System and method of automatic room segmentation for two-dimensional laser floorplans

Patent metadata

Field	Value
Publication number	US-11501478-B2
Application number	US-202117325947-A
Country	US
Kind code	B2
Filing date	May 20, 2021
Priority date	Aug 17, 2020
Publication date	Nov 15, 2022
Grant date	Nov 15, 2022

Abstract

A system for generating an automatically segmented and annotated two-dimensional (2D) map of an environment includes processors coupled to a scanner to convert a 2D map from the scanner into a 2D image. Further, a mapping system categorizes a first set of pixels from the image into one of room-inside, room-outside, and noise by applying a trained neural network to the image. The mapping system further categorizes a first subset of pixels from the first set of pixels based on a room type if the first subset of pixels is categorized as room-inside. The mapping system also determines the room type of a second subset of pixels from the first set of pixels based on the first subset of pixels by using a flooding algorithm. The mapping system further annotates a portion of the 2D map to identify the room type based on the pixels corresponding to the portion.

First claim

What is claimed is:

1. A system of generating an automatically segmented and annotated two-dimensional (2D) map of an environment, the system comprising:
a scanner configured to capture a 2D map comprising one or more point clouds comprising coordinate measurements of one or more points from the environment;
one or more processors operably coupled to the scanner, the one or more processors being responsive to executable instructions for converting the 2D map into a 2D image;
a portable computing device having a second image sensor, the portable computing device being coupled for communication to the one or more processors, wherein the one or more processors are responsive to correlate a location captured by a first image from the portable computing device with the location in the 2D map of the environment in response to the first image being acquired by the second image sensor; and
a mapping system configured to:
categorize a first set of pixels from the image into room-inside, room-outside, and noise by applying a trained neural network to the image;
further categorize a first subset of pixels from the first set of pixels based on a room type, the first subset of pixels comprising pixels that are categorized as room-inside;
determine the room type of a second subset of pixels from the first set of pixels based on the first subset of pixels by using a flooding algorithm; and
annotate a portion of the 2D map to identify the room type based on the room type associated with one or more pixels corresponding to the portion.

2. The system of claim 1 , wherein the mapping system is further configured to perform automatic segmentation of the 2D image subsequent to the categorization of the pixels from the image.

3. The system of claim 2 , wherein the automatic segmentation is performed using one or more of morphological segmentation, Voronoi segmentation, and distance-based segmentation.

4. The system of claim 1 , wherein the annotating further comprises determining a label that identifies a type of an object and adding the label to the 2D map proximate to a location of the object.

5. The system of claim 4 , wherein the label of the object is wall, the updating the 2D map includes adding the wall to the 2D map as a geometric element at the location.

6. The system of claim 1 , wherein the scanner is a 2D scanner disposed in a body of a housing, the housing being sized to be carried by a single person during operation, the body having a first plane extending there through.

7. A method for generating a two-dimensional (2D) map of an environment, the method comprising:
capturing, by a scanner, a 2D map comprising one or more point clouds comprising coordinate measurements of one or more points from the environment;
converting the 2D map into a 2D image by one or more processors operably coupled to the scanner, the one or more processors being responsive to executable instructions, wherein the one or more processors correlate a location captured by a first image by a portable computing device with the location in the 2D map of the environment in response to the first image being acquired by the portable computing device;
categorizing a first set of pixels from the image into room-inside, room-outside, and noise by applying a trained neural network to the image;
further categorizing a first subset of pixels from the first set of pixels based on a room type, the first subset of pixels comprising pixels that are categorized as room-inside;
determining the room type of a second subset of pixels from the first set of pixels based on the first subset of pixels by using a flooding algorithm; and
annotating a portion of the 2D map to identify the room type based on the room type associated with one or more pixels corresponding to the portion.

8. The method of claim 7 , further comprising performing automatic segmentation of the 2D image subsequent to the categorization of the pixels from the image.

9. The method of claim 8 , wherein the automatic segmentation is performed using one or more of morphological segmentation, Voronoi segmentation, and distance-based segmentation.

10. The method of claim 7 , wherein the annotating further comprises determining a label that identifies a type of an object and adding the label to the 2D map proximate to the location.

11. The method of claim 10 , wherein the label of the object is wall, the updating the 2D map includes adding the wall to the 2D map as a geometric element at a location of the object.

12. The method of claim 7 , wherein the scanner is a 2D scanner disposed in a body of a housing, the housing being sized to be carried by a single person during operation, the body having a first plane extending there through.

13. A computer program product comprising a memory device with computer executable instructions stored thereon, which when executed by one or more processing units causes the one or more processing units to execute a method for generating a two-dimensional (2D) map of an environment, the method comprising:
receiving a 2D map comprising one or more point clouds comprising coordinate measurements of one or more points from the environment captured by a scanner;
correlating a location captured by a first image by a portable computing device with the location in the 2D map of the environment in response to the first image being acquired by the portable computing device;
converting the 2D map into a 2D image by one or more processors operably coupled to the scanner;
categorizing a first set of pixels from the image into room-inside, room-outside, and noise by applying a trained neural network to the image;
further categorizing a first subset of pixels from the first set of pixels based on a room type, the first subset of pixels comprising pixels that are categorized as room-inside;
determining the room type of a second subset of pixels from the first set of pixels based on the first subset of pixels by using a flooding algorithm; and
annotating a portion of the 2D map to identify the room type based on the room type associated with one or more pixels corresponding to the portion.

14. The computer program product of claim 13 , wherein the method further comprises performing automatic segmentation of the 2D image subsequent to the categorization of the pixels from the image.

15. The computer program product of claim 13 , wherein the annotating further comprises determining a label that identifies a type of an object and adding the label to the 2D map proximate to a location of the object.

16. The computer program product of claim 15 , wherein the label of the object is wall, the updating the 2D map includes adding the wall to the 2D map as a geometric element at the location.

17. The computer program product of claim 13 , wherein the scanner is a 2D scanner disposed in a body of a housing, the housing being sized to be carried by a single person during operation, the body having a first plane extending there through.

18. A system of generating an automatically segmented and annotated two-dimensional (2D) map of an environment, the system comprising:
a scanner configured to capture a 2D map comprising one or more point clouds comprising coordinate measurements of one or more points from the environment;
one or more processors operably coupled to the scanner, the one or more processors being responsive to executable instructions for converting the 2D map into a 2D image;
a mapping system configured to:
categorize a first set of pixels from the image into room-inside, room-outside, and noise by applying a trai

Assignees

• Faro Tech Inc

Inventors

• Brenner Mark
• Frank Aleksej
• Zweigle Oliver
• Ramadneh Ahmad
• Waheed Mufassar

Classifications

• G01S17/42

  Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S17/46) · CPC title

• G06T2207/20156

  Automatic seed setting · CPC title

• G06V10/82

  using neural networks · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title