CN113219489A

CN113219489A - Method and device for determining point pair of multi-line laser, computer equipment and storage medium

Info

Publication number: CN113219489A
Application number: CN202110524505.8A
Authority: CN
Inventors: 杨琪; 冉国江; 翟剑坤; 黄国明; 刘国帅; 王佳轩
Original assignee: Shenzhen Shuma Electronic Technology Co ltd
Current assignee: Shenzhen Shuma Electronic Technology Co ltd
Priority date: 2021-05-13
Filing date: 2021-05-13
Publication date: 2021-08-06
Anticipated expiration: 2041-05-13
Also published as: CN113219489B

Abstract

The application relates to a method, a device, a computer device and a storage medium for determining the point pairs of a multi-line laser. The method comprises the following steps: acquiring a first laser line set in a first laser image and a second laser line set in a second laser image; the first laser line set comprises at least one first laser line, and the second laser line set comprises at least one second laser line; the first laser image and the second laser image are obtained by shooting the same scene; acquiring a first laser point set positioned on a first laser line and acquiring a second laser point set positioned on a second laser line; according to the laser transformation relation set, performing multiple uniqueness judgment on the first laser point set and the second laser point set to obtain a target point pair; and constructing according to the target point pair to obtain the three-dimensional point cloud. The method can improve the accuracy of the point pairs.

Description

Method and device for determining point pair of multi-line laser, computer equipment and storage medium

Technical Field

The present invention relates to the field of laser scanning technologies, and in particular, to a method and an apparatus for determining a point pair of a multi-line laser, a computer device, and a computer-readable storage medium.

Background

In laser real-time three-dimensional scanning, the determination of the corresponding point of the laser is very important. The traditional method for determining the multi-line laser point pairs is to extract left and right characteristic points of a laser image and obtain matched point pairs in a characteristic point matching mode. In the traditional multi-line laser point pair determination, the problem of inaccurate point pair exists.

Disclosure of Invention

In view of the above, it is necessary to provide a method, an apparatus, a computer device and a storage medium for determining a point pair of a multiline laser.

A method of point pair determination for a multiline laser, the method comprising:

acquiring a first laser line set in a first laser image and a second laser line set in a second laser image; the first laser line set comprises at least one first laser line, and the second laser line set comprises at least one second laser line; the first laser image and the second laser image are obtained by shooting the same scene;

acquiring a first laser point set positioned on a first laser line and acquiring a second laser point set positioned on a second laser line;

according to the laser transformation relation set, performing multiple uniqueness judgment on the first laser point set and the second laser point set to obtain a target point pair;

and constructing according to the target point pair to obtain the three-dimensional point cloud.

A device for point pair determination of multiline laser light, the device comprising:

the laser line acquisition module is used for acquiring a first laser line set in the first laser image and a second laser line set in the second laser image; the first laser line set comprises at least one first laser line, and the second laser line set comprises at least one second laser line; the first laser image and the second laser image are obtained by shooting the same scene;

the laser point acquisition module is used for acquiring a first laser point set positioned on a first laser line and acquiring a second laser point set positioned on a second laser line;

the multiple uniqueness judgment module is used for performing multiple uniqueness judgment on the first laser point set and the second laser point set according to a laser transformation relation set to obtain a target point pair;

and the construction module is used for constructing according to the target point pairs to obtain the three-dimensional point cloud.

A computer device comprising a memory storing a computer program and a processor implementing method embodiments of a method of point pair determination for multiline laser light when the computer program is executed.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out method embodiments of a method for point pair determination of a multiline laser.

The method, the device, the computer equipment and the storage medium for determining the point pairs of the multi-line laser can acquire a first laser line set in a first laser image and a second laser line set in a second laser image, acquire a first laser point set on the first laser line and acquire a second laser point set on the second laser line, and then can acquire enough points; according to the laser transformation relation set, multiple uniqueness judgment is carried out on the first laser point set and the second laser point set to obtain target point pairs, the target point pairs are constructed according to the target point pairs to obtain three-dimensional point clouds, unique corresponding point pairs can be obtained, stable high speed can be kept, more corresponding points are reserved, and the accuracy of the obtained target point pairs is improved, so that the reconstructed three-dimensional point clouds are more accurate, good in real-time performance and wide in practicability.

Drawings

FIG. 1 is a diagram of an exemplary embodiment of a multi-line laser scanning application;

FIG. 2 is a schematic flow chart of a method for determining a point pair for a multiline laser in one embodiment;

FIG. 3 is a schematic diagram of a first laser map in one embodiment;

FIG. 4 is a flow diagram illustrating one embodiment of obtaining one-to-one candidate point pairs;

FIG. 5 is a block diagram of a point pair determining apparatus for a multiline laser in one embodiment;

FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.

It should be noted that all directional indicators (such as upper, lower, left, right, front and rear … …) in the embodiment of the present invention are only used to explain the relative position relationship between the components, the movement situation, etc. in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indicator is changed accordingly, and the connection may be a direct connection or an indirect connection.

In addition, the descriptions related to "first", "second", etc. in the present invention are only for descriptive purposes and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, technical solutions between various embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination should not be considered to exist, and is not within the protection scope of the present invention.

In one embodiment, as shown in FIG. 1, a diagram of an environment in which the multi-line laser scanning is applied in one embodiment is provided. Included in fig. 1 are a first camera device 110, a second camera device 120, a laser transmitter 130, an object 140, and a computer device 150. The first image pickup device 110, the second image pickup device 120, and the laser transmitter 130 are respectively connected to a computer device 150. The first image capturing apparatus 110 and the second image capturing apparatus 120 are not the same image capturing apparatus, and the first image capturing apparatus 110 and the second image capturing apparatus 120 are located at different positions. The laser transmitter 130 transmits a plurality of laser beams to the object 140, scans the entire object 140, photographs the object 140 while scanning, photographs through the first photographing apparatus 110 to obtain a first laser beam pattern, and photographs through the second photographing apparatus 120 to obtain a second laser beam pattern. The computer device 150 acquires the first laser image and the second laser image in real time, and performs three-dimensional point cloud construction based on the first laser image and the second laser image. It is to be understood that the number of image pickup apparatuses may be configured as required. After the multi-line laser emitted by the laser emitter is reflected by an object, the number of laser lines in the first laser diagram may be different from the number of laser lines in the second laser diagram, and the computer device does not know which laser line applies which laser transformation relation, nor does it know which laser line in the first laser diagram corresponds to which laser line in the second laser diagram.

It will be appreciated that although the laser transmitter emits multiple laser beams, there may be only one visible laser line in the first laser beam pattern and the second laser beam pattern. A laser map containing only one laser line is also applicable to the point pair determination method of the multiline laser in the embodiment of the present application.

In one embodiment, as shown in fig. 2, a flow chart of a method for determining a point pair of a multi-line laser in one embodiment includes steps 202 to 208:

step 202, acquiring a first laser line set in a first laser map and a second laser line set in a second laser map; the first laser line set comprises at least one first laser line, and the second laser line set comprises at least one second laser line; the first laser image and the second laser image are obtained by shooting the same scene.

