Configuration of a 3D sensor for safe object tracking

Patent metadata

Field	Value
Publication number	US-12469174-B2
Application number	US-202418407849-A
Country	US
Kind code	B2
Filing date	Jan 9, 2024
Priority date	Jan 10, 2023
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

A method for the automatic configuration of a 3D sensor for safe object tracking in which at least one 3D image is recorded by the 3D sensor, a detection region in which safety related objects can move is fixed within a detection zone of the 3D sensor using the at least one 3D image, and the object tracking is restricted to the detection region. In this respect, picture elements corresponding to a reference area, in particular a floor level, are identified in the at least one 3D image and the detection region is fixed with respect to these picture elements.

First claim

The invention claimed is:

1 . A method for the automatic configuration of a 3D sensor for safe object tracking, comprising:
recording at least one 3D image with the 3D sensor; and fixing a detection region in which safety related objects can move using the at least one 3D image, the detection region being within a detection zone of the 3D sensor, and wherein the safe object tracking is restricted to the detection region; wherein picture elements corresponding to a reference area are identified in the at least one 3D image and the detection region is fixed with respect to the picture elements corresponding to the reference area, wherein the at least one 3D image comprises an image sequence having a plurality of 3D images which are recorded while the safety related objects move through the detection zone, and wherein the picture elements corresponding to the reference area are repeatedly identified using the 3D images from the image sequence to also detect temporarily covered parts of the reference area.

2 . The method in accordance with claim 1 ,
wherein the picture elements correspond to a floor level.

3 . The method in accordance with claim 1 ,
wherein a 3D location of the reference area is determined by means of adapting a level that coincides as best as possible with the picture elements recorded in the at least one 3D image.

4 . The method in accordance with claim 1 ,
wherein the 3D sensor measures its own orientation by means of a location sensor, wherein an expected orientation of the reference area is derived from the measured orientation, and the expected orientation is used to identify the picture elements corresponding to the reference area.

5 . The method in accordance with claim 1 ,
wherein picture elements are determined in which a respectively measured distance changes by more than a noise tolerance over the 3D images of the image sequence, and wherein these picture elements are added to the detection region.

6 . The method in accordance with claim 1 ,
wherein a check is made for adjacent picture elements of the picture elements corresponding to the reference area as to whether measured distance values are compatible with a distance from the reference area and an adjacent picture element to which this applies is added to the detection region.

7 . The method in accordance with claim 6 ,
wherein a distance tolerance is selected as large enough that a step or an inclined ramp is still compatible with the reference area.

8 . The method in accordance with claim 1 ,
wherein the configured detection region is displayed and an acknowledgment is requested before the detection region is used for the safe object tracking.

9 . The method in accordance with claim 1 ,
wherein subsequent to the configuration of the detection region, further 3D images are cyclically recorded and objects therein are tracked to output a respective current position of moving objects in the detection region as a safety function.

10 . The method in accordance with claim 9 ,
wherein the positions of the objects are evaluated to decide whether there is a risk of accident.

11 . The method in accordance with claim 10 ,
wherein the positions of the objects are evaluated while including the previous movement to decide whether there is a risk of accident.

12 . The method in accordance with claim 10 ,
wherein the positions of the objects are evaluated to set a monitored machine into a safe state.

13 . The method in accordance with claim 10 ,
wherein the positions of the objects are evaluated to set a monitored machine into a safe state if a decision is made that there is a risk of accident.

14 . A 3D sensor system, comprising:
a 3D sensor; and a control and evaluation unit configured for safe object tracking, the control and evaluation unit being further configured to: record at least one 3D image with the 3D sensor; and fix a detection region in which safety related objects can move using the at least one 3D image, the detection region being within a detection zone of the 3D sensor, and wherein the safe object tracking is restricted to the detection region, wherein picture elements corresponding to a reference area are identified in the at least one 3D image and the detection region is fixed with respect to the picture elements corresponding to the reference area, wherein the at least one 3D image comprises an image sequence having a plurality of 3D images which are recorded while the safety related objects move through the detection zone, and wherein the picture elements corresponding to the reference area are repeatedly identified using the 3D images from the image sequence to also detect temporarily covered parts of the reference area.

15 . The 3D sensor in accordance with claim 14 ,
wherein the control and evaluation is further configured to carry out the safe object tracking subsequent to the configuration by evaluating cyclically recorded 3D images.

16 . A method for the automatic configuration of a 3D sensor for safe object tracking, comprising:
recording at least one 3D image with the 3D sensor; and fixing a detection region in which safety related objects can move using the at least one 3D image, the detection region being within a detection zone of the 3D sensor, and wherein the safe object tracking is restricted to the detection region; wherein picture elements corresponding to a reference area are identified in the at least one 3D image and the detection region is fixed with respect to the picture elements corresponding to the reference area, wherein the at least one 3D image comprises an image sequence having a plurality of 3D images which are recorded while the safety related objects move through the detection zone, and wherein picture elements are determined in which a respectively measured distance changes by more than a noise tolerance over the 3D images of the image sequence, and wherein these picture elements are added to the detection region.

Assignees

• Sick Ag

Inventors

• Hammes Markus

Classifications

• G06T2207/30244

  Camera pose · CPC title

• G06T2207/20092

  Interactive image processing based on input by user · CPC title

• G06T2207/10021

  Stereoscopic video; Stereoscopic image sequence · CPC title

• G06T2200/04

  involving 3D image data · CPC title

• G06T7/20

  Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) · CPC title