# US 2023/0056676 A1 · System and Method for Object Detection and Dimensioning

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

- Number: US 2023/0056676 A1
- Document ID: US20230056676A1
- Status: Published application
- Kind: A1
- Application number: 17/405,919
- Filed: 2021-08-18
- Published: 2023-02-23
- Earliest claimed: 2021-08-18
- Assignees: Zebra Technologies Corporation
- Applicants: Zebra Technologies Corporation
- Inventors: Raveen T. Thrimawithana
- Web page: https://patentgrep.com/patent/US20230056676A1

## CPC

- G06F 18/2137: based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
- G06F 18/23: Clustering techniques
- G06T 7/11: Region-based segmentation
- G06T 7/194: involving foreground-background segmentation
- G06T 7/50: Depth or shape recovery
- G06V 10/757: {Matching configurations of points or features}
- G06V 10/762: using clustering, e.g. of similar faces in social networks
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A method for detecting and dimensioning a target object includes: obtaining depth data representing a target object; defining a mask having a structure which decreases in density away from a central point of the mask; overlaying the mask on the depth data and selecting a subset of the depth data comprising data points which contact the mask; detecting a cluster of data points from the subset; detecting, based on the cluster, the target object; and outputting a representation of the target object.

## Claims

### Claim 1

A method comprising:
obtaining, from a depth sensor, depth data representing a target object;
defining a mask having a structure which decreases in density away from a central point of the mask;
overlaying the mask on the depth data and selecting a subset of the depth data comprising data points having coordinates which match coordinates of a point of the mask;
detecting a cluster of data points from the subset;
detecting, based on the cluster, the target object; and
determining an object dimension of the target object based on the cluster.

### Claim 2 (depends on 1)

The method of claim 1, wherein the mask comprises an arrangement of lines which decrease in density away from the central point.

### Claim 3 (depends on 2)

The method of claim 2, wherein the lines extend radially from the central point of the mask.

### Claim 4 (depends on 1)

The method of claim 1, wherein overlaying the mask on the depth data comprises aligning the central point with a center of a frame of the depth data.

### Claim 5 (depends on 1)

The method of claim 1, wherein detecting the cluster of data points from the subset comprises:
selecting a starting point from the subset of the depth data, the starting point comprising a closest data point of the subset to the central point of the mask; and
selecting the cluster of data points about the starting point.

### Claim 6 (depends on 1)

The method of claim 1, wherein detecting the target object comprises applying plane and object detection to the detected cluster of data points.

### Claim 7 (depends on 1)

The method of claim 1, wherein detecting the target object comprises:
generating a bounding region for the cluster of data points;
selecting a second subset of the depth data comprising data points contained in the bounding region; and
applying plane and object detection to the second subset of depth data to identify the target object.

### Claim 8

A device comprising:
a depth sensor configured to obtain depth data representing a target object;
a memory;
a processor interconnected with the depth sensor and the memory, the processor configured to:
obtain, from the depth sensor, depth data representing the target object;
define a mask having a structure which decreases in density away from a central point of the mask;
overlay the mask on the depth data and select a subset of the depth data comprising data points having coordinates which match coordinates of a point of the mask;
detect a cluster of data points from the subset;
detect, based on the cluster, the target object; and
determine an object dimension of the target object based on the cluster.

### Claim 9 (depends on 8)

The device of claim 8, wherein the mask comprises an arrangement of lines which decrease in density away from the central point.

### Claim 10 (depends on 9)

The device of claim 9, wherein the lines extend radially from the central point of the mask.

### Claim 11 (depends on 8)

The device of claim 8, wherein, to overlay the mask on the depth data, the processor is configured to align the central point of the mask with a center of a frame of the depth data.

### Claim 12 (depends on 8)

The device of claim 8, wherein, to detect the cluster of data points from the subset, processor is configured to:
select a starting point from the subset of the depth data, the starting point comprising a closest data point of the subset to the central point of the mask; and
select the cluster of data points about the starting point.

### Claim 13 (depends on 8)

The device of claim 8, wherein to detect the target object, the processor is configured to: apply plane and object detection to the detected cluster of data points.

### Claim 14 (depends on 8)

The device of claim 8, wherein to detect the target object, the processor is configured to:
generate a bounding region for the cluster of data points;
select a second subset of the depth data comprising data points contained in the bounding region; and
apply plane and object detection to the second subset of depth data to identify the target object.

### Claim 16

A non-transitory computer-readable medium storing a plurality of computer-readable instructions executable by a processor, wherein execution of the instructions configures the processor to:
obtain, from a depth sensor, depth data representing a target object;
define a mask having a structure which decreases in density away from a central point of the mask;
overlay the mask on the depth data and select a subset of the depth data comprising data points having coordinates which match coordinates of a point of the mask;
detect a cluster of data points from the subset;
detect, based on the cluster, the target object; and
determine an object dimension of the target object based on the cluster.

### Claim 17 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the mask comprises an arrangement of lines which decrease in density away from the central point.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the lines extend radially from the central point of the mask.

### Claim 19 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein, to overlay the mask on the depth data, the instructions configure the processor to: align the central point of the mask with a center of a frame of the depth data.

### Claim 20 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein, to detect the cluster of data points from the subset, instructions configure the processor to:
select a starting point from the subset of the depth data, the starting point comprising a closest data point of the subset to the central point of the mask; and
select the cluster of data points about the starting point.

### Claim 21 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein to detect the target object, the instructions configure the processor to: apply plane and object detection to the detected cluster of data points.

### Claim 22 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein to detect the target object, the instructions configure the processor to:
generate a bounding region for the cluster of data points;
select a second subset of the depth data comprising data points contained in the bounding region; and
apply plane and object detection to the second subset of depth data to identify the target object.

### Claim 23 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the instructions further configure the processor to: dimension the target object and output dimensions of the target object.