# US 2023/0264302 A1 · Repair welding segment detection method and repair welding segment detection device

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

- Number: US 2023/0264302 A1
- Document ID: US20230264302A1
- Status: Published application
- Kind: A1
- Application number: 18/140,213
- Filed: 2023-04-27
- Published: 2023-08-24
- Earliest claimed: 2020-10-28
- Assignees: Panasonic Intellectual Property Management Co., Ltd.
- Applicants: Panasonic Intellectual Property Management Co., Ltd.
- Inventors: Katsuaki OKUMA
- Web page: https://patentgrep.com/patent/US20230264302A1

## CPC

- B23K 31/125: {Weld quality monitoring}
- B23K 9/0956: {using sensing means, e.g. optical}
- B23K 26/032: {using optical means}
- G01N 21/88: Investigating the presence of flaws or contamination
- G06T 7/001: {using an image reference approach}
- G06V 10/255: {Detecting or recognising potential candidate objects based on visual cues, e.g. shapes}
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G06T 2207/30152: Solder
- G06V 2201/121: using special illumination
- B23K: SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A repair welding segment detection method includes generating, based on a result of the inspection determination, shape mismatch data obtained by extracting a shape mismatch portion of the weld bead, dividing the shape mismatch data into N. where N is an integer of 2 or more, equally divided windows in a direction perpendicular to a welding direction of the weld bead, setting a shift region formed by i, where i: an integer of 1 or more, continuous windows among the N windows, separately calculating volumes of (N-i+1) shift regions obtained by shifting one by one the i windows forming the shift region in the welding direction, and determining that a shift region having a volume of a predetermined value or more among the calculated volumes of the (N-i+1) respective shift regions is a defective segment of the weld bead.

## Claims

### Claim 1

A repair welding segment detection method comprising:
entering input data regarding a weld bead of a workpiece produced by welding;
executing inspection determination regarding a shape of the weld bead by using the input data and master data of a non-defective workpiece:
generating, based on a result of the inspection determination, shape mismatch data obtained by extracting a shape mismatch portion of the weld bead;
dividing the shape mismatch data into N, where N is an integer of 2 or more, equally divided windows in a direction perpendicular to a welding direction of the weld bead;
setting a shift region formed by i, where i: an integer of 1 or more, continuous windows among the N windows;
separately calculating volumes of (N-i+1 ) shift regions obtained by shifting one by one the i windows forming the shift region in the welding direction; and
determining that a shift region having a volume of a predetermined value or more among the calculated volumes of the (N-i+1) respective shift regions is a defective segment of the weld bead.

### Claim 2 (depends on 1)

The repair welding segment detection method according to claim 1, wherein
the processings of calculating the volumes of the (N-i+1) respective shift regions and determining the defective segment are sequentially executed (N-i+1) times along the welding direction.

### Claim 3 (depends on 1)

The repair welding segment detection method according to claim 1, further comprising:
determining, as a start point of the defective segment, a first window among the i windows included in the shift region determined to be the defective segment, and
determining an i-th window as an end point of the defective segment.

### Claim 4 (depends on 2)

The repair welding segment detection method according to claim 2, further comprising:
in a case that two or more continuous shift regions are determined to be the defective segment, determining, as a start point of the defective segment, a first window among the i windows forming a shift region firstly determined as the defective segment among the two or more continuous shift regions, and determining, as an end point of the defective segment, an i-th window among the i windows included in a shift region last determined as the defective segment.

### Claim 5 (depends on 2)

The repair welding segment detection method according to claim 2,further comprising:
in a case that a first shift region among the (N-i+1) shift regions is determined as the defective segment, determining, as a start point of the defective segment, a first window among the i windows forming the first shift region.

### Claim 6 (depends on 2)

The repair welding segment detection method according to claim 2, further comprising:
in a case that an (N-i+1)-th shift region among the (N-i+1) shift regions is determined as the defective segment, determining, as an end point of the defective segment, an i-th window among the i windows forming the (N-i+1)-th shift region.

