Estimation method, information processing apparatus, and non-transitory computer-readable storage medium for estimating motion of animal

Patent metadata

Field	Value
Publication number	US-11129552-B2
Application number	US-201816110334-A
Country	US
Kind code	B2
Filing date	Aug 23, 2018
Priority date	Feb 29, 2016
Publication date	Sep 28, 2021
Grant date	Sep 28, 2021

Abstract

An estimation method performed by a computer includes: executing a first identifying process that includes identifying a candidate for a limb having a problem by evaluating left and right balance at the time of movement based on measurement data of a motion sensor attached to an animal moving on four limbs, and executing a second identifying process that includes identifying the limb having a problem among the limbs which are considered as the candidates for a limb having a problem based on a detection result of a nutation movement by detecting the nutation movement from the measurement data.

First claim

What is claimed is:

1. An estimation method performed by a computer, the method comprising:
executing a first identifying process that includes identifying a candidate limb possibly having a problem by evaluating left and right balance at the time of movement based on measurement data of a motion sensor attached in front of a chest region of an animal moving on four limbs, the measurement data indicating a movement of the chest region of the animal which has been measured by the motion sensor; and
executing a second identifying process that includes identifying, based on a detection result obtained by detecting a nutation movement from the measurement data, a limb having the problem from among limbs which are considered as the candidate limb in the first identifying process, the nutation movement being an upward nutation movement or a downward nutation movement; and
outputting a result of the second identifying process, the outputting of the result including the candidate limb identified to have the problem,
wherein the evaluating of the left and right balance includes:
mapping a trajectory using the measurement data in measurement time order onto an X-Y coordinate plane for each of a plurality of walking completions of the animal, the plane having an X axis representing a movement amount in the horizontal direction and a Y axis representing a movement amount in an up-down direction;
determining a maximum point in the trajectory at which a position of a next point in the up-down direction first drops,
determining a minimum point in the trajectory at which a position of a next point in the up-down direction first rises, where the minimum point occurring after the maximum point in the measurement time order is positioned on a right of the maximum point in the plane, the maximum point is deemed to be on a left side and the minimum point is deemed to be on a right side, and where the minimum point occurring after the maximum point in the measurement time order is positioned on a left of the maximum point in the plane, the maximum point is deemed to be on the right side and the minimum point is deemed to be on the left side;
determining in the trajectory a first point, a second point, a third point, and a fourth point, the first point being a maximum point on the left side, the second point being a minimum point on the right side, the third point being a maximum point on the right side, and the fourth point being a minimum point on the left side;
generating a first line segment, a second line segment, an intersection point, and a center line in the plane, the first line segment connecting the first point and the second point to each other, the second line segment connecting the third point and the fourth point to each other, the intersection point being at an intersection between the first line segment and the second line segment, and the center line being a line extending in the Y-axis direction that passes through the intersection point;
generating a first indicator, a second indicator, and a third indicator for the trajectory, the first indicator being a ratio of a left area to a right area of the trajectory with respect to the center line, the second indicator being a ratio of a length of portions of the first line segment and the second line segment that are on the left of the center line to a length of portions of the first line segment and the second line segment that are on the right of the center line, the third indicator being a ratio of a decrease amount of the second point from the intersection point to a decrease amount of the fourth point from the intersection point, with a value less than one for each of the first, second, and third indicators representing an evaluation of the let and right balance to be tilted to the right side, and a value greater than one for each of the first, second, and third indicators representing the evaluation of the left and right balance to be tilted to the left side; and
determining the left and right balance using a majority of the evaluations represented by the first, second, and third indicators, and

wherein the detecting of the nutation movement is configured to:
determine that the upward nutation movement has occurred in a case where the measurement data equal to or greater than a predetermine value is generated in an upward direction; and
determine that the downward nutation movement has occurred in a case where the measurement data equal to or greater than the predetermine value is generated in a downward direction.

2. The estimation method according to claim 1 ,
wherein the first identifying process is configured to
evaluate the left and right balance,
identify a left forelimb and a right hindlimb as the candidate limb when the left and right balance is evaluated to be tilted to the right side, and
identify a right forelimb and a left hindlimb as the candidate limb when the left and right balance is evaluated to be tilted to the left side.

3. The estimation method according to claim 2 ,
wherein the second identifying process is configured to
identify a forelimb identified as the candidate limb as a limb having a problem when the upward nutation movement is detected upon movement of the forelimb and a hindlimb which are considered as the candidate limb, and
identify the hindlimb identified as the candidate limb as a limb having a problem when the downward nutation movement is detected, from the measurement data.

4. The estimation method according to claim 1 , the method further comprising:
executing a determining process that includes determining a gait of the animal based on the measurement data,
wherein the first identifying process is configured to identify the candidate limb from the result of evaluating the left and right balance when the animal moves in the gait of walk or trot.

5. The estimation method according to claim 4 ,
wherein the second identifying process is configured to
identify a hindlimb identified as the candidate limb as a limb having a problem when the left and right balance is reversed in the gait of trot and the gait of walk, and
identify a forelimb identified as the candidate limb as a limb having a problem when the left and right balance is reversed in the gait of trot and the gait of walk.

6. The estimation method according to claim 4 ,
wherein the first identifying process is configured to
determine the presence or absence of lameness based on the symmetry of motion waveforms in any one of the up-down direction, the left-right direction, and the yaw axis direction when the animal moves in the gait of trot,
identify the left forelimb and the right hindlimb as the candidate limb when the yaw axis direction turns rightward at a timing at which the motion waveform becomes asymmetric, and
identify the right forelimb and the left hindlimb as the candidate limb when the left turn at a timing at which the motion waveform becomes asymmetric.

7. The estimation method according to claim 4 ,
wherein the second identifying process is configured to
determine, when the nutation movement is not detected, whether there is a first timing at which the motion waveform becomes asymmetric upon movement of the animal in the gait of walk,
identify a forelimb considered as the candidate limb as a limb having a problem when the first timing is interlocked with the movement of the forelimb, and
identify a hindlimb considered as the candidate limb as a limb having a problem when the first timing is not interlocked with the movement of the forelimb.

8. An information processing apparatus for estimating a motion of an animal, the apparatus comprising:
a memory; and
a processor coupled to the memory and configured to
execute a first identifying process that includ

Assignees

• Fujitsu Ltd

Inventors

• Yamamoto Kiyoko

Classifications

• A01K29/005 Primary

  Monitoring or measuring activity · CPC title

• A61B5/1122

  of movement trajectories · CPC title

• A01K29/00

  Other apparatus for animal husbandry · CPC title

• A61B5/112 Primary

  Gait analysis · CPC title

• A01K67/00

  Rearing or breeding animals, not otherwise provided for; New or modified breeds of animals · CPC title