The first laser image and the second laser image are laser images at different positions of the same scene. The first laser pattern and the second laser pattern are not the same image. The first laser pattern may include a plurality of first laser lines and the second laser pattern may include a plurality of second laser lines.

Specifically, the computer device acquires a first laser image in real time from the first image pickup device, and acquires a second laser image in real time from the second image pickup device. The computer device can extract the first laser line in the first laser map using a Steger algorithm. The computer device can extract the second laser line in the second laser map using a Steger algorithm. Fig. 3 is a schematic diagram of a first laser map in one embodiment. Included in fig. 3 is a first laser line 310.

The laser line may be a laser centerline. The laser center line is a line of the laser center. I.e., the first laser line may be a first laser centerline and the second laser line may be a second laser centerline. And the laser line may be a straight line or a curved line.

Step 204, a first set of laser spots on the first laser line is acquired, and a second set of laser spots on the second laser line is acquired.

The first laser point set may include each first laser point located on the first laser line, or may be a laser point corresponding to a first coordinate value located on the first laser line being a positive integer value. The second set of laser points may include each second laser point located on the second laser line, or may be laser points corresponding to the first coordinate value located on the second laser line being a positive integer value. In a computer device, the laser points may be represented by laser point coordinates.

Specifically, the computer device acquires a first set of laser spots located on a first laser centerline and acquires a second set of laser spots located on a second laser line.

And step 206, performing multiple uniqueness judgment on the first laser point set and the second laser point set according to the laser transformation relation set to obtain a target point pair.

And before three-dimensional scanning, the laser transformation relation set is calibrated. The laser transformation relation refers to the transformation relation corresponding to the laser line. The laser transformation relation set comprises the laser transformation relation corresponding to each laser line. Namely, the set of laser transformation relations used in the known laser map, if necessary, the laser transformation relation H₁Laser transformation relation H₂Laser transformation relation H₃And the like are not limited thereto.

The laser conversion relation comprises at least one of a first laser conversion relation and a second laser relation. The first laser transformation relationship is used to project points in the first laser map into the second laser map. The second laser transformation relationship is used to project points in the second laser map into the first laser map.

The multiple uniqueness judgment means two or more uniqueness judgments. The multiple uniqueness judgment includes corresponding point uniqueness judgment and laser conversion relation uniqueness judgment. The corresponding point uniqueness judgment is based on the principle that one point in the first laser diagram has one and only one corresponding point in the second laser diagram, and similarly, one point in the second laser diagram has one and only one corresponding point in the first laser diagram. The principle according to which the uniqueness of the laser transformation relation is determined comprises that the laser transformation relations corresponding to each laser line in the laser graph are different, and the laser transformation relation of one laser line should be unique.

The target point pair refers to a point pair for constructing a three-dimensional point cloud. One point of the pair of target points is located in the first laser map and the other point is located in the second laser map. And constructing the three-dimensional point cloud through the target point pairs. The laser spots in the target spot pair are unique and non-repetitive.

Specifically, the computer device performs multiple uniqueness judgment on the first laser point set and the second laser point set according to a laser transformation relation set required by conversion of the first laser image and the second laser image, and obtains a target point pair. The computer equipment can also obtain the respective corresponding target laser transformation relation of the target point pairs.

And 208, constructing according to the target point pairs to obtain the three-dimensional point cloud.

Specifically, the computer equipment is constructed according to target point pairs based on a binocular vision principle to obtain three-dimensional point cloud.

In the method for determining a pair of multi-line laser points in this embodiment, a first laser line set in a first laser diagram and a second laser line set in a second laser diagram are obtained, a first laser point set located on a first laser line is obtained, and a second laser point set located on a second laser line is obtained, so that enough points can be obtained; according to the laser transformation relation set, multiple uniqueness judgment is carried out on the first laser point set and the second laser point set to obtain target point pairs, the target point pairs are constructed according to the target point pairs to obtain three-dimensional point clouds, unique corresponding point pairs can be obtained, stable high speed can be kept, more corresponding points are reserved, and the accuracy of the obtained target point pairs is improved, so that the reconstructed three-dimensional point clouds are more accurate, good in real-time performance and wide in practicability.

In one embodiment, performing multiple uniqueness decisions on the first set of laser points and the second set of laser points according to a set of laser transformation relationships to obtain pairs of target points comprises:

according to the laser transformation relation set, corresponding point uniqueness judgment is carried out on the first laser point set and the second laser point set, and one-to-one candidate point pairs corresponding to the first laser image and the second laser image are obtained;

and according to the laser transformation relation corresponding to the one-to-one candidate point pair, performing laser transformation relation uniqueness judgment on the one-to-one point pair to obtain a target point pair.

The corresponding point uniqueness judgment comprises a first candidate point pair judgment and a second candidate point pair judgment, wherein the first candidate point pair judgment is used for judging the candidate laser point in the second laser map corresponding to the first laser point so as to form the first candidate point pair. The second candidate point pair determination is used for determining a candidate laser point in the first laser map corresponding to the second laser point, thereby forming a second candidate point pair.

A one-to-one candidate point pair is one and only one pair stress spot in the first laser map and the second laser map. And a one-to-one pair of candidate points corresponds to a laser transform relationship.

Specifically, the computer device respectively transforms a first laser point set and a second laser point set according to the laser transformation relation set, and determines a first candidate point pair corresponding to the first laser point and a second candidate point pair corresponding to the second laser point; and screening the first candidate point pair and the second candidate point pair to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image.

And according to the laser transformation relation corresponding to the one-to-one candidate point pair, the computer equipment performs laser transformation relation uniqueness judgment on the one-to-one candidate point pair to obtain a target point pair.

Optionally, the computer device screens out a target point pair with a unique laser transformation relation under the condition that the first coordinate values are the same from the one-to-one candidate point pairs according to the laser transformation relation corresponding to the one-to-one candidate point pair.

Optionally, the computer device screens out a target point pair corresponding to the unique corresponding laser transformation relation on the same laser line from the one-to-one candidate point pair according to the laser transformation relation corresponding to the one-to-one candidate point pair.

In this embodiment, based on the principle that one point in the first laser map has only one corresponding point in the second laser map, a one-to-one candidate point pair corresponding to the first laser map and the second laser map can be obtained by determining the uniqueness of the corresponding point; and based on the uniqueness of the laser transformation relation, the uniqueness judgment of the laser transformation relation is carried out on the one-to-one candidate point pair, more error points are deleted, more point pairs meeting the construction conditions can be screened out, and the processing efficiency is high.

In one embodiment, the set of laser transformation relationships includes a first set of laser transformation relationships and a second set of laser transformation relationships. As shown in fig. 4, a schematic flow chart of obtaining one-to-one candidate point pairs in one embodiment, where the method performs corresponding point uniqueness determination on a first laser point set and a second laser point set according to a set of laser transformation relations to obtain one-to-one candidate point pairs corresponding to a first laser map and a second laser map, and includes steps 402 to 410:

step 402, for a first laser point in the first laser point set, transforming the first laser point according to the first laser transformation relation set to obtain a transformed first laser point.

