# US 2021/0281712 A1 · Reading device, image processing apparatus, method of detecting feature amount, and non-transitory recording medium

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

- Number: US 2021/0281712 A1
- Document ID: US20210281712A1
- Status: Published application
- Kind: A1
- Application number: 17/161,704
- Filed: 2021-01-29
- Published: 2021-09-09
- Earliest claimed: 2020-03-05
- Assignees: Ricoh Company, Ltd.
- Inventors: Shogo NAKAMURA; Ayumu HASHIMOTO; Masamoto NAKAZAWA; Tadaaki OYAMA
- Web page: https://patentgrep.com/patent/US20210281712A1

## CPC

- H04N 1/00718: {Skew}
- H04N 1/3878: {Skew detection or correction}
- H04N 1/00737: {using the scanning elements as detectors}
- H04N 1/00748: {Detecting edges, e.g. of a stationary sheet}
- H04N 1/0075: {Detecting a change in reflectivity}
- H04N 1/00755: {Detecting an interruption of light}
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A reading device includes a light source, an image sensor, a background board, and an image processing circuitry. The light source irradiates visible light and invisible light to a subject. The image sensor receives the visible light and the invisible light, each of which is reflected from the subject, to capture a visible image and an invisible image. The background board is provided in an image capturing range of the image sensor. The background board is a background portion. The image processing circuitry detects a feature amount of the subject and the background portion from at least one of the visible image and the invisible image.

## Claims

### Claim 1

A reading device, comprising:
a light source that irradiates visible light and invisible light to a subject;
an image sensor configured to receive the visible light and the invisible light, each of which is reflected from the subject, to capture a visible image and an invisible image;
a background board provided in an image capturing range of the image sensor, the background board being a background portion; and
an image processing circuitry configured to detect a feature amount of the subject and the background portion from at least one of the visible image or the invisible image.

### Claim 2 (depends on 1)

The reading device, according to claim 1,
wherein the light source irradiates infrared light as the invisible light, and
wherein the image sensor captures an infrared image as the invisible image.

### Claim 3 (depends on 1)

The reading device, according to claim 1,
wherein the image processing circuitry detects the feature amount of the subject and the background portion from at least one of the visible image or the invisible image, and performs selection processing or combination processing depending on at least one of the feature amount detected from the visible image or the feature amount detected from the invisible image.

### Claim 4 (depends on 1)

The reading device, according to claim 1,
wherein the image processing circuitry compares, with respect to the invisible light and the visible light, differences in spectral reflection characteristics between the background portion and the subject, and
the feature amount detected from the visible image including a component having the largest difference from the invisible light among a plurality of components of the visible light.

### Claim 5 (depends on 1)

The reading device, according to claim 1,
wherein the background portion includes an invisible light low reflection portion that diffusely reflects the visible light and reflects the invisible light at a reflectance lower than the reflectance of the visible light.

### Claim 6 (depends on 3)

The reading device, according to claim 3,
wherein the image processing circuitry detects, as the feature amount, an edge of the subject, the edge being at least one of a first edge that is detected from the invisible image or a second edge that is detected from the visible image.

### Claim 7 (depends on 6)

The reading device, according to claim 6,
wherein the image processing circuitry performs OR processing on the first edge detected from the invisible image and the second edge detected from the visible image to combine the first edge and second edge.

### Claim 8 (depends on 6)

The reading device, according to claim 6,
wherein, in a case where the first edge is detected normally from the invisible image, the image processing circuitry selects the first edge,
wherein, in a case where the first edge is not detected normally from the invisible image, the image processing circuitry selects the second edge.

### Claim 9 (depends on 6)

The reading device, according to claim 6,
wherein, in a case where the image processing circuitry fails to normally detect the first edge from the invisible image and the second edge from the visible image, the image processing circuitry performs OR processing on the first edge and the second edge to combine the first edge and second edge.

### Claim 10 (depends on 6)

The reading device, according to claim 6,
wherein the image processing circuitry detects a size of the subject based on the edge.

### Claim 11 (depends on 6)

The reading device, according to claim 6,
wherein the image processing circuitry corrects an inclination and a position of the subject based on the edge.

### Claim 12 (depends on 6)

The reading device, according to claim 6,
wherein the image processing circuitry cuts out the subject based on the edge.

### Claim 13 (depends on 1)

The reading device, according to claim 1,
wherein the image processing circuitry corrects at least one of the visible image or the invisible image.

### Claim 14 (depends on 1)

An image processing apparatus, comprising:
the reading device according to claim 1; and
an image forming device provided below the reading device and configured to form an image.

### Claim 15

A method of detecting a feature amount comprising
detecting a feature amount of a subject and a background portion from at least one of a visible image or an invisible image, each of the visible image and the invisible image being captured by an image sensor receiving visible light and invisible light that is irradiated from a light source and reflected from the subject, the background portion being provided in an image capturing range of the image sensor.

### Claim 16

Anon-transitory recording medium storing a plurality of instructions which, when executed by one or more processors, cause the processors to perform a method comprising
detecting a feature amount of a subject and a background portion from at least one of a visible image or an invisible image, each of the visible image and the invisible image being captured by an image sensor receiving visible light and invisible light that is irradiated from a light source and reflected from the subject, the background portion being provided in an image capturing range of the image sensor.
