Pet status assessment system, pet camera, server, pet status assessment method, and program

Patent metadata

Field	Value
Publication number	US-2023320328-A1
Application number	US-202118022074-A
Country	US
Kind code	A1
Filing date	Aug 20, 2021
Priority date	Sep 1, 2020
Publication date	Oct 12, 2023
Grant date	—

Abstract

A pet status assessment system includes an area detector, an information generator, and an assessor. The area detector detects, in image data, a specific area representing at least a part of appearance of a pet as a subject. The information generator generates pet information. The pet information includes posture information which is based on a learned model and the image data. The learned model has been generated by learning the posture of the pet to recognize, on an image, the posture of the pet. The assessor assesses, based on the pet information, a pet status concerning at least one of an emotion of the pet or an action of the pet.

First claim

1 . A pet status assessment system comprising:
an area detector configured to detect, in image data, a specific area representing at least a part of appearance of a pet as a subject; an information generator configured to generate pet information including posture information about at least a posture of the pet, the posture information being based on a learned model and the image data, the learned model having been generated by learning the posture of the pet to recognize, on an image, the posture of the pet; and an assessor configured to assess, based on the pet information, a pet status concerning at least one of an emotion of the pet presented in the specific area or an action of the pet presented in the specific area.

2 . The pet status assessment system of claim 1 , wherein
the assessor is configured to assess the pet status based on the pet information and condition information about at least one of a particular action of the pet or a particular emotion of the pet.

3 . The pet status assessment system of claim 1 , wherein
the area detector is configured to detect the specific area based on a learned model, the learned model having been generated by learning an appearance factor of a predetermined type of pet to recognize, on the image, the predetermined type of pet.

4 . The pet status assessment system of claim 1 , wherein
the area detector is configured to detect, based on a learned model, a head area representing a head region of the subject, the learned model having been generated by learning an appearance factor of the head region of a predetermined type of pet to recognize, on the image, the head region of the predetermined type of pet.

5 . The pet status assessment system of claim 4 , wherein
the information generator includes a facing direction determiner configured to determine, based on the image data in which the specific area has been detected, a direction that the pet is facing in the image data, and the pet information further includes a result of determination made by the facing direction determiner.

6 . The pet status assessment system of claim 5 , wherein
the facing direction determiner is configured to determine, based on at least a relative location of the head area with respect to the specific area, the direction that the pet is facing.

7 . The pet status assessment system of claim 5 , wherein
the assessor is configured to assess the pet status based on the pet information and condition information about at least one of a particular action of the pet or a particular emotion of the pet, the condition information includes facing direction information in which a plurality of directions that the pet is facing and a plurality of pet statuses are associated with each other, and the assessor is configured to assess the pet status based on a result of determination made by the facing direction determiner and the facing direction information.

8 . The pet status assessment system of claim 5 , further comprising an output interface configured to output results of assessment made by the assessor, wherein
the output interface is configured to, when the results of assessment made by the assessor with respect to multiple frames of the image data indicate that the pet is facing an identical direction successively a predetermined number of times, restrict outputting the results of assessment made by the assessor.

9 . The pet status assessment system of claim 1 , further comprising an object detector configured to detect an object area representing, in the image data, a particular object other than the pet, wherein
the information generator includes a distance determiner configured to determine a relative distance of the pet with respect to the object area, the pet information further includes a result of determination made by the distance determiner, the assessor is configured to assess the pet status based on the pet information and condition information about at least one of a particular action of the pet or a particular emotion of the pet, the condition information includes information in which multiple types of particular objects and a plurality of threshold values about distances between the pet and the multiple types of particular objects are associated with each other, and the assessor is configured to assess the pet status by comparing a result of determination made by the distance determiner with the plurality of threshold values.

10 . The pet status assessment system of claim 9 , wherein
the object detector is configured to detect the object area based on a learned model, the learned model having been generated by learning an appearance factor of a predetermined type of particular object to recognize, on the image, the predetermined type of particular object.

11 . The pet status assessment system of claim 9 , wherein
the assessor is configured to, when the particular object presented in the object area detected by the object detector is a bowl and the relative distance determined by the distance determiner is equal to or less than a predetermined threshold value, assess the pet status concerning the action of the pet to be eating.

12 . The pet status assessment system of claim 1 , further comprising an output interface configured to output a result of assessment made by the assessor while associating the result of assessment with the image data in which the specific area that forms a basis of the result of assessment has been detected.

13 . The pet status assessment system of claim 1 , wherein
the area detector is configured to detect, based on a learned model, the specific area of the pet assuming a particular posture in the image data, the learned model having been generated by learning the posture of the pet to recognize, on the image, the posture of the pet.

14 . A pet camera comprising:
the pet status assessment system of claim 1 ; and an image capture device configured to capture the image data.

15 . A server configured to communicate with a pet camera, the pet camera being equipped with the information generator and the assessor of the pet status assessment system of claim 1 ,
the server being equipped with the area detector.

16 . A server configured to communicate with a pet camera, the pet camera being equipped with the area detector of the pet status assessment system of claim 1 ,
the server being equipped with the information generator and the assessor.

17 . A pet status assessment method comprising:
a pet detection step including detecting, in image data, a specific area representing at least a part of appearance of a pet as a subject; an information generation step including generating pet information including posture information about at least a posture of the pet, the posture information being based on a learned model and the image data, the learned model having been generated by learning the posture of the pet to recognize, on an image, the posture of the pet; and an assessment step including assessing, based on the pet information, a pet status concerning at least one of an emotion of the pet presented in the specific area or an action of the pet presented in the specific area.

18 . A non-transitory storage medium that stores a program designed to cause one or more processors to perform the pet status assessment method of claim 17 .

Assignees

• Panasonic Ip Man Co Ltd

Inventors

• Nakamura Yuka
• Omayu Hiroaki
• Uesaka Yasushi
• Sato Masafumi

Classifications

• A01K29/005 Primary

  Monitoring or measuring activity · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06V10/774

  Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V20/50

  Context or environment of the image · CPC title

• G06V40/10

  Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title