Generating 2D mapping using 3D data

Patent metadata

Field	Value
Publication number	US-12100101-B2
Application number	US-202117409918-A
Country	US
Kind code	B2
Filing date	Aug 24, 2021
Priority date	Aug 24, 2021
Publication date	Sep 24, 2024
Grant date	Sep 24, 2024

Abstract

In an aspect for generating a 2-dimensional (2D) building mapping, a computer-implemented method may include one or more processors configured for receiving scanning data in a first file format corresponding to 3-dimensional (3D) aerial Light Detection and Ranging (LiDaR) images. Further, the method may include transforming the first file format to a second file format and transmitting the scanning data to a building semantic segmentation component. Further, the computer-implemented method may include generating semantic segmentation data corresponding to building structures defined by point clusters; generating instance segmentation data corresponding to top surfaces of the building structures; generating building boundary data for each one of the point clusters based on one or more of the semantic segmentation data and the instance segmentation data; and processing the building boundary data to generate a 2D building mapping comprising refined boundary lines for each of the point clusters.

First claim

What is claimed is:

1. A computer-implemented method for generating a 2-dimensional (2D) building mapping, comprising:
identifying, by one or more processors, scanning data in a second file format;
generating, by the one or more processors, semantic segmentation data corresponding to one or more building structures defined by one or more point clusters;
generating, by the one or more processors, instance segmentation data corresponding to one or more top surfaces of the one or more building structures;
filtering, by the one or more processors, one or more noise point clusters from the one or more point clusters, the one or more noise point clusters satisfying a first condition;
generating, by the one or more processors, building boundary data for each one of the one or more point clusters by identifying concave hull points using alpha shape, fitting boundary points into segmented lines using a random sample consensus (RANSAC) algorithm, processing the semantic segmentation data to filter out wall points according to normal vectors, filter out and down-sample roof points, denoise 3D points to keep only 2D information, and processing the instance segmentation data; and
processing, by the one or more processors, the building boundary data to generate a 2D building mapping comprising refined boundary lines for each of the one or more point clusters.

2. The computer-implemented method of claim 1 , further comprising:
receiving, by the one or more processors, scanning data corresponding to 3-dimensional (3D) aerial Light Detection and Ranging (LiDaR) images, wherein the scanning data is in a first file format; and
transforming, by the one or more processors, the scanning data from the first file format to the second file format.

3. The computer-implemented method of claim 1 , wherein generating the semantic segmentation data further comprises:
processing, by the one or more processors, the scanning data in the second file format using a deep learning (DL) model to generate building structure data identifying the one or more building structures based at least on the one or more point clusters.

4. The computer-implemented method of claim 3 , wherein generating the instance segmentation data further comprises:
processing, by the one or more processors, one or more of the scanning data and the semantic segmentation data to generate top surface data identifying the one or more top surfaces of the one or more building structures.

5. The computer-implemented method of claim 4 , further comprising:
identifying, by the one or more processors, one or more individual buildings based on the building structure data and the top surface data.

6. The computer-implemented method of claim 1 , wherein generating the building boundary data further comprises:
discriminating, by the one or more processors, wall data represented in the semantic segmentation data and roof data represented in the instance segmentation data;
separately projecting, by the one or more processors, the wall data and the roof data into a 2D space of the 2D building mapping; and
generating, by the one or more processors, a convex polygon based on the wall data and the roof data in the 2D space.

7. A computer program product for generating a 2D building mapping, the computer program product comprising:
one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to identify scanning data in a second file format;
program instructions to generate semantic segmentation data corresponding to one or more building structures defined by one or more point clusters;
program instructions to generate instance segmentation data corresponding to one or more top surfaces of the one or more building structures;
program instructions to filter one or more noise point clusters from the one or more point clusters, the one or more noise point clusters satisfying a first condition;
program instructions to generate building boundary data for each one of the one or more point clusters by processing the semantic segmentation data to filter out wall points according to normal vectors, filter out and down-sample roof points, denoise 3D points to keep only 2D information, and processing the instance segmentation data; and
program instructions to process the building boundary data to generate a 2D building mapping comprising refined boundary lines for each of the one or more point clusters.

8. The computer program product of claim 7 , further comprising:
program instructions to receive the scanning data corresponding to 3D aerial Light Detection and Ranging (LiDaR) images, wherein the scanning data is in a first file format; and
program instructions to transform the scanning data from the first file format to the second file format.

9. The computer program product of claim 7 , wherein the program instructions to generate the semantic segmentation data further comprise:
program instructions to process the scanning data in the second file format using a DL model to generate building structure data identifying the one or more building structures based at least on the one or more point clusters.

10. The computer program product of claim 9 , wherein the program instructions to generate the instance segmentation data further comprise:
program instructions to process one or more of the scanning data and the semantic segmentation data to generate top surface data identifying the one or more top surfaces of the one or more building structures.

11. The computer program product of claim 10 , further comprising:
program instructions to identify one or more individual buildings based on the building structure data and the top surface data.

12. The computer program product of claim 7 , wherein the program instructions to generate the building boundary data further comprise:
program instructions to discriminate wall data represented in the semantic segmentation data and roof data represented in the instance segmentation data;
program instructions to separately project the wall data and the roof data into a 2D space of the 2D building mapping; and
program instructions to generate a convex polygon based on the wall data and the roof data in the 2D space.

13. A computer system for generating a 2D building mapping, the computer system comprising:
one or more computer processors;
one or more computer readable storage media;
program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to identify scanning data in a second file format;
program instructions to generate semantic segmentation data corresponding to one or more building structures defined by one or more point clusters;
program instructions to generate instance segmentation data corresponding to one or more top surfaces of the one or more building structures;
program instructions to filter one or more noise point clusters from the one or more point clusters, the one or more noise point clusters satisfying a first condition;
program instructions to generate building boundary data for each one of the one or more point clusters by processing the semantic segmentation data to filter out wall points according to normal vectors, filter out and down-sample roof points, denoise 3D points to keep only 2D information, and processing the instance segmentation data; and
program instructions to process the building boundary data to generate a 2D building mapping comprising refined boundary lines for eac

Assignees

• IBM

Inventors

• Min Xu
• Zhang Dan
• Liu Xue Ping
• Ding Yuan Yuan
• Zhang Hong Bing
• Xin Chao
• Li Fan

Classifications

• G06V10/26

  Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion · CPC title

• G06V10/762

  using clustering, e.g. of similar faces in social networks · CPC title

• G06V20/64

  Three-dimensional [3D] objects · CPC title

• G06T17/00 Primary

  Three-dimensional [3D] modelling for computer graphics · CPC title

• G06T3/06

  Topological mapping of higher dimensional structures onto lower dimensional surfaces · CPC title