### Claim 7 (depends on 2)

The repair welding segment detection method according to claim 2, further comprising:
in a case that two or more continuous shift regions are determined to be the defective segment, determining, as a start point of the defective segment, a center position of a shift region firstly determined as the defective segment, and determining, as an end point, a center position of the shift region last determined as the defective segment.

### Claim 8 (depends on 2)

The repair welding segment detection method according to claim 2, further comprising:
in a case that any k-th, where k is an integer of 1 or more, shift region among the (N-i+1) shift regions is determined to be the defective segment, setting, as a start point of the defective segment, any window included in the shift region and determining whether a volume of a (k+1)-th shift region continuous to the shift region is equal to or smaller than a second predetermined value smaller than the predetermined value; and
in a case that the volume of the (k+1)-th shift region is equal to or smaller than the second predetermined value, determining, as an end point of the defective segment, any window forming the (k+1)-th shift region.

### Claim 9 (depends on 8)

The repair welding segment detection method according to claim 8, further comprising:
in a case that the volume of the (k+1)-th shift region is not equal to or smaller than the second predetermined value, determining whether a volume of a (k+2)-th shift region continuous to the (k+1)-th shift region is equal to or smaller than the second predetermined value; and
repeatedly executing the determination until any one of calculated volumes of shift regions is determined to be equal to or smaller than the second predetermined value, and determining, as the end point of the defective segment, any window forming the shift region having a volume determined to be equal to or smaller than the second predetermined value.

### Claim 10 (depends on 9)

The repair welding segment detection method according to claim 9, further comprising:
in a case that a volume of an (N-i+1)-th shift region is determined to be not equal to or smaller than the second predetermined value by the determination, determining, as the end point of the defective segment, any window forming the (N-i+1)-th shift region.

### Claim 11 (depends on 3)

The repair welding segment detection method according to claim 3,further comprising:
determining, as coordinates of the start point, barycentric coordinates of a window including the start point of the defective segment.

### Claim 12 (depends on 11)

The repair welding segment detection method according to claim 11, further comprising:
determining, as the coordinates of the start point, coordinates of a middle point between a point closest to and a point farthest from a welding trajectory of the weld bead in the window including the start point.

### Claim 13 (depends on 3)

The repair welding segment detection method according to claim 3, further comprising:
determining, as coordinates of the end point, barycentric coordinates of a window including the end point of the defective segment.

### Claim 14 (depends on 13)

The repair welding segment detection method according to claim 13, further comprising:
determining, as the coordinates of the end point, coordinates of a middle point between a point closest to and a point farthest from a welding trajectory of the weld bead in the window including the end point.

### Claim 15 (depends on 3)

The repair welding segment detection method according to claim 3, further comprising:
determining, as coordinates of the start point, coordinates of a point closest to a cut surface of the shape mismatch data corresponding to a window including the start point of the defective segment.

### Claim 16 (depends on 3)

The repair welding segment detection method according to claim 3, further comprising:
determining, as coordinates of the end point, coordinates of a point closest to a cut surface of the shape mismatch data corresponding to a next window adjacent to a window including the end point of the defective segment.

### Claim 17 (depends on 1)

The repair welding segment detection method according to claim 1, further comprising:
outputting, as a repair welding condition, the defective segment and information regarding the volume in the defective segment in association with each other.

### Claim 18

A repair welding segment detection device comprising:
an input unit configured to enter input data regarding a weld bead of a workpiece produced by welding:
a determination unit configured to execute inspection determination regarding a shape of the weld bead by using the input data and master data of a non-defective workpiece:
a data generation unit configured to generate, based on an inspection determination result obtained by the determination unit, shape mismatch data obtained by extracting a shape mismatch portion of the weld bead;
a calculation unit configured to divide the shape mismatch data into N, where N is an integer of 2 or more, equally divided windows in a direction perpendicular to a welding direction of the weld bead, set a shift region formed by i, where i is an integer of 1 or more, continuous windows among the respective windows, and separately calculate volumes of (N-i+1) respective shift regions obtained by shifting one by one the i windows forming the shift region in the welding direction; and
a generation unit configured to determine that a shift region having a volume of a predetermined value or more among the calculated volumes of the (N-i+1) shift regions is a defective segment of the weld bead, and generate information regarding the defective segment.