The laser transformation relation set comprises a first laser transformation relation set and a second laser transformation relation set. The number of transformation relations in the first set of laser transformation relations and the second set of laser transformation relations matches the number of laser lines in the laser map. The first laser transformation relationship refers to the transformation relationship required to map a laser line in a first laser map to a second laser map. The second laser transformation relationship refers to the transformation relationship required to map the laser lines in the second laser map to the first laser map. Each first laser line in the first laser graph corresponds to a different first laser transformation relation. For example, laser line 1 corresponds to the first laser transformation relationship H1 and laser line 2 corresponds to the second laser transformation relationship H2. Similarly, each second laser line in the second laser map corresponds to a different second laser transformation relationship.

Specifically, for each first laser spot in the first laser spot set, the first laser spots are respectively transformed according to each first laser transformation relation in the first laser transformation set, and the transformed first laser spots are respectively obtained. For example, the first set of laser transform relationships H ═ H₁，H₂，H₃……H_nThen the first laser point a follows H₁Once transformed in accordance with H₂Once transformed in accordance with H₃Once … … according to H_nAnd the conversion is performed once.

In step 404, when the transformed first laser point matches with a second laser point in the second laser point set, the first laser point and the matching second laser point are taken as a first candidate point pair.

Wherein one point of the first candidate pair is a first laser point and the other point is a second laser point matching the first laser point. The number of first candidate point pairs corresponding to one first laser point may be plural.

Specifically, the computer device matches the transformed first laser point with a second laser point in the second set of laser points. The matching mode may specifically be that the computer device performs difference calculation on the transformed second coordinate value of the first laser point and the transformed second coordinate value of the second laser point to obtain a difference value; and when the difference value meets the difference threshold condition, determining that the first laser point is matched with the second laser point. In the case where the transformed first laser point matches a second laser point in the second set of laser points, the computer device takes the first laser point and the matching second laser point as a first candidate point pair.

And 406, for a second laser point in the second laser point set, transforming the second laser point according to the second laser transformation relation set to obtain a transformed second laser point.

In step 408, when the transformed second laser point matches the first laser point in the first laser point set, the second laser point and the matching first laser point are used as a second candidate point pair.

And step 410, screening the first candidate point pair and the second candidate point pair to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image.

The laser points in the one-to-one candidate point pair corresponding to the first laser map and the second laser map are only one, namely, the first candidate point pair corresponding to the first laser point is only one, the second candidate point pair corresponding to the first laser point is only one, and the first candidate point pair corresponding to the first laser point is matched with the second candidate point pair.

Specifically, the computer device deletes the candidate point pairs including the laser point many-to-one relationship and the laser point one-to-many relationship, retains the candidate point pairs of the mutually corresponding laser point one-to-one relationship in the first candidate point pair and the second candidate point pair, and obtains the mutually corresponding one-to-one candidate point pair of the first laser map and the second laser map.

For example, the first laser spot L₁The corresponding first candidate point pair has (L)₁，R₁) And (L)₁，R₂) The corresponding second candidate point pair has (R)₃，L₁) And (R)₂，L₁) Then the first laser spot L₁And deleting all the corresponding candidate point pairs. First laser spot L₃The corresponding first candidate point pair is only (L)₃，R₃) The corresponding second candidate point pair is only (R)₃，L₃) Then, a one-to-one candidate point pair (L) where the first laser points B correspond to each other is obtained₃，R₃)。

In this embodiment, the first laser point is transformed according to the first laser transformation relationship to obtain a transformed first laser point, and when the transformed first laser point is matched with a second laser point in the second laser point set, the first laser point and the matched second laser point are used as a first candidate point pair, so as to obtain a one-to-many or one-to-one point pair of the first laser map to the second laser map; converting the second laser points according to the second laser conversion relation set to obtain converted second laser points, and taking the second laser points and the matched first laser points as second candidate point pairs under the condition that the converted second laser points are matched with the first laser points in the first laser point set, so as to obtain one-to-many or one-to-one point pairs of the second laser image to the first laser image; then, since the first laser point has only one point corresponding to the second laser map, the first candidate point pair and the second candidate point pair need to be screened, so as to obtain the one-to-one candidate point pair corresponding to the first laser map and the second laser map, and at this time, more points are retained, and the accuracy of point cloud construction can also be improved.

In one embodiment, the first laser map and the second laser map are both maps obtained after binocular vision correction;

for a first laser point in the first laser point set, transforming the first laser point according to the first laser transformation relation set to obtain a transformed first laser point, including:

and (a1) for the first laser points with the same first coordinate value in the first laser point set, respectively transforming the first laser points with the same first coordinate value point according to the first transformation relation in the first transformation relation set to obtain transformed first laser points.

And the binocular vision correction is used for enabling the laser points matched in the first laser image and the second laser image to be on the same first coordinate value. Binocular vision correction is an important step in the calibration process of the camera equipment. The binocular vision correction includes binocular epipolar correction.

The first coordinate value and the second coordinate value are not the same coordinate value. For example, when the first coordinate value is a y coordinate value, the second coordinate value is an x coordinate value. It is understood that the first coordinate value and the second coordinate value may be coordinate values in other two-dimensional coordinate systems.

Specifically, for the first laser points in the first laser point set and having the same first coordinate value, the computer device transforms the first laser points having the same first coordinate value according to the first transformation relation in the first transformation relation set, and obtains the transformed first laser points.

When the transformed first laser point is matched with a second laser point in the second laser point set, taking the first laser point and the matched second laser point as a first candidate point pair, including:

and (a2) determining a corresponding first difference value according to the transformed second coordinate value corresponding to the first laser point and the second coordinate value corresponding to the second laser point located on the same first coordinate value.

The first difference value is used for representing the difference between the second coordinate value of the transformed first laser point and the second coordinate value corresponding to the second laser point. The difference value may be a difference value, a proportional value, or the like.

Specifically, the computer device calculates a difference between a second coordinate value corresponding to the transformed first laser point and a second coordinate value corresponding to a second laser point located on the same first coordinate value, and obtains a corresponding difference value.

And (a3) when the first difference value satisfies the difference threshold condition, setting the first laser point and the second laser point satisfying the difference threshold condition as the first candidate point pair.

The difference threshold condition may be that the difference value is smaller than a preset difference value, or that the difference value is smaller than or equal to the preset difference value. The difference threshold condition may be set as desired.

Specifically, in a case where the disparity value satisfies the disparity threshold condition, the computer device takes the first laser point and the second laser point pair satisfying the disparity threshold condition as the first candidate point pair.

For example, the first laser points with the same first coordinate value may be the first laser points with the same y value, i.e. the first laser points on the same row, and then the first laser point a of the row is set H ═ H according to the first laser transform relationship set₁，H₂，H₃……H_nObtaining a transformed first laser point A'₁、A’₂、A’₃……A’_n. A is the second laser point with the same first coordinate value₁、a₂、a₃……a_mTransformed first laser point A'₁For example, then calculate A'₁And a₁Value of difference between, A'₁And a₂Value of difference between, A'₁And a₃Value of difference between … … A'₁And a_mA 'to'₁And a second laser spot a satisfying a difference threshold condition₃As a first candidate point pair.

