# US 2016/0253580 A1 · Image processing apparatus and method for detecting transparent object in image

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

- Number: US 2016/0253580 A1
- Document ID: US20160253580A1
- Status: Published application
- Kind: A1
- Application number: 15/151,866
- Filed: 2016-05-11
- Published: 2016-09-01
- Earliest claimed: 2012-10-24
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Seung Won JUNG; Seung Kyu LEE; Ouk CHOI; Do Kyoon KIM; Kee Chang LEE
- Web page: https://patentgrep.com/patent/US20160253580A1

## CPC

- H04N 23/00: Cameras or camera modules comprising electronic image sensors; Control thereof
- G06F 18/21: Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Provided is an image processing apparatus and method for detecting a transparent image from an input image. The image processing apparatus may include an image segmenting unit to segment an input image into a plurality of segments, a likelihood determining unit to determine a likelihood that a transparent object is present between adjacent segments among the plurality of segments, and an object detecting unit to detect the transparent object from the input image based on the likelihood.

## Claims

### Claim at position 1

1-28. (canceled)

### Claim 29

An apparatus, comprising:
one or more cameras; and
one or more processors configures to,
obtain an image via the one or more cameras,
segment the image into a plurality of areas including a first area and a second area;
determine whether an area corresponding to a transparent object is present between the first area and the second area, and
detect the area from the plurality of areas based at least in part on a determination that the area is present between the first area and the second area.

### Claim 30 (depends on 29)

The apparatus of claim 29, wherein the one or more processors segments the input image based on at least one of a depth image and an amplitude image included in the input image.

### Claim 31 (depends on 29)

The apparatus of claim 29, wherein:
the one or more processors determines whether a area pair including the first area and the second area belong to a group associated with a transparency attribute, and
the group associated with the transparency attribute comprises a area pair in which at least one area is transparent.

### Claim 32 (depends on 29)

The apparatus of claim 29, wherein the one or more processors determines whether a area pair including the first area and the second area belong to a group associated with a transparency attribute, using a learning model associated with feature extraction.

### Claim 33 (depends on 32)

The apparatus of claim 32, wherein the learning model extracts a feature based on amplitude and depth of the respective areas constituting the area pair using a test image.

### Claim 34 (depends on 29)

The apparatus of claim 29, wherein the one or more processors detects the area using one of a global optimization scheme and a local optimization scheme.

### Claim 35

An apparatus, comprising:
one or more cameras; and
one or more processors configured to,
obtain a test image via the one or more cameras,
segment the test image into a plurality of areas including a first area and a second area,
determine a group associated with a transparency attribute that is belonged by a area pair including the first area and the second area among the plurality of segments, and
extract a feature associated with the area pair based on the group belonged by the area pair.

### Claim 36 (depends on 35)

The apparatus of claim 35, wherein the group determining unit determines whether at least one of the first area and the second area included in the area pair is associated with a transparent object, based on information associated with the transparent object of the test image, and determines the group belonged by the area pair based on the determination result.

### Claim 37

An apparatus, comprising:
one or more cameras; and
one or more processors configured to,
perform a learning process of extracting a feature associated with a transparent object using a test image,
obtain an input image via the one or more cameras,
determine whether an area corresponding to a transparent object is present between the first area and the second area using a learned feature, and
detect the area from the plurality of areas based at least in part on a determination that the area is present between the first area and the second area.

### Claim 38 (depends on 37)

The apparatus of claim 37, wherein the one or more processors
obtain a test image via the one or more cameras;
segment the test image into a plurality of areas including a first area and a second area;
determine a group associated with a transparency attribute that is belonged by a area pair including the first area and the second area among the plurality of segments; and
extract a feature associated with the area pair based on the group belonged by the area pair.

### Claim 39 (depends on 38)

The apparatus of claim 38, wherein the one or more processors determines whether at least one of the first area and the second area included in the area pair is associated with the transparent object, based on information associated with the transparent object of the test image, and determines the group belonged by the area pair.

### Claim 40 (depends on 37)

The apparatus of claim 37, wherein the one or more processors segments the input image based on at least one of a depth image and an amplitude image included in the input image.

### Claim 41 (depends on 37)

The apparatus of claim 37, wherein the one or more processors detects the area using one of a global optimization scheme and a local optimization scheme.
