Granted Patent B2

US 12,450,879 · App. 18/249,928 · Granted Oct 21, 2025

Inventors: Taichi Sato (Kyoto, JP); Ryosuke Goto (Osaka, JP); Yuto Kitagawa (Osaka, JP)

Assignee: PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO., LTD.

G06V10/772G06V10/44

Abstract

Claims (63)

1. A data creation system configured to create, based on first image data, second image data for use as learning data to generate a learned model about an object, the data creation system comprising:

a deformer configured to generate, based on the first image data including a pixel region representing the object, the second image data by causing deformation about the object to the first image data;

a remover configured to remove a linear first feature present in the first image data;

a feature acquirer configured to acquire a second feature; and

a feature assigner configured to assign the second feature to the second image data.

2. The data creation system of claim 1 , wherein

the second feature includes the linear first feature present in the first image data, and

the feature acquirer is configured to acquire, as the second feature, the first feature from the first image data.

3. The data creation system of claim 2 , wherein

the remover is configured to remove the first feature from the first image data,

the deformer is configured to generate the second image data based on the first image data from which the first feature has been removed, and

the feature assigner is configured to assign, to the second image data, the first feature that the feature acquirer has acquired from the first image data.

4. The data creation system of claim 2 , wherein

the remover is configured to remove the first feature subjected to the deformation from the second image data, and

the feature assigner is configured to assign, to the second image data from which the first feature has been removed, the first feature that the feature acquirer has acquired from the first image data.

5. The data creation system of claim 1 , wherein

the second feature includes a linear third feature present in additional image data other than the first image data, and

the feature acquirer is configured to acquire, as the second feature, the third feature from the additional image data.

6. The data creation system of claim 1 , wherein

the linear first feature includes a straight feature.

7. The data creation system of claim 6 , wherein

the straight feature concerns a boundary between a plurality of scan lines on the first image data depending on a line sensor camera that has shot the object.

8. The data creation system of claim 7 , wherein

the straight feature is a feature based on a pixel value difference corresponding to a height difference between the plurality of scan lines, the height difference being caused, in a situation, depending on a difference in a distance from the line sensor camera to the object.

9. The data creation system of claim 1 , wherein

the deformation about the object includes rotation of the object.

10. The data creation system of claim 1 , wherein

the deformation about the object includes deformation of a shape of the object.

11. A learning system configured to generate the learned model using a learning data set, the learning data set including the learning data as the image data created by the data creation system of claim 1 .

12. An estimation system configured to estimate a particular condition of the object as an object to be recognized using the learned model generated by the learning system of claim 11 .

13. The data creation system of claim 1 , further including a feature extractor configured to extract the first feature from the first image data, wherein

the feature extractor is included in a first processing device,

the feature assigner is included in a second processing device different from the first processing device, and

the first processing device is configured to transmit information representing the first feature to the second processing device.

14. A processing device functioning as the first processing device of the data creation system of claim 13 .

15. A processing device functioning as the second processing device of the data creation system of claim 13 .

16. A data creation system configured to create image data for use as learning data to generate a learned model about an object, the data creation system comprising:

a feature acquirer configured to acquire a linear feature; and

a feature assigner configured to assign the linear feature to the image data including a pixel region representing the object,

wherein the linear feature concerns a boundary between a plurality of scan lines depending on a line sensor camera to shoot the object.

17. A data creation method for creating, based on first image data, second image data for use as learning data to generate a learned model about an object, the data creation method comprising:

a deforming step including generating, based on the first image data including a pixel region representing the object, the second image data by causing deformation about the object to the first image data;

a removing step including removing a linear first feature present in the first image data;

a feature acquiring step including acquiring a second feature; and

a feature assigning step including assigning the second feature to the second image data.

18. A non-transitory storage medium storing a program designed to cause one or more processors to perform the data creation method of claim 17 .

19. A data creation method for creating image data for use as learning data to generate a learned model about an object, the data creation method comprising:

a feature acquiring step including acquiring a linear feature; and

a feature assigning step including assigning the linear feature to the image data including a pixel region representing the object,

wherein the linear feature concerns a boundary between a plurality of scan lines depending on a line sensor camera to shoot the object.

20. An evaluation system configured to estimate a particular condition of an object, the evaluation system comprising a processing device and an estimation system,

the processing device being configured to: extract a linear first feature present in first image data including a pixel region representing an object; and output information representing the first feature thus extracted,

the estimation system being configured to output, in response to image data created by assigning a second feature to second image data, an estimation result similar to a situation where the first image data is subjected to estimation made about a particular condition of the object, the second image data being generated by removing the first feature from the first image data and causing deformation about the object to the first image data.

21. The evaluation system of claim 20 , wherein

the deformation about the object includes at least one selected from the group consisting of rotation, flipping, translation, scaling up, and scaling down of the object.

22. The evaluation system of claim 20 , wherein

the estimation system is configured to estimate the particular condition of the object using a learned model,

the evaluation system further includes a learning system configured to generate the learned model, and

the learned model being configured to output, in response to image data created by assigning a second feature to second image data, an estimation result similar to a situation where the first image data is subjected to estimation made about the particular condition of the object,

the second image data being generated by removing the first feature from the first image data and causing deformation about the object to the first image data.

23. A learning system functioning as the learning system of the evaluation system of claim 22 .

24. A processing device functioning as the processing device of the evaluation system according to claim 20 .

25. An estimation system functioning as the estimation system of the evaluation system according to claim 20 .

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (34)