# US 2021/0174536 A1 · Queue analyzing method and image monitoring apparatus

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

- Number: US 2021/0174536 A1
- Document ID: US20210174536A1
- Status: Published application
- Kind: A1
- Application number: 16/831,849
- Filed: 2020-03-27
- Published: 2021-06-10
- Earliest claimed: 2019-12-10
- Assignees: Vivotek Inc.
- Applicants: Vivotek Inc.
- Inventors: Cheng-Chieh Liu
- Web page: https://patentgrep.com/patent/US20210174536A1

## CPC

- G06V 20/53: {Recognition of crowd images, e.g. recognition of crowd congestion}
- G06F 16/53: Querying
- G06T 7/70: Determining position or orientation of objects or cameras (camera calibration G06T7/80)
- G06T 2207/30204: Marker
- G06T 2207/30242: Counting objects in image
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A queue analyzing method is applied to an image monitoring apparatus and can automatically generate an interval threshold according to position variation of objects. The queue analyzing method includes computing a plurality of intervals between all objects inside an image, dividing the plurality of intervals at least into a first group corresponding to a low interval range and a second group corresponding to a high interval range, computing an interval mean value and an interval amending value of objects inside the first group, utilizing the interval mean value and the interval amending value to generate the interval threshold, and marking some adjacent objects conforming to the interval threshold within the image.

## Claims

### Claim 1

A queue analyzing method of automatically generating an interval threshold for queue analysis in accordance with position variation of objects, the queue analyzing method comprising:
computing a plurality of intervals between all the objects inside an image;
dividing the plurality of intervals at least into a first group corresponding to a low interval range and a second group corresponding to a high interval range;
computing an interval mean value and an interval amending value of objects inside the first group;
utilizing the interval mean value and the interval amending value to generate the interval threshold; and
marking adjacent objects conforming to the interval threshold inside the image.

### Claim 2 (depends on 1)

The queue analyzing method of claim 1, wherein computing the plurality of intervals between all the objects inside the image comprises:
defining a major object and other auxiliary objects inside the image;
dividing the auxiliary objects at least into a first set close to the major object and a second set distant from the major object; and
computing intervals between the major object and all objects inside the first set for being the plurality of intervals.

### Claim 3 (depends on 1)

The queue analyzing method of claim 1, wherein the interval amending value is an interval standard deviation, and the interval threshold is a computation result generated by the interval standard deviation and the interval mean value.

### Claim 4 (depends on 1)

The queue analyzing method of claim 1, wherein the interval amending value is a weighting, and the interval threshold is a weighted result generated by the weighting and the interval mean value.

### Claim 5 (depends on 1)

The queue analyzing method of claim 1, further comprising:
removing some objects not conforming to the interval threshold; and
utilizing surplus objects inside the image to acquire the plurality of intervals and the related interval threshold.

### Claim 6 (depends on 1)

The queue analyzing method of claim 1, further comprising:
reserving some objects inside the image according to an input command; and
utilizing the reserved objects to acquire the plurality of intervals and the related interval threshold.

### Claim 7 (depends on 1)

The queue analyzing method of claim 1, further comprising:
drawing an indication pattern inside the image according to an input command;
marking several objects with distances relative to the indication pattern conforming to a specific condition; and
utilizing the marked objects to acquire the plurality of intervals and the related interval threshold.

### Claim 8 (depends on 1)

The queue analyzing method of claim 1, further comprising:
defining a region of interest inside the image; and
numbering the adjacent objects conforming to the interval threshold in sequence based on the region of interest.

### Claim 9 (depends on 1)

The queue analyzing method of claim 1, further comprising:
computing and displaying an actual interval between the adjacent objects when the adjacent objects conforming to the interval threshold are marked.

### Claim 10 (depends on 1)

The queue analyzing method of claim 1, further comprising:
marking the adjacent objects conforming to the interval threshold to form a queue;
defining at least one ending object in the queue; and
generating a stretching region via the interval threshold inside the image based on the at least one ending object.

### Claim 11

An image monitoring apparatus, comprising:
an image receiver adapted to receive an image; and
an operation processor electrically connected to the image receiver and adapted to analyze the image for finding all objects, the operation processor being further adapted to compute a plurality of intervals between all the objects inside an image, divide the plurality of intervals at least into a first group corresponding to a low interval range and a second group corresponding to a high interval range, compute an interval mean value and an interval amending value of objects inside the first group, utilize the interval mean value and the interval amending value to generate the interval threshold, and mark adjacent objects conforming to the interval threshold inside the image.

### Claim 12 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to define a major object and other auxiliary objects inside the image, divide the auxiliary objects at least into a first set close to the major object and a second set distant from the major object, and compute intervals between the major object and all objects inside the first set for being the plurality of intervals.

### Claim 13 (depends on 11)

The image monitoring apparatus of claim 11, wherein the interval amending value is an interval standard deviation, and the interval threshold is a computation result generated by the interval standard deviation and the interval mean value.

### Claim 14 (depends on 11)

The image monitoring apparatus of claim 11, wherein the interval amending value is a weighting, and the interval threshold is a weighted result generated by the weighting and the interval mean value.

### Claim 15 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to remove some objects not conforming to the interval threshold, and utilize surplus objects inside the image to acquire the plurality of intervals and the related interval threshold.

### Claim 16 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to reserve some objects inside the image according to an input command, and utilize the reserved objects to acquire the plurality of intervals and the related interval threshold.

### Claim 17 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to draw an indication pattern inside the image according to an input command, mark several objects with distances relative to the indication pattern conforming to a specific condition, and utilize the marked objects to acquire the plurality of intervals and the related interval threshold.

### Claim 18 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to define a region of interest inside the image, and number the adjacent objects conforming to the interval threshold in sequence based on the region of interest.

### Claim 19 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to compute and displaying an actual interval between the adjacent objects when the adjacent objects conforming to the interval threshold are marked.

### Claim 20 (depends on 11)

The image monitoring apparatus of claim 11, wherein the operation processor is further adapted to mark the adjacent objects conforming to the interval threshold to form a queue, define at least one ending object in the queue, and generate a stretching region via the interval threshold inside the image based on the at least one ending object.
