# US 10,311,317 B2 · Teacher data generation method

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

- Number: US 10,311,317 B2
- Document ID: US10311317B2
- Status: Granted
- Kind: B2
- Application number: 15/434,950
- Filed: 2017-02-16
- Published: 2017-09-28
- Granted: 2019-06-04
- Earliest claimed: 2016-03-25
- Assignees: Honda Motor Co., Ltd.
- Applicants: Honda Motor Co., Ltd.
- Inventors: Tomoyuki Kanamaru
- Web page: https://patentgrep.com/patent/US10311317B2

## CPC

- G06V 10/22: by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06F 18/28: Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
- G06V 10/772: Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Provided is a method for generating teacher data for image recognition while reducing the number of images used as the basis. A captured image is obtained by imaging an object by an imaging device C arranged at a first designated position Pi. A basic image region Si is extracted from the captured image. The teacher data is generated as a result of coordinate transformation of the basic image region Si from one image coordinate system to a coordinate system of a captured image by the imaging device C on the assumption that the imaging device C is arranged at a second designated position Pj which is different from the first designated position Pi.

## Claims

### Claim 1

A method for generating teacher data for image recognition of an object, comprising:
a step of obtaining a plurality of captured images by imaging the object by an imaging device arranged at each of a plurality of first designated positions;
a step of recognizing, by an operator via an input interface, a region that is a substantially flat surface as a basic image region, and extracting the basic image region from each of the plurality of captured images;
a step of generating the teacher data as a result of coordinate transformation of the basic image region from one image coordinate system to each coordinate system of the plurality of captured images by the imaging device on an assumption that the imaging device is arranged at each of a plurality of second designated positions which are different from the plurality of first designated positions; and
a step of using the teacher data for image recognition,
wherein
the step of obtaining the plurality of captured images includes obtaining the plurality of captured images by capturing the object by the imaging device which is arranged at positions of each of a plurality of points on a hemisphere face having an origin of a three-dimensional polar coordinate system as a center, as the each of the plurality of first designated positions, and an imaging device posture is adjusted so that an optical axis thereof is directed to the origin of the three-dimensional polar coordinate system, and
the step of generating the teacher data includes generating, as the teacher data, each of a plurality of results of the coordinate transformation of each of a plurality of the basic image region individually from the one image coordinate system to the each of coordinate system of the plurality of captured images by the imaging device on an assumption that the imaging device is arranged at positions of each of a plurality of points on the hemisphere face as the each of the plurality of second designated positions and that the imaging device posture is adjusted so that the optical axis thereof is directed to the origin of the three-dimensional polar coordinate system.

### Claim 2

A method for generating teacher data for image recognition of an object, comprising:
a step of obtaining a plurality of captured images by imaging the object by an imaging device arranged at each of a plurality of first designated positions;
a step of extracting a basic image region from each of the plurality of captured images; and
a step of generating the teacher data as a result of coordinate transformation of the basic image region from one image coordinate system to each coordinate system of the plurality of captured images by the imaging device on an assumption that the imaging device is arranged at each of a plurality of second designated positions which are different from the plurality of first designated positions,
wherein
the step of obtaining the plurality of captured images includes obtaining the plurality of captured images by capturing the object by the imaging device which is arranged at positions of each of a plurality of points on a hemisphere face having an origin of a three-dimensional polar coordinate system as a center, as the each of the plurality of first designated positions, and an imaging device posture is adjusted so that an optical axis thereof is directed to the origin of the three-dimensional polar coordinate system, and
the step of generating the teacher data includes generating, as the teacher data, each of a plurality of results of the coordinate transformation of each of a plurality of the basic image region individually from the one image coordinate system to the each of coordinate system of the plurality of captured images by the imaging device on an assumption that the imaging device is arranged at positions of each of a plurality of points on the hemisphere face as the each of the plurality of second designated positions and that the imaging device posture is adjusted so that the optical axis thereof is directed to the origin of the three-dimensional polar coordinate system.