For a second laser point in the second laser point set, transforming the second laser point according to a second laser transformation relation set to obtain a transformed second laser point, including:

and (a4) for the second laser points with the same first coordinate value in the second laser point set, respectively transforming the second laser points with the same first coordinate value according to a second transformation relation in the second transformation relation set to obtain transformed second laser points.

When the transformed second laser point is matched with the first laser point in the first laser point set, the step of taking the second laser point and the matched first laser point as a second candidate point pair includes:

and (a5) calculating a corresponding second difference value according to the first coordinate value corresponding to the transformed second laser point and the first coordinate value corresponding to the first laser point located at the same first coordinate value.

And (a6) when the second difference value satisfies the difference threshold condition, setting the second laser point and the first laser point satisfying the difference threshold condition as a second candidate point pair.

In this embodiment, laser points of the same first coordinate value are transformed according to the transformation relation in the transformation relation set to obtain transformed laser points, and the difference between the transformed laser points and the laser points on another graph of the same first coordinate value is calculated.

In one embodiment, the laser transformation relationship uniqueness judgment is performed on the one-to-one point pair according to the laser transformation relationship corresponding to the one-to-one candidate point pair, and the target point pair is obtained, including:

screening out reference point pairs with non-repeated laser transformation relations under the condition that the first coordinate values are the same from the one-to-one candidate point pairs according to the laser transformation relations corresponding to the one-to-one candidate point pairs;

and screening out the target point pair corresponding to the unique corresponding laser transformation relation on the same laser line from the reference point pair according to the corresponding laser transformation relation of the reference point pair.

The same laser line can be a first laser line, a second laser line, a first laser line and a second laser line. The specific method can be adjusted according to the actual configuration.

Specifically, in the multi-line laser map, the laser points having the same first coordinate value cannot be subjected to the situation where the stress light point is obtained by the same laser line variation relationship. Taking the first coordinate value as the y-axis coordinate value as an example, the extracted laser points in the same row belong to different laser lines, so the laser line transformation relationship is not repeated. And screening out reference point pairs which do not have repeated laser transformation relations under the condition that the first coordinate values are the same from the one-to-one candidate point pairs according to the laser transformation relations corresponding to the one-to-one candidate point pairs by the computer equipment.

In the multiline laser map, the laser mapping relationship for each laser line should be unique. The only references herein mean that the laser transformation relationship required to map a laser line in a first laser map into a second laser map is unique, or that a laser line in a second laser map into a first laser map is unique. Then, the computer device screens out the unique corresponding laser transformation relation on the same laser line from the reference point pair according to the corresponding laser transformation relation of the reference point pair, and acquires the laser point pair corresponding to the unique corresponding laser transformation relation to obtain the target point pair.

In this embodiment, since the laser point with the same first coordinate value cannot obtain the stress light point through the same laser line variation relationship, a reference point pair with a non-repeated laser conversion relationship under the condition that the first coordinate value is the same is screened out from the one-to-one candidate point pair according to the laser conversion relationship corresponding to the one-to-one candidate point pair, so that some points unsuitable for point cloud construction can be excluded; because the laser transformation relation of the same laser line is unique, the target point pair corresponding to the unique corresponding laser transformation relation on the same laser line is screened out according to the laser transformation relation corresponding to the reference point pair, and the point suitable for point cloud construction is obtained through another round of screening, so that the accuracy of point cloud construction is improved.

In one embodiment, screening out a reference point pair whose laser transformation relationship does not overlap when the first coordinate values are the same from the one-to-one candidate point pairs according to the laser transformation relationship corresponding to the one-to-one candidate point pair includes:

according to the laser transformation relation corresponding to the one-to-one candidate point pair, determining the point pair with the same first coordinate value and the same corresponding laser transformation relation from the one-to-one candidate point pair;

screening out unique point pairs with the same first coordinate values and the same corresponding laser transformation relations from the point pairs with the same first coordinate values and the same corresponding laser transformation relations;

and combining the unique point pairs with one-to-one candidate point pairs except for the point pairs which have the same first coordinate values and correspond to the same laser transformation relations to obtain reference point pairs which do not have the repeated laser transformation relations under the condition that the first coordinate values are the same.

The one-to-one candidate point pair other than the point pair having the same first coordinate value and the same corresponding laser conversion relationship is a one-to-one candidate point pair having no repetition of the laser conversion relationship when the first coordinate values are the same.

Specifically, the computer device determines, from the one-to-one candidate point pairs, a point pair having the same first coordinate value and the same corresponding laser transformation relationship, according to the laser transformation relationship to which the one-to-one candidate point pair corresponds. The corresponding laser transformation relations are the same, and may be the same corresponding to the first laser transformation relations, or the same corresponding to the second laser transformation relations, or both the corresponding first laser transformation relations and the corresponding second laser transformation relations are the same.

And acquiring a difference value corresponding to each point pair with the same first coordinate value and the same corresponding laser transformation relation, and taking the point pair with the minimum difference value as a unique point pair with the same first coordinate value and the same corresponding laser transformation relation by the computer equipment. The difference value may refer to a first difference value or a second difference value. The first difference value is obtained by calculating the difference between the second laser point and the transformed first laser point; the transformed first laser point is obtained by transforming the first laser point according to the corresponding laser transformation relation, wherein the first laser point and the second laser point are both laser points in a point pair with the same first coordinate value and the same corresponding laser transformation relation. The second difference value is obtained by calculating the difference between the first laser point and the transformed second laser point; the transformed second laser point is obtained by transforming the second laser point according to the corresponding laser transformation relation, wherein the first laser point and the second laser point are both laser points in the point pair with the same first coordinate value and the same corresponding laser transformation relation.

The computer device combines the unique point pair and a one-to-one candidate point pair except for the point pair which has the same first coordinate value and the same corresponding laser transformation relation, and obtains a reference point pair whose laser transformation relation is not repeated in the case that the first coordinate values are the same.

In this embodiment, according to the laser transformation relationship corresponding to the one-to-one candidate point pair, the point pair having the same first coordinate value and the same corresponding laser transformation relationship is determined from the one-to-one candidate point pair, that is, the point pair having the same laser transformation relationship in the same row or the same column is determined, and only the unique point pair having the same first coordinate value and the same corresponding laser transformation relationship is retained, so that some error points are excluded, thereby improving the accuracy of point cloud construction.

In one embodiment, screening out a target point pair corresponding to a laser transformation relation corresponding to the same laser line from the reference point pair according to the laser transformation relation corresponding to the reference point pair includes: determining the laser transformation relation corresponding to the reference point pairs on the same laser line according to the laser transformation relation corresponding to the reference point pairs;

determining the laser transformation relation of the maximum number on the same laser line according to the corresponding laser transformation relation of the reference point pairs on the same laser line;

and determining a target point pair according to the point pair corresponding to the maximum number of laser transformation relations.

The laser lines in the laser transformation relation corresponding to the point pairs on the same laser line can be a first laser line or a second laser line; when the laser line is the first laser line, the corresponding laser conversion relation is the first laser conversion relation; when the laser line is the second laser line, the corresponding laser conversion relation is the second laser conversion relation.

