Apparatus and method for object tracking in welding process

Patent metadata

Field	Value
Publication number	US-11446755-B2
Application number	US-202016950920-A
Country	US
Kind code	B2
Filing date	Nov 18, 2020
Priority date	Sep 18, 2020
Publication date	Sep 20, 2022
Grant date	Sep 20, 2022

Abstract

According to an embodiment, an object tracking device in a welding process tracks and outputs a predetermined object in a welding image. The object tracking device comprises a camera device capturing the welding image including a base material and a welding torch for welding the base material, a controller receiving a plurality of camera control parameter-varied images from the camera device, identifying the predetermined object in the received images, and generating an object tracking image, the plurality of camera control parameter-varied images having varied camera control parameters of the camera device, and an output device outputting the welding image captured by the camera device, the plurality of images received by the controller, or the object tracking image generated by the controller.

First claim

What is claimed is:

1. An object tracking device in a welding process, the object tracking device tracking and outputting predetermined, at least one object in a welding image, the object tracking device comprising:
a camera device capturing the welding image including a base material and a welding torch for welding the base material;
a controller setting or changing at least one camera control parameter according to a welding environment input by a user, controlling the camera device to capture a plurality of images based on the at least one camera control parameter, receiving the captured images from the camera device, identifying the at least one object in the received images, and generating an object tracking image including coordinates or a trajectory of the at least one object; and
an output device outputting the welding image, the plurality of images, or the object tracking image, wherein the controller identifies coordinates of a first object of the at least one object according to first settings of the at least one camera control parameter, identifies coordinates of a second object of the at least one object according to second settings of the at least one camera control parameter, and identifies coordinates of a third object of the at least one object according to third settings of the at least one camera control parameter.

2. The object tracking device of claim 1 , wherein
the base material includes a marker that may be identified by the controller.

3. The object tracking device of claim 2 , wherein
the at least one object includes the welding torch, a welding tip, and the marker.

4. The object tracking device of claim 1 , wherein
the at least one camera control parameter includes at least one of a camera exposure value, a camera sensor color gain value, a gamma value, and a black level.

5. The object tracking device of claim 2 , wherein
the marker has an opposite reflection spectrum from a spectrum of welding light.

6. The object tracking device of claim 1 , wherein
the controller performs pre-treatment to remove, subtract, or reduce a black image captured by the camera device according to each camera control parameter from the received images to remove noise.

7. A tracking controller tracking predetermined, at least one object in a welding image obtained using a camera, the tracking controller comprising:
a user input unit receiving an input for a welding environment from a user;
a camera controller setting or changing at least one camera control parameter of the camera according to the welding environment received by the user, controlling the camera to a plurality of images based on the at least one camera control parameter, and receiving the captured images from the camera;
a learning unit designating the at least one object in a plurality of welding images and generating learning data based on feature information about the at least one object;
a memory storing the at least one camera control parameter according to the welding environment and the plurality of welding images by which the learning unit is able to generate the learning data; and
an object tracking unit identifying the at least one object from the plurality of images received by the camera controller from the camera and generating an object tracking image including coordinates or a trajectory of the at least one object, wherein the controller identifies coordinates of a first object of the at least one object according to first settings of the at least one camera control parameter, identifies coordinates of a second object of the at least one object according to second settings of the at least one camera control parameter, and identifies coordinates of a third object of the at least one object according to third settings of the at least one camera control parameter.

8. The tracking controller of claim 7 , wherein the camera controller varies the at least one camera control parameter of the camera into the at least one camera control parameter according to the welding environment and receives the plurality of camera control parameter-varied images from the camera.

9. The tracking controller of claim 7 , wherein
the at least one object includes a welding torch, a welding tip, and a marker included in a base material.

10. The tracking controller of claim 7 , wherein
the at least one camera control parameter includes at least one of a camera exposure value, a camera sensor color gain value, a gamma value, and a black level.

11. An object tracking method in a welding process to track and output predetermined, at least one object in a welding image, the object tracking method comprising:
capturing the welding image using a camera device, the welding image including a base material and a welding torch for welding the base material;
setting or changing at least one camera control parameter of the camera device according to a welding environment input by a user;
controlling the camera device to capture a plurality of images based on the at least one camera control parameter;
receiving the captured images from the camera device;
identifying the at least one object from the plurality of received images and generating an object tracking image including coordinates or a trajectory of the at least one object; and
outputting the captured images, the received images, or the generated object tracking image, wherein the controller identifies coordinates of a first object of the at least one object according to first settings of the at least one camera control parameter, identifies coordinates of a second object of the at least one object according to second settings of the at least one camera control parameter, and identifies coordinates of a third object of the at least one object according to third settings of the at least one camera control parameter.

12. The object tracking method of claim 11 , wherein
the base material includes a marker that may be identified by a controller.

13. The object tracking method of claim 11 , wherein
the at least one object includes the welding torch, a welding tip, and a marker included in the base material.

Assignees

• Korea Photonics Tech Inst

Inventors

• Kim Hoe Min
• Chun Sung Kuk
• Lee Dong Kil
• Lee Kwang Hoon
• Kim Seon Man

Classifications

• B23K26/032 Primary

  using optical means · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title

• B23K26/21

  by welding · CPC title

• G06T2207/30204

  Marker · CPC title

• B23K9/0953

  using computing means · CPC title