Point cloud analysis device, method, and program

Patent metadata

Field	Value
Publication number	US-12094153-B2
Application number	US-201917608963-A
Country	US
Kind code	B2
Filing date	May 8, 2019
Priority date	May 8, 2019
Publication date	Sep 17, 2024
Grant date	Sep 17, 2024

Abstract

Provided is a point cloud analysis device that curbs a decrease in model estimation accuracy due to a laser measurement point cloud. A clustering unit ( 30 ) clusters a point cloud representing a three-dimensional point on an object obtained by a measurement unit mounted on a moving body and performing measurement while scanning a measurement position, within a scan line, to obtain a point cloud cluster. A central axis direction estimation unit ( 32 ) estimates a central axis direction based on the point cloud cluster. A direction-dependent local effective length estimation unit ( 34 ) estimates a local effective length based on an estimated central axis direction and an interval of scan lines, the local effective length being a length when a length of projection of the point cloud cluster in a central axis direction for each of the point cloud clusters is interpolated by an amount of a loss part of the point cloud.

First claim

The invention claimed is:

1. A point cloud analysis device for estimating presence or absence of a linear structure or/and a region in which the linear structure is present from point cloud data obtained by measuring a real space, the point cloud analysis device comprising:
a linear structure estimator configured to estimate the presence or absence of the linear structure or/and the region in which the linear structure is present from the point cloud data using a property in the real space common to the linear structure,
wherein the property includes a length of the linear structure and a relationship between divided regions when the linear structure is divided into predetermined units,
a three-dimensional data store configured to store a point cloud representing three-dimensional points;
a cluster configured to cluster the point clouds to obtain a point cloud cluster;
a central axis direction estimator configured to estimate a central axis direction based on the point cloud cluster; and
a direction-dependent local effective length estimator configured to estimate a local effective length for each of the point cloud clusters and the estimated central axis direction, the local effective length being a length when a length of projection of a point cloud belonging to the point cloud cluster in the central axis direction is interpolated by an amount of a loss part of the point cloud,
wherein the linear structure estimator uses the local effective length as a length of the linear structure to estimate a model parameter representing a region in which the linear structure is present.

2. The point cloud analysis device according to claim 1 ,
wherein the linear structure estimator further uses a relationship between the linear structure and a ground in the real space, or a relationship between the linear structure and an artificial structure present near the linear structure.

3. The point cloud analysis device according to claim 1 , wherein the linear structure estimator estimates the model parameter representing the region in which the linear structure represented by the point cloud cluster is present based on the local effective length estimated for each of the point cloud clusters and the estimated central axis direction.

4. The point cloud analysis device according to claim 3 , wherein the linear structure estimator uses an evaluation function for evaluating the model parameter, including a penalty term based on a positional relationship between the surrounding structure obtained in advance and the linear structure represented by the point cloud cluster to estimate the model parameter of the linear structure represented by the point cloud cluster.

5. The point cloud analysis device according to claim 3 , wherein the linear structure estimator simultaneously estimates model parameters of the linear structure corresponding to the number of the linear structures using the number of nearby linear structures obtained in advance.

6. The point cloud analysis device according to claim 3 , further comprising:
a model reliability determiner configured to determine model reliability based on a length of a range of a surrounding point cloud consisting of three-dimensional points around the linear structure represented by the estimated model parameter and a length when the linear structure represented by the estimated model parameter is extended to a surrounding structure obtained in advance, and
a remeasurement alerter configured to notify that remeasurement of point cloud data is necessary for a linear structure having the model reliability which is equal to or lower than a threshold value.

7. A point cloud analysis method performed by a point cloud analysis device for estimating presence or absence of a linear structure or/and a region in which the linear structure is present from point cloud data obtained by measuring a real space, the point cloud analysis method comprising:
estimating, by a linear structure estimator, the presence or absence of the linear structure or/and the region in which the linear structure is present from the point cloud data using a property in the real space common to the linear structure,
wherein the property includes a length of the linear structure and a relationship between divided regions when the linear structure is divided into predetermined units,
storing a point cloud representing three-dimensional points;
clustering the point clouds to obtain a point cloud cluster;
estimating a central axis direction based on the point cloud cluster; and
estimating a local effective length for each of the point cloud clusters and the estimated central axis direction, the local effective length being a length when a length of projection of a point cloud belonging to the point cloud cluster in the central axis direction is interpolated by an amount of a loss part of the point cloud,
wherein the linear structure estimator uses the local effective length as a length of the linear structure to estimate a model parameter representing a region in which the linear structure is present.

8. A computer-readable non-transitory recording medium storing computer-executable program instructions for estimating presence or absence of a linear structure or/and a region in which the linear structure is present from point cloud data obtained by measuring a real space, the program instructions when executed by a processor cause a computer system to:
estimate, by a linear structure estimator, the presence or absence of the linear structure or/and the region in which the linear structure is present from the point cloud data using a property in the real space common to the linear structure,
wherein the property includes a length of the linear structure and a relationship between divided regions when the linear structure is divided into predetermined units,
storing, by a three-dimensional data store, a point cloud representing three-dimensional points;
clustering, by a cluster, the point clouds to obtain a point cloud cluster;
estimating, by a central axis direction estimator a central axis direction based on the point cloud cluster; and
estimating, by a direction-dependent local effective length estimator configured to estimate a local effective length for each of the point cloud clusters and the estimated central axis direction, the local effective length being a length when a length of projection of a point cloud belonging to the point cloud cluster in the central axis direction is interpolated by an amount of a loss part of the point cloud,
wherein the linear structure estimator uses the local effective length as a length of the linear structure to estimate a model parameter representing a region in which the linear structure is present.

9. The point cloud analysis method according to claim 7 , wherein the linear structure estimator further uses a relationship between the linear structure and a ground in the real space, or a relationship between the linear structure and an artificial structure present near the linear structure.

10. The point cloud analysis method according to claim 7 , wherein the linear structure estimator estimates the model parameter representing the region in which the linear structure represented by the point cloud cluster is present based on the local effective length estimated for each of the point cloud clusters and the estimated central axis direction.

11. The point cloud analysis method according to claim 10 , wherein the linear structure estimator uses an evaluation function for evaluating the model parameter, including a penalty term based on a positional relationship between the surrounding structure obtained in advance and the linear structure represented by the p

Assignees

• Nippon Telegraph & Telephone

Inventors

• Niigaki Hitoshi
• Yao Yasuhiro
• Inoue Masaaki
• Shimizu Tomoya
• Goto Yukihiro
• Matsuda Shigehiro
• Honda Ryuji
• Oshida Hiroyuki
• Kurata Kana
• Ando Shingo
• Sagata Atsushi

Classifications

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title

• G06T7/0002

  Inspection of images, e.g. flaw detection · CPC title

• G06T2207/30184

  Infrastructure · CPC title

• G06T7/70 Primary

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

• G01C15/00

  Surveying instruments or accessories not provided for in groups G01C1/00 - G01C13/00 · CPC title