, and the second condition being a condition of using both the first portion and the second portion of the damage for the determination;
- in response to receiving the at least one user input, in a case where the at least one user input sets the first condition,
- generating a first modified image based on the image, wherein generating the first modified image includes highlighting the first portion in the image without highlighting the second portion in the image,
- determining, using a trained machine-learning model, how much the structure is damaged using the first portion and without using the second portion, and
- generating a first display, wherein the first display includes the first modified image, and wherein the first display includes an indicator of how much the structure is damaged as determined using the first portion and without using the second portion;
- and
- in response to receiving the at least one user input, in a case where the at least one user input sets the second condition,
- generating a second modified image based on the image, wherein generating the second modified image includes highlighting the first portion and highlighting the second portion in the image,
- determining, using the trained machine-learning model, how much the structure is damaged using the first portion and using the second portion, and
- generating a second display, wherein the second display includes the second modified image, and wherein the second display includes an indicator of how much the structure is damaged as determined using the first portion and using the second portion.
9 more elements
Dependencies are inferred from the claim text.