Specifically, the computer device classifies the laser transformation relations corresponding to the reference point pairs, and determines the laser transformation relations corresponding to the reference point pairs on the same laser line; the computer equipment counts the number of the laser transformation relations corresponding to the reference point pairs on the same laser line according to the laser transformation relations corresponding to the reference point pairs on the same laser line, and determines the laser transformation relation with the maximum number on the same laser line according to the number of the laser transformation relations. The computer device may take the reference point pairs corresponding to the maximum number of laser transformation relations as the target point pairs. Or, when the maximum number meets the number threshold condition, the point pairs corresponding to the maximum number of laser conversion relations are taken as target point pairs.

In this embodiment, the laser transformation relationship corresponding to the reference points on the same laser line is determined according to the laser transformation relationship corresponding to the reference point pairs, the maximum number of laser transformation relationships on the same laser line is determined according to the laser transformation relationship corresponding to the reference point pairs on the same laser line, and the target point pair is determined according to the reference point pairs corresponding to the maximum number of laser transformation relationships, that is, the laser transformation relationships of the point pairs corresponding to the same laser line are ensured to be consistent, error points are excluded, and the accuracy of three-dimensional point cloud construction is improved.

In one embodiment, determining the target point pair according to the reference point pairs corresponding to the maximum number of laser transformation relations comprises:

determining the relation proportion of the maximum number of laser transformation relations in the laser transformation relations on the same laser line;

and when the relation proportion meets the relation proportion threshold condition and the maximum number meets the number threshold condition, taking the point pairs corresponding to the laser conversion relations with the maximum number as target point pairs.

Both the relation proportion threshold condition and the quantity threshold condition are used for judging the reliability of the laser conversion relation of the maximum quantity. The relational proportion threshold condition may be satisfied specifically by being greater than the relational proportion threshold. The number threshold condition may be satisfied specifically by being greater than a number threshold.

Specifically, the proportion of the maximum number of laser transformation relations in the laser transformation relations on the same laser line is consistent with the calculation mode of the proportion of the maximum number of reference point pairs on the same laser line determined by the computer device. For example, if the maximum number is 70 and the number of reference point pairs is 100, the relationship ratio is 70/100 × 100% to 70%.

In this embodiment, since the laser transformation relations corresponding to the reference point pairs on the same laser line may be inconsistent and the laser transformation relations corresponding to the same laser line should be consistent, the relationship proportion occupied by the maximum number of laser transformation relations in the laser transformation relations on the same laser line is determined, and when the relationship proportion satisfies the relationship proportion threshold condition and the maximum number satisfies the number threshold condition, it is indicated that the reliability of the maximum number of laser transformation relations is high, and therefore, the point pairs corresponding to the maximum number of laser transformation relations are taken as target point pairs, and point pairs conforming to the uniqueness principle are screened out, thereby improving the accuracy of the three-dimensional point cloud.

In one embodiment, determining the target point pair according to the point pairs corresponding to the maximum number of laser transformation relations includes:

when the maximum number meets the number threshold condition, converting the point pairs on the same laser line according to the maximum number of laser conversion relations, and determining the difference value corresponding to the converted point pairs;

and taking the reference point pairs which are positioned on the same laser line and the difference value of which meets the difference threshold value condition as target point pairs.

Wherein the quantity threshold condition is a threshold range pre-stored in the computer device. The number threshold condition may be determined based on the number of pixels of the first laser map or the number of pixels of the second laser map.

Specifically, a maximum number of laser conversion relationships corresponding to the first laser map will be described as an example. When the maximum number meets the number threshold condition, the computer equipment transforms a first laser point in a point pair on the same first laser line according to the maximum number of laser transformation relations corresponding to the first laser graph to obtain a transformed first laser point; calculating a first difference value between the transformed first laser point and a second laser point in the point pair on the same first laser line; and taking the point pairs which are positioned on the same first laser line and the first difference value meets the difference threshold value condition as target point pairs.

The maximum number of laser conversion relationships corresponding to the second laser map will be described as an example. When the maximum number meets the number threshold condition, the computer equipment transforms the second laser point in the point pair on the same second laser line according to the maximum number of laser transformation relations corresponding to the second laser graph to obtain a transformed second laser point; calculating a second difference value between the transformed second laser point and the first laser point in the point pair on the same second laser line; and taking the point pairs which are positioned on the same second laser line and the second difference value meets the difference threshold value condition as target point pairs.

In this embodiment, when the maximum number satisfies the number threshold condition, it indicates that the reliability of the laser transformation relationship corresponding to the maximum number is high, and then the laser transformation relationship corresponding to the maximum number is transformed, and the difference value corresponding to the transformed point pair is determined, i.e., it can be determined whether each point pair matches when the laser transformation relationship is selected, and when the difference value is smaller than the difference threshold condition, it indicates that the point pairs match with each other, so that the point pair can be used as the target point pair. By screening the number of the point pairs on the same laser line and carrying out difference screening according to the laser transformation relation, the correctness of the target point pairs can be ensured, and the accuracy of the three-dimensional point cloud is improved.

In one embodiment, constructing from the target point pairs, obtaining a three-dimensional point cloud, comprises:

and (b1) when an error laser line which has an intersection with the first coordinate value contained in the laser line corresponding to the target point pair in the same laser map and has the same laser transformation relation exists, determining the average difference value corresponding to the laser line corresponding to the target point pair and the error laser line respectively.

The first coordinate value included in the laser line corresponding to the target point pair may refer to all the first coordinate values included in the laser line.

The average difference value refers to the value of the average difference for each target point pair of a laser line. The average difference value is calculated according to the difference value corresponding to the target point pairs on the laser line and the number of the target point pairs. For example, the average difference value may be a logarithmic average value. The logarithmic mean is obtained by dividing the sum of the difference values corresponding to the target point pairs on the laser line by the number of the target point pairs and the natural logarithm of the number of the target point pairs. The average difference value may also be the sum of the differences corresponding to the target point pairs on the laser line divided by the number of the target point pairs.

Specifically, when obtaining the target point pair, the computer device also obtains a target laser transformation relationship corresponding to the target point pair. When an error laser line which has intersection with a first coordinate value contained in a laser line corresponding to a target point pair in the same laser image and has the same laser change relation exists, acquiring a difference value corresponding to the target point pair, and calculating according to the difference value corresponding to the target point pair and the number of the target point pair on the laser line to obtain an average difference value; and obtaining difference values corresponding to the target point pairs on the error laser line, and calculating to obtain an average difference value according to the sum of the difference values corresponding to the target point pairs on the error laser line and the number of the target point pairs on the error laser line.

And (b2) acquiring a candidate laser transformation relation corresponding to the laser line with the maximum average difference value.

The laser line with the largest average difference value represents that the average value of the difference values obtained by calculation after transformation according to the same laser transformation relation is the largest.

