# US 2021/0081652 A1 · Image processing apparatus and image processing method

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

- Number: US 2021/0081652 A1
- Document ID: US20210081652A1
- Status: Published application
- Kind: A1
- Application number: 16/996,962
- Filed: 2020-08-19
- Published: 2021-03-18
- Earliest claimed: 2019-09-13
- Assignees: Fujitsu Limited
- Applicants: Fujitsu Limited
- Inventors: Tomoaki Matsunami
- Web page: https://patentgrep.com/patent/US20210081652A1

## CPC

- G06V 40/171: {Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships}
- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06V 40/168: {Feature extraction; Face representation}
- G06V 40/169: {Holistic features and representations, i.e. based on the facial image taken as a whole}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An image processing apparatus includes memory that stores a plurality of images acquired by image-capturing a face of a person; and a processor, coupled to the memory, configured to: acquire a pixel value pattern indicating a relationship between positions in a given direction and pixel values in an area having parts of the face of each of the plurality of images, and select an image to be processed from the plurality of images based on the pixel value patterns of the plurality of images.

## Claims

### Claim 1

An image processing apparatus, comprising:
a memory that stores a plurality of images acquired by image-capturing a face of a person; and
a processor, coupled to the memory, configured to:
acquire a pixel value pattern indicating a relationship between positions in a given direction and pixel values in an area having parts of the face of each of the plurality of images, and
select an image to be processed from the plurality of images based on the pixel value patterns of the plurality of images.

### Claim 2 (depends on 1)

The image processing apparatus according to claim 1,
wherein, at positions of pixels in the given direction, the processor calculates a statistic on pixel values from pixel values of a plurality of pixels aligned in a direction perpendicular to the given direction, acquires, as the pixel value pattern, a pattern indicating a relationship between positions of pixels in the given direction and the statistic, calculates a characteristic value by using the pixel value pattern, and selects the image to be processed based on the characteristic values of the plurality of images.

### Claim 3 (depends on 2)

The image processing apparatus according to claim 2,
wherein, in an area including eyes, a nose, and a mouth as the parts of the face, the processor acquires the pixel value pattern by using a top-bottom direction of the face as the given direction and using intensity values as the pixel values and calculates a third harmonic component in a frequency characteristic of the pixel value pattern as the characteristic value.

### Claim 4 (depends on 2)

The image processing apparatus according to claim 2,
wherein, in an area including eyes as the parts of the face, the processor acquires the pixel value pattern by using a right-left direction of the face as the given direction and using intensity values as the pixel values and calculates a second harmonic component in a frequency characteristic of the pixel value pattern as the characteristic value.

### Claim 5 (depends on 2)

The image processing apparatus according to claim 2,
wherein, in an area including eyes, a nose, and a mouth as the parts of the face, the processor acquires a pixel value pattern in a top-bottom direction of the face by using the top-bottom direction of the face as the given direction and using intensity values as the pixel values and calculates a third harmonic component in a frequency characteristic of the pixel value pattern in the top-bottom direction of the face as the characteristic value, and, in an area including the eyes as the parts of the face, acquires a pixel value pattern in a right-left direction of the face by using the right-left direction of the face as the given direction and using intensity values as the pixel values, calculates a second harmonic component in the frequency characteristic of the pixel value pattern in the right-left direction of the face as the characteristic value and selects the image to be processed based on the third harmonic component of each of the plurality of images and the second harmonic component of each of the plurality of images.

### Claim 6 (depends on 2)

The image processing apparatus according to claim 2, wherein the processor
calculates a frontal degree indicating how much the face included in each of the plurality of images turns to the front, and
selects an image to be processed based on the characteristic value of each of the plurality of images and the frontal degree of each of the plurality of images.

### Claim 7 (depends on 6)

The image processing apparatus according to claim 6,
wherein the processor selects a given number of images as the images to be processed by repeating operations of calculating a statistic of each of the plurality of images, referring to characteristic values of the plurality of images in decreasing order of the frontal degrees, and, when a difference between the referred characteristic value and the statistic on the characteristic value is higher than a threshold value, referring to the characteristic value of a next image, and when the difference is lower than the threshold value, selecting the image having the referred characteristic value as the image to be processed.

### Claim 8 (depends on 1)

The image processing apparatus according to claim 1,
wherein the processor obtains the plurality of images by repeating operations of referring to a captured time of an image including the face in order of captured times, and when a difference between the referred captured time and the captured time of an immediately previously obtained image is shorter than a given time, referring to a captured time of a next image, and when the difference between the referred captured time and the captured time of the immediately previously obtained image is longer than the given time, obtaining the image having the referred captured time.

### Claim 9 (depends on 1)

The image processing apparatus according to claim 1,
wherein the image to be processed is an image to be registered or to be authenticated.

### Claim 10

A non-transitory computer-readable storage medium having stored therein a program for causing a processor to execute a process, the process comprising:
acquiring a pixel value pattern indicating a relationship between positions in a given direction and pixel values in an area having parts of a face of each of the plurality of images acquired by image-capturing a face of a person; and
selecting an image to be processed from the plurality of images based on the pixel value patterns of the plurality of images.

### Claim 11 (depends on 10)

The non-transitory computer-readable storage medium according to claim 10,
wherein, at positions of pixels in the given direction, the processor calculates a statistic on pixel values from pixel values of a plurality of pixels aligned in a direction perpendicular to the given direction, acquires, as the pixel value pattern, a pattern indicating a relationship between positions of pixels in the given direction and the statistic, calculates a characteristic value by using the pixel value pattern, and selects the image to be processed based on the characteristic values of the plurality of images.

### Claim 12 (depends on 10)

The non-transitory computer-readable storage medium according to claim 10,
wherein, in an area including eyes, a nose, and a mouth as the parts of the face, the processor acquires the pixel value pattern by using a top-bottom direction of the face as the given direction and using intensity values as the pixel values and calculates a third harmonic component in a frequency characteristic of the pixel value pattern as the characteristic value.

### Claim 13

A method for causing a computer to execute processing, comprising:
acquiring a pixel value pattern indicating a relationship between positions in a given direction and pixel values in an area having parts of a face of each of the plurality of images acquired by image-capturing a face of a person; and
selecting an image to be processed from the plurality of images based on the pixel value patterns of the plurality of images.

### Claim 14 (depends on 13)

The method according to claim 13,
wherein, at positions of pixels in the given direction, the processor calculates a statistic on pixel values from pixel values of a plurality of pixels aligned in a direction perpendicular to the given direction, acquires, as the pixel value pattern, a pattern indicating a relationship between positions of pixels in the given direction and the statistic, calculates a characteristic value by using the pixel value pattern, and selects the image to be processed based on the characteristic values of the plurality of images.

### Claim 15 (depends on 13)

The method according to claim 13,
wherein, in an area including eyes, a nose, and a mouth as the parts of the face, the processor acquires the pixel value pattern by using a top-bottom direction of the face as the given direction and using intensity values as the pixel values and calculates a third harmonic component in a frequency characteristic of the pixel value pattern as the characteristic value.
