# US 10,769,497 B2 · Learning device, imaging device, and learning method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,769,497 B2
- Document ID: US10769497B2
- Status: Granted
- Kind: B2
- Application number: 16/043,057
- Filed: 2018-07-23
- Published: 2019-06-27
- Granted: 2020-09-08
- Earliest claimed: 2017-12-24
- Assignees: Olympus Corporation
- Applicants: Olympus Corporation
- Inventors: Kazuhiro Haneda; Hisashi Yoneyama; Atsushi Kohashi; Zhen Li; Dai Ito; Yoichi Yoshida; Kazuhiko Osa; Osamu Nonaka
- Web page: https://patentgrep.com/patent/US10769497B2

## CPC

- G06N 3/08: Learning methods
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06V 10/776: Validation; Performance evaluation
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A learning device, comprising a reception circuit that receives requests indicating photographs the user likes from external device, a machine learning processor that extracts images that match the requests and that have received a given evaluation from a third party, from within an image database, performs machine learning using these images that have been extracted, and outputs an inference model, and a transmission circuit that transmits an inference model that has been output from the learning processor to the external device.

## Claims

### Claim 1

An imaging device, comprising:
an image input section that generates image data;
a setting circuit that sets a request including a target third party evaluator profile specified based on a user input;
a transmission circuit that transmits the request to an external machine learning device;
a reception circuit that receives at least one inference model that was transmitted directly or indirectly from the external machine leaning device, wherein the at least one inference model was generated by the external machine learning device using image training data which was retrieved from a database using the third party evaluator profile included in the request;
an inference engine that provides an inference result using the received at least one inference model and the image data generated by the image input section; and
a display that displays the inference result that was provided by the inference engine.

### Claim 2 (depends on 1)

The imaging device of claim 1, wherein the image data generated by the image data input section includes through image data, and wherein the display further displays the through image data.

### Claim 3 (depends on 1)

The imaging device of claim 1, wherein:
the external machine learning device, when an inference model, that has reliability in range of image data for which the machine learning has been decided, has been generated, outputs that inference model as an interim inference model.

### Claim 4 (depends on 1)

The imaging device of claim 1, wherein:
the external machine learning device creates image data constituting a population based on the request, from image data that has been stored in an image database.

### Claim 5 (depends on 1)

The imaging device of claim 1, wherein:
the external machine learning device creates advice information for shooting images that fit the inference model, and
the transmission circuit transmits the advice information to the external machine learning device.

### Claim 6 (depends on 1)

The imaging device of claim 1, wherein:
the request further includes a theme of a photograph the user wants.

### Claim 7 (depends on 1)

The imaging device of claim 1, wherein:
the external machine learning device extracts data that matches the requests, and that includes evaluation information for which there is a difference in subjective evaluation for each data, from within a database, performs machine learning using groups of this extracted data, and outputs inference models for subjective evaluation prediction.

### Claim 8 (depends on 7)

The imaging device of claim 7, wherein:
the external machine learning device respectively extracts groups of data for which a value representing evaluation is higher than a given value and groups of data for which a value representing evaluation is lower than a given value, from within the database, performs deep learning, and outputs inference models.

### Claim 9

An imaging method, comprising:
generating image data;
setting a request including a target third party evaluator profile specified based on a user input;
transmitting the request to an external machine learning device;
receiving at least one inference model that was transmitted directly or indirectly from the external machine learning device, wherein the at least one inference model was generated by the external machine learning device using image training data which was retrieved from a database using the third party evaluator profile included in the request;
providing an inference result using the received at least one inference model and the image data generated; and
displaying the inference result that was provided.

### Claim 10 (depends on 9)

The imaging method of claim 9, wherein the image data generated includes through image data, the imaging method further comprising:
displaying the through image data.

### Claim 11 (depends on 9)

The imaging method of claim 9, further comprising:
responsive to generating an inference model that has reliability in range of image data for which the machine learning has been decided, outputting that inference model as an interim inference model.

### Claim 12 (depends on 9)

The imaging method of claim 9, further comprising:
creating image data constituting a population based on the request, from image data that has been stored in an image database.

### Claim 13 (depends on 9)

The imaging method of claim 9, further comprising:
creating advice information for shooting images that fit the inference model; and
transmitting the advice information to the external machine learning device.

### Claim 14 (depends on 9)

The imaging method of claim 9, wherein:
the request further includes a theme of a photograph the user wants.

### Claim 15 (depends on 9)

The imaging method of claim 9, further comprising:
extracting data that matches the requests, and that includes evaluation information for which there is a difference in subjective evaluation for each data, from within a database;
performing machine learning using groups of this extracted data; and
outputting inference models for subjective evaluation prediction.

### Claim 16 (depends on 15)

The imaging method of claim 15, further comprising:
extracting groups of data for which a value representing evaluation is higher than a given value and groups of data for which a value representing evaluation is lower than a given value, from within the database;
performing deep learning, and
outputting inference models.

### Claim 17

A non-transitory computer-readable medium storing processor-executable instructions which, when executed by at least one processor, cause the at least one processor to perform an imaging method, including:
generating image data;
setting a request including a target third party evaluator profile specified based on a user input;
transmitting the request to an external machine learning device;
receiving at least one inference model the external machine learning device, wherein the at least one inference model was generated by the external machine learning device using image training data which was retrieved from a database using the third party evaluator profile included in the request;
providing an inference result using the received at least one inference model and the image data generated; and
displaying the inference result that was provided.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the image data generated includes through image data, the imaging method further including:
displaying the through image data.

### Claim 19 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the method further includes:
creating advice information for shooting images that fit the inference model; and
transmitting the advice information to the external machine learning device.

### Claim 20 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein:
the request further includes a theme of a photograph the user wants.