# US 2010/0245554 A1 · Vision watching system and method for safety hat

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

- Number: US 2010/0245554 A1
- Document ID: US20100245554A1
- Status: Published application
- Kind: A1
- Application number: 12/588,664
- Filed: 2009-10-22
- Published: 2010-09-30
- Earliest claimed: 2009-03-24
- Assignees: Ajou University Industry-Academic Cooperation
- Inventors: Yun Young Nam; Moo Rim Kim; Suck Chul Kim; We Duke Cho
- Web page: https://patentgrep.com/patent/US20100245554A1

## CPC

- H04N 7/18: Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- G06T 7/0002: {Inspection of images, e.g. flaw detection}
- G06T 2207/30196: Human being; Person
- G06T 2207/30232: Surveillance
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Provided are vision watching system and method for a safety hat. The vision watching system includes: a receiver which receives an image signal of a watching zone of a camera; a processor which determines whether an object of the image signal is a person, based on a size and an outline of the object and determines whether the person wears a safety hat, based on at least one or more of red (R), green (G), and blue (B) values of a head part of the person and a Hue value of Hue, Saturation, Value (HSV) of the head part; and a transmitter which outputs a Pan-Tilt-Zoom (PTZ) control signal of the camera to obtain a thumbnail image of the person when determining the person who does not wear the safety hat.

## Claims

### Claim 1

A vision watching system for a safety hat, comprising:
a receiver which receives an image signal of a watching zone of a camera;
a processor which determines whether an object of the image signal is a person, based on a size and an outline of the object and determines whether the person wears a safety hat, based on at least one or more of red (R), green (G), and blue (B) values of a head part of the person and a Hue value of Hue, Saturation, Value (HSV) of the head part; and
a transmitter which outputs a Pan-Tilt-Zoom (PTZ) control signal of the camera to obtain a thumbnail image of the person when determining the person who does not wear the safety hat.

### Claim 2 (depends on 1)

The vision watching system of claim 1, wherein the receiver receives the thumbnail image of the person obtained by the PTZ control signal of the camera, and the transmitter transmits the thumbnail image.

### Claim 3

A vision watching system for a safety hat, comprising:
a camera which photographs a watching zone;
an image signal processor which receives an image signal of the camera watching area, determines whether an object of the image signal is a person, based on a size and an outline of the object, and determines whether the person wears a safety hat, based on at least one or more of RGB values of a head part of the person and a H value of HSV of the head part to output a PTZ control signal of the camera for obtaining a thumbnail image of the person when determining the person; and
a controller which displays the thumbnail image on a screen, generates events on the track of a moving path of the person, and communicates the fact that the person does not wear the safety hat to the person in real time.

### Claim 4 (depends on 1)

The vision watching system of claim 1, wherein the camera is a camera having a resolution less than or equal to a predetermined reference value, and the image signal processor determines whether the person wears the safety hat, based on comparison results between the RGB values and pre-set reference RGB values or a comparison result between the H value of the HSV and a H value of pre-set reference HSV.

### Claim 5 (depends on 1)

The vision watching system of claim 1, wherein a position of a pixel is extracted from two points having greatest sizes of the object to detect the head part of the person.

### Claim 6 (depends on 3)

The vision watching system of claim 3, wherein the moving path of the person is tracked based on a central point, a histogram, a direction, and a size of the person.

### Claim 7

A vision watching method for a safety hat, comprising:
receiving an image signal of a watching zone of a camera;
determining whether an object of the image signal is a person, based on a size and an outline of the object and determining whether the person wears a safety hat, based on at least one or more of RGB values of a head part of the person and a H value of HSV of the head part;
outputting a PTZ control signal of the camera for obtaining a thumbnail image of the person when determining the person;
receiving the thumbnail image of the person; and
displaying the thumbnail image on a screen, generating events on the track of a moving path of the person, and communicating the fact that the person does not wear the safety hat to the person in real time.

### Claim 8 (depends on 7)

The vision watching method of claim 7, wherein the camera is a camera having a resolution less than or equal to a predetermined reference value, and the processor determines whether the person wears the safety hat, based on comparison results between the RGB values and pre-set reference RGB values or a comparison result between the H value and a H value of a pre-set reference HSV.

### Claim 9 (depends on 7)

The vision watching method of claim 7, wherein a position of a pixel is extracted from two points having two greatest sizes of the object to detect the head part of the person.

### Claim 10 (depends on 7)

The vision watching method of claim 7, wherein the moving path of the person is tracked based on a central point, a histogram, a direction, and a size of the person.
