Method and device for generating position information of target object

Patent metadata

Field	Value
Publication number	US-10655982-B2
Application number	US-201715784908-A
Country	US
Kind code	B2
Filing date	Oct 16, 2017
Priority date	Jul 21, 2017
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

Disclosed embodiments include a method and a device for generating position information of a target object. In some embodiments, the method comprises: obtaining target point cloud data of the target object, acquired by a target scanner at a target position, and position information of the target scanner, and obtaining reference point cloud data of the target object, acquired by a reference scanner at a reference position, and position information and an Euler angle of the reference scanner; obtaining an Euler angle of the target scanner based on the target point cloud data, the reference point cloud data, and the Euler angle of the reference scanner; and generating the position information of the target object based on the target point cloud data, the position information and the Euler angle of the target scanner, the reference point cloud data, and the position information and the Euler angle of the reference scanner.

First claim

What is claimed is:

1. A method for generating position information of a target object, the method comprising:
obtaining target point cloud data of the target object, acquired by a target scanner at a target position, and position information of the target scanner, and obtaining reference point cloud data of the target object, acquired by a reference scanner at a reference position, and position information and an Euler angle of the reference scanner, wherein the target point cloud data includes offsets between a center of the target scanner and each target point in a target point cloud, and the reference point cloud data includes offsets between a center of the reference scanner and each reference point in a reference point cloud;
obtaining an Euler angle of the target scanner based on the target point cloud data, the reference point cloud data and the Euler angle of the reference scanner; and
generating the position information of the target object based on the target point cloud data, the position information and the Euler angle of the target scanner, the reference point cloud data, and the position information and the Euler angle of the reference scanner.

2. The method according to claim 1 , wherein the obtaining the Euler angle of the target scanner based on the target point cloud data, the reference point cloud data and the Euler angle of the reference scanner comprises:
partitioning the target point cloud and the reference point cloud into a plurality of cubes with a preset size respectively;
executing following determination steps: for each partitioned cube, fitting a plane by using points in the cube, determining whether standard deviations of the distances from the points in the cube to the fitted plane is smaller than a standard deviation threshold, if the standard deviations are smaller than the standard deviation threshold, using the fitted plane as a first plane, determining whether the cube meets a preset condition at the same time: a number of points in the cube is smaller than a preset number of points and side lengths of the cube are smaller than a preset side length, and if the preset condition is met at the same time, using the first plane as a second plane;
obtaining centroid coordinates and normal vectors of obtained second planes, wherein the obtained second planes comprise second planes corresponding to the target point cloud and second planes corresponding to the reference point cloud;
generating an optimal Euler angle difference between the target scanner and the reference scanner based on the centroid coordinate and the normal vectors of the obtained second planes; and
generating the Euler angle of the target scanner based on the Euler angle of the reference scanner, and the optimal Euler angle difference between the target scanner and the reference scanner.

3. The method according to claim 2 , wherein the obtaining the Euler angle of the target scanner based on the target point cloud data, the reference point cloud data and the Euler angle of the reference scanner further comprises:
in response to determining that the standard deviations of the distances from the points in the cube to the fitted plane are not smaller than the standard deviation threshold, or in response to determining that the cube does not meet the preset condition, partitioning the cube into a preset number of cubes with an identical size, and continuing executing the determination steps.

4. The method according to claim 2 , wherein the generating an optimal Euler angle difference between the target scanner and the reference scanner based on the centroid coordinates and the normal vectors of the obtained second planes comprises:
constructing a sum function of distances between the second planes corresponding to the target point cloud and the second planes corresponding to the reference point cloud by use of the centroid coordinates and the normal vectors of the obtained second planes, and the Euler angle difference between the target scanner and the reference scanner; and
determining an Euler angle difference minimizing a value of the sum function of the distances, and using the Euler angle difference as the optimal Euler angle difference.

5. The method according to claim 1 , wherein the generating the position information of the target object, based on the target point cloud data, the position information and the Euler angle of the target scanner, the reference point cloud data, and the position information and the Euler angle of the reference scanner comprises:
converting the target point cloud data into position information of target points in the target point cloud using the position information and the Euler angle of the target scanner, and converting the reference point cloud data into position information of reference points in the reference point cloud using the position information and the Euler of the reference scanner;
rendering a point cloud of the target object based on the position information of the target points in the target point cloud, and the position information of the reference points in the reference point cloud;
fitting a plane of the target object using the point cloud of the target object; and
obtaining a centroid coordinate of the plane of the target object, and using the centroid coordinate as the position information of the target object.

6. The method according to claim 2 , wherein a method adopted for fitting the plane comprises a random sample consensus algorithm or a least square method.

7. The method according to claim 1 , further comprising:
searching a position indicated by the position information of the target object on a pre-generated map, and marking the target object at the searched position.

8. A device for generating position information of a target object, the device comprising:
at least one processor; and
a memory storing instructions, which when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:
obtaining target point cloud data of the target object, acquired by a target scanner at a target position, and position information of the target scanner, and obtaining reference point cloud data of the target object, acquired by a reference scanner at a reference position, and position information and an Euler angle of the reference scanner, wherein the target point cloud data includes offsets between a center of the target scanner and each target point in a target point cloud, and the reference point cloud data includes offsets between a center of the reference scanner and each reference point in a reference point cloud;
obtaining an Euler angle of the target scanner based on the target point cloud data, the reference point cloud data and the Euler angle of the reference scanner; and
generating the position information of the target object based on the target point cloud data, the position information and the Euler angle of the target scanner, the reference point cloud data, and the position information and the Euler angle of the reference scanner.

9. The device according to claim 8 , wherein the obtaining the Euler angle of the target scanner based on the target point cloud data, the reference point cloud data and the Euler angle of the reference scanner comprises:
partitioning the target point cloud and the reference point cloud into a plurality of cubes with a preset size respectively;
executing following determination steps: for each partitioned cube, fitting a plane by using points in the cube, determining whether the standard deviations of the distances from the points in the cube to the fitted plane are smaller than a standard deviation threshold, if the standard deviation is smaller than the standard deviat

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Peng Weilin
• Yu Li
• Xu Shengpan
• Tian Hailong
• Song Shiyu
• Dong Fangfang

Classifications

• G06T7/70

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

• G01C3/08

  Use of electric radiation detectors · CPC title

• G06T7/50

  Depth or shape recovery · CPC title

• G06T2207/10028

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

• G06T7/73 Primary

  using feature-based methods · CPC title