IP Library › Granted Patent US 12,008,784
Granted Patent B2
US 12,008,784 · App. 17/171,100 · Granted Jun 11, 2024

Systems and methods for image-based electrical connector assembly detection

Inventors: Raj Sohmshetty (Canton, MI); Vikas Rajendra (Novi, MI)
Assignee: Ford Global Technologies, LLC
G06T7/73G06T7/0002G06T7/11G06T7/13G06T2207/20081G06T2207/20084G06T2207/30108
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,008,784
App. No.
17/171,100
Granted
Jun 11, 2024
Kind
B2
Abstract

A method includes defining, in an image, bounding boxes about an electrical connector assembly, identifying an edge of each of the bounding boxes, and determining one or more metrics based on the edge of each the bounding boxes, where the one or more metrics indicate a positional relationship between the edges of the bounding boxes. The method includes determining a state of the electrical connector assembly based on the one or more metrics and transmitting a notification based on the state.

Claims (34)

1. A method comprising:

defining, in an image and by a convolutional neural network, bounding boxes about an electrical connector assembly, wherein the bounding boxes include a first component bounding box and a second component bounding box, wherein the first component bounding box is associated with a first component of the electrical connector assembly, and wherein the second component bounding box is associated with a second component of the electrical connector assembly;

identifying an edge of each of the bounding boxes;

determining a distance and an angle between on the edge of each of the bounding boxes, wherein the distance and the angle between the edge of each of the bounding boxes define a positional relationship between the edges of the bounding boxes;

determining a state of the electrical connector assembly based on the positional relationship; and

transmitting a notification based on the state.

2. The method of claim 1 , wherein the first component bounding box and the second component bounding boxes are rotated bounding boxes.

3. The method of claim 1 , wherein the edge of each of the bounding boxes includes a first adjacent edge of the first component bounding box and a second adjacent edge of the second component bounding box.

4. The method of claim 3 , wherein the distance is a distance between the first adjacent edge and the second adjacent edge.

5. The method of claim 4 , wherein the state is a connected state in response to the distance between the first adjacent edge and the second adjacent edge being less than a threshold distance.

6. The method of claim 4 , wherein the state is an unconnected state in response to the distance between the first adjacent edge and the second adjacent edge being greater than a threshold distance.

7. The method of claim 3 , wherein the angle is defined between the first adjacent edge and the second adjacent edge.

8. The method of claim 7 , wherein the state is a connected state in response to the angle defined between the first adjacent edge and the second adjacent edge being less than a threshold angle.

9. The method of claim 7 , wherein the state is an unconnected state in response to the angle defined between the first adjacent edge and the second adjacent edge being greater than a threshold angle.

10. A system comprising:

a processor; and

a nontransitory computer-readable medium including instructions that are executable by the processor, wherein the instructions include:

defining, in an image and by a convolutional neural network, bounding boxes about an electrical connector assembly, wherein the bounding boxes include a first component bounding box and a second component bounding box, wherein the first component bounding box is associated with a first component of the electrical connector assembly, and wherein the second component bounding box is associated with a second component of the electrical connector assembly;

identifying an edge of each bounding box;

determining a distance and an angle between the edge of each bounding box, wherein the distance and the angle between each edge of each bounding box define a positional relationship between each edge of each bounding box;

determining a connection state of the electrical connector assembly based on the positional relationship; and

transmitting a notification based on the connection state.

11. The system of claim 10 , wherein the edge of each of the bounding boxes includes a first adjacent edge of the first component bounding box and a second adjacent edge of the second component bounding box.

12. The system of claim 11 , wherein the distance between each edge is further based on a distance between the first adjacent edge and the second adjacent edge.

13. The system of claim 11 , wherein the angle between each edge is further based on an angle defined between the first adjacent edge and the second adjacent edge.

14. A method comprising:

defining, in an image and by a convolutional neural network, a first component bounding box and a second component bounding box about an electrical connector assembly, wherein the first component bounding box is associated with a first component of the electrical connector assembly, and the second component bounding box is associated with a second component of the electrical connector assembly;

identifying a first adjacent edge of the first component bounding box and a second adjacent edge of the second component bounding box;

determining a distance and an angle between the first adjacent edge and the second adjacent edge, wherein the distance and the angle between the first adjacent edge and the second adjacent edge define a positional relationship between the first adjacent edge and the second adjacent edge;

determining a state of the electrical connector assembly based on the positional relationship; and

transmitting a notification based on the state.

15. The method of claim 14 , wherein the state is a connected state in response to the distance between the first adjacent edge and the second adjacent edge being less than a threshold distance, the angle defined between the first adjacent edge and the second adjacent edge being less than a threshold angle, or a combination thereof.

16. The method of claim 14 , wherein the state is an unconnected state in response to the distance between the first adjacent edge and the second adjacent edge being greater than a threshold distance, the angle defined between the first adjacent edge and the second adjacent edge being greater than a threshold angle, or a combination thereof.

17. The method of claim 14 , wherein the neutral network is a region-based convolutional neural network.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 10, 2021
From: SOHMSHETTY, RAJ; RAJENDRA, VIKAS
To: FORD GLOBAL TECHNOLOGIES, LLC
Reel/Frame 056187/0167 →
Continuity (1)
Related Publication 20220254055A1 · Aug 11, 2022