Monitoring a transverse position of a conveyor belt and its material load by digital image analysis

Patent metadata

Field	Value
Publication number	US-11401113-B2
Application number	US-202016925840-A
Country	US
Kind code	B2
Filing date	Jul 10, 2020
Priority date	Jul 11, 2019
Publication date	Aug 2, 2022
Grant date	Aug 2, 2022

Abstract

A method, system and computer program product are provided for monitoring a transverse position of a conveyor belt or its material load. A processor receives a digital video or digital images capturing movement of the conveyor belt and the material load. The processor segments the images or frames into a group of contiguous pixels representative of the conveyor belt, and the material load, such as by moving object detection, using background segmentation and threshold processing, pixel intensity-based segmentation or image texture-based segmentation. The processor determines a pixel coordinate of the group of contiguous pixels, which is indicative of the transverse position of the conveyor belt or the material load. The processor generates an alarm or report or a data signal, which depends, directly or indirectly, on the determined pixel coordinates.

First claim

The invention claimed is:

1. A method for monitoring a transverse position of a conveyor belt, the method implemented by a processor and comprising the steps of:
(a) receiving at least one digital image of at least a portion of the conveyor belt;
(b) for each digital image:
(i) segmenting the digital image into a group of contiguous pixels representative of at least the portion of the conveyor belt; and
(ii) determining a transverse pixel or actual distance between the portion of the conveyor belt and a stationary reference point, based on the group of contiguous pixels; and

(c) generating either an alarm or report that is audible or visible to a human, or a data signal, wherein generation of the alarm, the report or the data signal, or a characteristic of the generated alarm, report or data signal depends, directly or indirectly, on the determined transverse pixel or actual distance.

2. The method of claim 1 , wherein the receiving step (a) comprises receiving a time sequenced succession of digital images or digital video frames, and the segmenting step (b)(i) comprises performing moving object detection on the frame to differentiate the pixels representative of at least the portion of the conveyor belt from other pixels of the frame, based on apparent motion of at least the portion of the conveyor belt between successive digital images or digital video frames.

3. The method of claim 2 , wherein the moving object detection comprises background subtraction.

4. The method of claim 2 , wherein the segmenting step (b)(i) further comprises performing threshold processing on the digital image or digital video frame after the moving object detection.

5. The method of claim 1 , wherein the segmenting step (b)(i) comprises one or a combination of pixel intensity-based segmentation, or image texture-based segmentation.

6. The method of claim 1 , the portion of conveyor belt comprises an edge of the conveyor belt.

7. The method of claim 1 , wherein the segmenting step (b)(i) comprises one or a combination of pixel intensity-based segmentation, or image texture-based segmentation.

8. A method for monitoring a transverse position of a material load on a conveyor belt, the method implemented by a processor and comprising the steps of:
(a) receiving at least one digital image of the conveyor belt and the material load on the conveyor belt;
(b) for each digital image:
(i) segmenting the digital image into a group of contiguous pixels representative of the material load; and
(ii) determining a pixel coordinate indicative of the transverse position of the material load, based on the group of contiguous pixels; and

(c) generating either an alarm or report that is audible or visible to a human, or a data signal, wherein generation of the alarm, the report or the data signal, or a characteristic of the generated alarm, report or data signal depends, directly or indirectly, on the determined pixel coordinates.

9. The method of claim 8 , wherein the receiving step (a) comprises receiving a time sequenced succession of digital images or digital video frames, and the segmenting step comprises performing moving object detection on the digital image or digital video frames to differentiate the pixels representative of the material load from other pixels of the frame, based on apparent motion of the material load between successive digital images or digital video frames of the digital video.

10. The method of claim 9 , wherein the moving object detection comprises background subtraction.

11. The method of claim 9 , wherein the segmenting step (b)(i) further comprises performing threshold processing on the frame after the moving object detection.

12. The method of claim 8 , wherein step (b) further comprises determining a pixel distance or an actual distance between the material load and an edge of the conveyor belt, based on the determined pixel coordinate.

13. The method of claim 12 , wherein in step (c), generation of the alarm, the report or the data signal, or the characteristic of the generated alarm, report or data signal depends, directly or indirectly, on the pixel distance or the actual distance between the material load and the edge of the conveyor belt.

Assignees

• Syncrude Canada Ltd
• Syncrude Canada Ltd In Trust For The Owners Of

Inventors

• Polak Mark
• Carniato Michael

Classifications

• B05D7/22

  to internal surfaces, e.g. of tubes · CPC title

• B05D7/02

  to macromolecular substances, e.g. rubber (treatment or coating of shaped articles made of macromolecular substances C08J7/00) · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• B65G2203/041

  Camera · CPC title

• B65G43/02 Primary

  detecting dangerous physical condition of load carriers, e.g. for interrupting the drive in the event of overheating · CPC title