The candidate laser transformation relation may be a next largest number of laser transformation relations. For example, the laser line 1 corresponds to the largest number of laser transformation relations H₁The relationship of the laser conversion with the large number of orders is H₂. Then the candidate laser transform relationship is H₂. Alternatively, the computer device may select a laser transformation relation of which the number is the largest from among the laser transformation relations of which the average difference value satisfies the difference threshold value, as the candidate laser transformation relation.

And (b3) correcting the target point pair corresponding to the laser line with the maximum average difference value according to the candidate laser transformation relation to obtain the point pair to be constructed.

And (b4) constructing according to the point pairs to be constructed and the target point pairs to obtain the three-dimensional point cloud.

In this embodiment, when there are laser lines having the same laser transformation relationship and intersecting with the first coordinate value included in the laser line corresponding to the target point pair in the same laser map, it is described that at least two laser lines having the same laser transformation relationship appear in the same map, and then there is an error in the target point pair, so that an average difference value corresponding to the laser line corresponding to the target point pair and the error laser line respectively is determined, and a candidate laser transformation relationship corresponding to the laser line having the largest average difference value is determined, and the target point pair is corrected to obtain a point pair to be constructed, and a three-dimensional point cloud is obtained by constructing the point pair to be constructed and the target point pair, so that the target point pair can be finally corrected after uniqueness judgment, and accuracy of the obtained target point pair is improved.

In one embodiment, according to the candidate laser transformation relationship, modifying the target point pair corresponding to the laser line with the largest average difference value to obtain a point pair to be constructed includes:

according to the candidate laser transformation relation, transforming the target point pair corresponding to the laser line with the maximum average difference value, and determining the difference value of the transformed point pair;

and taking the point pairs with the difference values meeting the difference threshold value condition as the point pairs to be constructed.

It is understood that the candidate laser transformation relationship may be a first candidate laser transformation relationship or a second candidate laser transformation relationship. The first candidate laser transformation relationship is used to project points in the first laser map into the second laser map. The second candidate laser transformation relationship is used to project points in the second laser map into the first laser map.

Specifically, taking the candidate laser transformation relationship as a first candidate laser transformation relationship as an example for explanation, according to the first candidate laser transformation relationship, transforming a first laser point in a target point pair corresponding to a first laser line with the largest average difference value to obtain a transformed first laser point; and calculating a difference value between the transformed first laser point and a second laser point in the target point pair corresponding to the first laser line with the maximum average difference value, and taking the point pair with the difference value meeting the difference threshold condition as a point pair to be constructed.

In this embodiment, according to the candidate laser transformation relationship, the target point pair corresponding to the laser line with the largest average difference value is transformed, the difference value of the transformed point pair is determined, the point pair with the difference value satisfying the difference threshold condition is used as the point pair to be constructed, the target point pair obtained through uniqueness judgment and screening can be corrected, and the accuracy of the obtained three-dimensional point cloud is improved.

In one embodiment, acquiring a first set of laser points located on a first laser line comprises:

when a first coordinate value between two adjacent first laser points on the first laser line spans a preset number of positive integer values, acquiring the spanned first positive integer values, and determining a second coordinate value corresponding to the first positive integer values;

correspondingly storing the plurality of first positive integer values and second coordinate values corresponding to the first positive integer values to obtain a first laser point set;

acquiring a second set of laser points located on a second laser line, comprising:

when a first coordinate value between two adjacent second laser points on the second laser line spans a preset number of positive integer values, acquiring the spanned second positive integer values, and determining a second coordinate value corresponding to the second positive integer values;

and correspondingly storing the plurality of second positive integer values and second coordinate values corresponding to the second positive integer values to obtain a second laser point set.

The difference between the laser points extracted by the Steger algorithm is not more than 3 pixel points, so that the preset number can be 0, 1 or 2. And a coordinate point formed by the first positive integer value and the corresponding second coordinate value is positioned on the first laser line. And a coordinate point formed by the second positive integer value and the corresponding second coordinate value is positioned on the second laser line. The second coordinate values corresponding to the first positive integer value and the second positive integer value respectively may be sub-pixel coordinate values.

Specifically, when a first coordinate value between two adjacent first laser points on the first laser line crosses a preset number of positive integer values, the computer device obtains the crossed first positive integer values, performs interpolation or fitting according to coordinates of the two adjacent first laser points, and determines a second coordinate value corresponding to the first positive integer value. And the computer equipment correspondingly stores a plurality of first positive integer values and all first positive integer values on the first laser line to obtain a first laser point set.

When the first coordinate value between two adjacent second laser points on the second laser line crosses a preset number of positive integer values, acquiring the crossed second positive integer values, and performing interpolation or fitting according to the coordinates of the two adjacent second laser points, and determining the second coordinate value corresponding to the second positive integer values. And the computer equipment correspondingly stores the plurality of second positive integer values and all the second positive integer values on the second laser line to obtain a second laser point set.

For example, two adjacent points L₁And L₂And point a coordinates (5,5.2) and point B coordinates (7,6.4), if the first coordinate value of point a is 5.2, the first coordinate value of point B is 6.4, and the spanned positive integer value is 6, then linear interpolation is performed according to (5,5.2) and (7,6.4) by using a linear interpolation method, so as to obtain a second coordinate value corresponding to a second positive integer value.

In this embodiment, the expression form of the first laser spot set may be a first laser interpolation matrix, and the expression form of the second laser spot set may be a second laser interpolation matrix. The computer device may apply for a space equivalent to the size of the first laser image and an empty set matrix space equivalent to the size of the second laser image, the elements being empty for recording the first laser interpolation matrix and the second laser interpolation matrix, respectively. The rows of the matrix in the empty set matrix space are the rows of the image and the initialization of the elements of the matrix is empty. And traversing all the first laser points of each first laser line by the computer equipment, and when the first coordinate values between two adjacent first laser points span 1 or 2 first positive integer values, calculating second coordinate values corresponding to the first positive integer values by utilizing linear interpolation, and then putting the second coordinate values into a first laser interpolation matrix. And performing linear interpolation on the vertical coordinate between two adjacent points on each laser line segment to obtain an integer vertical coordinate and a sub-pixel horizontal coordinate, and filling the sub-pixel horizontal coordinate of the corresponding interpolation into a row (vertical coordinate) corresponding to the matrix space to obtain a laser matrix.

Similarly, the computer device traverses all the second laser points of each second laser line, and when the first coordinate value between two adjacent second laser points spans 1 or 2 second positive integer values, the computer device calculates the second coordinate values corresponding to the second positive integer values by using linear interpolation, and then puts the second coordinate values into a second laser interpolation matrix.

In general, the first laser image and the second laser image calibrated by binocular vision are actually the same in the center position of a pixel point, for example, the integer part of the y-axis coordinate value is the same, however, the laser point extracted from the laser line is a sub-pixel point, and therefore, a certain error exists in a mode of matching the pixel point only by the sub-pixel value. In this embodiment, when a first coordinate value between two adjacent laser points on the laser line crosses a preset number of positive integer values, the crossed positive integer values are obtained, a second coordinate value corresponding to the positive integer values is determined, and the positive integer values and the second coordinate values corresponding to the positive integer values are stored correspondingly, so that multiple uniqueness judgments can be performed based on more accurate laser points, the obtained point pairs are more accurate, and the obtained three-dimensional point cloud is more accurate.

In one embodiment, a method for determining a point pair of a multiline laser includes:

and (c1) acquiring a first laser line set in the first laser image and a second laser line set in the second laser image. The first set of laser lines includes at least one first laser line and the second set of laser lines includes at least one second laser line. The first laser image and the second laser image are obtained by shooting the same scene.

And (c2) when the first coordinate value between two adjacent first laser points on the first laser line spans a preset number of positive integer values, acquiring the spanned first positive integer value, and determining a second coordinate value corresponding to the first positive integer value.

And (c3) storing the plurality of first positive integer values in association with second coordinate values corresponding to the first positive integer values, thereby obtaining a first laser spot set.

And (c4) when the first coordinate value between two adjacent second laser points on the second laser line spans a preset number of positive integer values, acquiring the spanned second positive integer values, and determining a second coordinate value corresponding to the second positive integer values.

And (c5) storing the plurality of second positive integer values in association with second coordinate values corresponding to the second positive integer values, thereby obtaining a second laser spot set.

And (c6) for the first laser points with the same first coordinate in the first laser point set, respectively transforming the first laser points with the same first coordinate value according to the first transformation relation in the first transformation relation set to obtain transformed first laser points.

And (c7) determining a corresponding first difference value according to the transformed second coordinate value corresponding to the first laser point and the second coordinate value corresponding to the second laser point located on the same first coordinate value.

And (c8) when the first difference value satisfies the difference threshold condition, setting the first laser point and the second laser point satisfying the difference threshold condition as the first candidate point pair.

And (c9) for the second laser points with the same first coordinate value in the second laser point set, respectively transforming the second laser points with the same first coordinate value according to a second transformation relation in the second transformation relation set to obtain transformed second laser points.

And (c10) calculating a corresponding second difference value according to the first coordinate value corresponding to the transformed second laser point and the first coordinate value corresponding to the first laser point located at the same first coordinate value.

And (c11) when the second difference value satisfies the difference threshold condition, setting the second laser point and the first laser point satisfying the difference threshold condition as a second candidate point pair.

And (c12) screening the first candidate point pairs and the second candidate point pairs to obtain one-to-one candidate point pairs corresponding to the first laser image and the second laser image.

And (c13) determining the point pairs with the same first coordinate values and the same corresponding laser transformation relations from the one-to-one candidate point pairs according to the laser transformation relations corresponding to the one-to-one candidate point pairs.

And (c14) screening out unique point pairs with the same first coordinate values and the same corresponding laser transformation relations from the point pairs with the same first coordinate values and the same corresponding laser transformation relations.

And (c15) combining the unique point pairs with one-to-one candidate point pairs except for the point pairs with the same first coordinate values and the same corresponding laser transformation relations to obtain reference point pairs with non-repeated laser transformation relations under the condition that the first coordinate values are the same.

And (c16) determining the laser transformation relation corresponding to the reference point pairs on the same laser line according to the laser transformation relation corresponding to the reference point pairs.

And (c17) determining the maximum number of laser transformation relations on the same laser line according to the corresponding laser transformation relations of the reference point pairs on the same laser line.

And (c18) when the maximum number meets the number threshold condition, transforming the point pairs on the same laser line according to the maximum number of laser transformation relations, and determining the difference value corresponding to the transformed point pairs.

And (c19) taking the point pairs which are positioned on the same laser line and the difference value meets the difference threshold condition as target point pairs.

And (c20) when there is an error laser line which has an intersection with the first coordinate value contained in the laser line corresponding to the target point pair in the same laser map and has the same laser transformation relation, determining the average difference value corresponding to the laser line corresponding to the target point pair and the error laser line respectively.

And (c21) acquiring a candidate laser transformation relation corresponding to the laser line with the maximum average difference value.

And (c22) transforming the target point pair corresponding to the laser line with the maximum average difference value according to the candidate laser transformation relation, and determining the difference value of the transformed point pair.

And (c23) taking the point pairs with the difference values meeting the difference threshold value condition as the point pairs to be constructed.

And (c24) constructing the point pairs to be constructed and the target point pairs to obtain the three-dimensional point cloud.

In the method for determining a pair of multi-line laser points in this embodiment, the first laser point set and the second laser point set are subjected to four-layer screening, and the screened pair of points satisfies four conditions: one laser spot on the left image can only correspond to one unique laser spot on the right image; one laser point of the right graph can only be corresponded by only one laser point of the left graph; the laser points with the same ordinate can not generate stress light points through the same laser line change relation; the corresponding relation of one laser line segment should be unique, error points can be deleted, target point pairs of the first laser image and the second laser image can be effectively screened, the target point pairs can be constructed according to the target point pairs, three-dimensional point cloud can be obtained, the deviation of effective points and the three-dimensional point cloud can be achieved, the stable high speed can be kept, meanwhile, more points are kept, and the real-time performance is good and the practicability is wide.

In one embodiment, a method for determining a point pair of multi-line laser, where a first camera is a left camera, a second camera is a right camera, a first laser line is a left laser line segment, and a second laser line is a right laser line segment, is described as an example, and includes:

1. obtaining calibration parameters: left camera laser line transformation relation H ═ H₁,H₂,...,H_nConverting relation with the right camera laser line;

2. the left camera and the right camera shoot the laser line image, the steger algorithm is utilized to extract the laser central line, and left and right laser line segments { l } are obtained_i},{r_jAnd the laser spot on the line segment;

3. setting a difference threshold delta x, a laser line segment average error threshold dx and an effective laser point proportion threshold s;

4. for { l_i},{r_jCarrying out fixed-space fast interpolation, wherein the specific interpolation process is as follows:

firstly, applying 2 spaces with the same size as an image, wherein elements are null, and respectively recording a left image laser interpolation matrix and a right image laser interpolation matrix;

② traverse { l_iEach line l_iIf the vertical coordinate between two adjacent points spans 1 or two positive integer values, the horizontal coordinate corresponding to the integer values is calculated by linear interpolation, and then the horizontal coordinate is put into an interpolation matrix L to traverse { r } in the same way_jObtaining an interpolation matrix R;

5. traversing each line L of the interpolation matrix L of the left graph_iR of the right-view laser matrix R corresponding to the same row_i，

To (L)_i,R_i) The following operations are carried out:

firstly, carrying out first weight uniqueness judgment, L_iEach point in the graph corresponds to R_iOne or no corresponding point, specifically L_iEach point in the set of points is according to H ═ H₁,H₂,...,H_nIs transformed into a new point, if the new point and R_iIf the difference of a certain point is less than the difference threshold value delta x, the point is a possible corresponding point, and the corresponding relation and the applied H are recorded_i；

Secondly, a left-to-right many-to-one corresponding relation is obtained in the previous step, second uniqueness judgment from right to left is carried out, namely the right point cannot be simultaneously corresponding to the left points, the corresponding point pair with the minimum difference is selected at the moment, and H is recorded simultaneously_i；

Thirdly, the correspondence obtained in the previous step after two uniqueness determinations needs to be specific to H_iMaking a third reduplicative decision, i.e. L_iIn which it is impossible to use the same H for two points_iAnd finding a corresponding point. Selecting the corresponding point with the minimum difference as reasonable H_iIn the case of corresponding point pairs, the remaining point pairs correspond to H using the next smallest difference from the previous step_iAnd the next smallest difference is smaller than the difference threshold deltax, and the third step is repeated until L_iIn which two points cannot appearUsing the same H_iFinding out corresponding points, wherein the difference of each group of corresponding points is the best minimum value detected by the corresponding point;

fourthly, all corresponding points and H obtained in the previous step_iTo be placed on the laser line segment l_iPerforming fourth uniqueness judgment, specifically: l_iThe same H should be applied to the laser spot above the laser line_j. The previous step gives_iAll corresponding point pairs above, according to H_jClassifying the point pairs, selecting the class with the largest proportion, and if the proportion is larger than a proportion threshold value s, determining that the most probable H is_jClass laser line, thereby extracting l_iUpper utilization of H_jCalculating the average difference of the corresponding points, and selecting H_kIn which H is_kAct on_iRatio of H to H_jLess and average difference ratio H_jAnd correspondingly large.

If l appears_iAnd l_kWith intersection in ordinate and using the same H_jIf the corresponding proportion is greater than s and the average difference is smaller than the average threshold value dx, the smallest average difference is selected as the effective laser line segment, and the next smallest H_uH corresponding to the other laser line segment_u。

Fifthly, circularly performing the fourth step until each laser line segment with intersection of the ordinate has only one H_iThe relationship is transformed until the calculated values all satisfy the threshold.

6. And (5) performing three-dimensional reconstruction on the effective line segments and the effective points obtained in the step 5 to obtain point cloud.

It should be understood that, although the steps in the flowcharts of fig. 2 and 4 are sequentially shown as indicated by arrows, and the steps from step (a1) to step (a6), from step (b1) to step (b4), and from step (c1) to step (c24) are sequentially shown as indicated by reference numerals, the steps are not necessarily performed sequentially in the order indicated by arrows or numerals. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2 and 4 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 5, a block diagram of a point pair determining apparatus for a multiline laser in one embodiment is provided, which includes: a laser line acquisition module 502, a laser point acquisition module 504, a multiplicity uniqueness determination module 506, and a construction module 508, wherein:

a laser line obtaining module 502, configured to obtain a first laser line set in a first laser map and a second laser line set in a second laser map; the first laser line set comprises at least one first laser line, and the second laser line set comprises at least one second laser line; the first laser image and the second laser image are obtained by shooting the same scene;

a laser spot acquisition module 504, configured to acquire a first set of laser spots on a first laser line and a second set of laser spots on a second laser line;

a multiple uniqueness judgment module 506, configured to perform multiple uniqueness judgment on the first laser point set and the second laser point set according to the laser transformation relationship set to obtain a target point pair;

and the constructing module 508 is configured to construct the target point pairs to obtain a three-dimensional point cloud.

In the device for determining a pair of multi-line laser points in this embodiment, if a first laser line set in a first laser diagram and a second laser line set in a second laser diagram are obtained, a first laser point set located on the first laser line is obtained, and a second laser point set located on the second laser line is obtained, then enough points can be obtained; according to the laser transformation relation set, multiple uniqueness judgment is carried out on the first laser point set and the second laser point set to obtain target point pairs, the target point pairs are constructed according to the target point pairs to obtain three-dimensional point clouds, unique corresponding point pairs can be obtained, stable high speed can be kept, more corresponding points are reserved, and the accuracy of the obtained target point pairs is improved, so that the reconstructed three-dimensional point clouds are more accurate, good in real-time performance and wide in practicability.

In one embodiment, the multiple uniqueness determination module 506 includes a corresponding point uniqueness determination unit and a laser transformation relation uniqueness determination unit; the corresponding point uniqueness judging unit is used for judging the corresponding point uniqueness of the first laser point set and the second laser point set according to the laser transformation relation set to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image;

and the laser conversion relation uniqueness judging unit is used for judging the uniqueness of the laser conversion relation of the one-to-one point pair according to the laser conversion relation corresponding to the one-to-one candidate point pair to obtain the target point pair.

In one embodiment, the set of laser transformation relationships includes a first set of laser transformation relationships and a second set of laser transformation relationships. The corresponding point uniqueness judgment unit is used for transforming the first laser points in the first laser point set according to the first laser transformation relation set to obtain transformed first laser points; under the condition that the transformed first laser point is matched with a second laser point in a second laser point set, taking the first laser point and the matched second laser point as a first candidate point pair; for a second laser point in the second laser point set, converting the second laser point according to a second laser conversion relation set to obtain a converted second laser point; under the condition that the transformed second laser point is matched with the first laser point in the first laser point set, taking the second laser point and the matched first laser point as a second candidate point pair; and screening the first candidate point pair and the second candidate point pair to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image.

In one embodiment, the first laser map and the second laser map are both maps obtained after binocular vision correction. The corresponding point uniqueness judgment unit is used for respectively transforming first laser points with the same first coordinate value point in the first laser point set according to a first transformation relation in the first transformation relation set for the first laser points with the same first coordinate value point in the first laser point set to obtain transformed first laser points; determining a corresponding first difference value according to a second coordinate value corresponding to the transformed first laser point and a second coordinate value corresponding to a second laser point positioned on the same first coordinate value; under the condition that the first difference value meets a difference threshold condition, taking the first laser point and the second laser point meeting the difference threshold condition as a first candidate point pair; for second laser points with the same first coordinate value in the second laser point set, respectively transforming the second laser points with the same first coordinate value according to a second transformation relation in the second transformation relation set to obtain transformed second laser points; calculating a corresponding second difference value according to the first coordinate value corresponding to the transformed second laser point and the first coordinate value corresponding to the first laser point positioned at the same first coordinate value; and under the condition that the second difference value meets the difference threshold condition, taking the second laser point and the first laser point meeting the difference threshold condition as a second candidate point pair.

In one embodiment, the laser transformation relation uniqueness determination unit is configured to screen out, from the one-to-one candidate point pairs, reference point pairs whose laser transformation relations are not repeated when the first coordinate values are the same, according to the laser transformation relations corresponding to the one-to-one candidate point pairs;

In one embodiment, the laser transformation relation uniqueness determination unit is configured to determine, from the pair of one-to-one candidate points, a point pair having the same first coordinate value and the same corresponding laser transformation relation according to the laser transformation relation corresponding to the pair of one-to-one candidate points;

In one embodiment, the laser transformation relation uniqueness judgment unit is configured to determine the laser transformation relation corresponding to the reference point pairs located on the same laser line according to the laser transformation relation corresponding to the reference point pairs;

In one embodiment, the laser transformation relation uniqueness determination unit is further configured to determine a relation proportion of the maximum number of laser transformation relations among the laser transformation relations on the same laser line;

In one embodiment, the laser transformation relation uniqueness judgment unit is further configured to, when the maximum number meets a number threshold condition, transform the point pairs located on the same laser line according to the maximum number of laser transformation relations, and determine a difference value corresponding to the transformed point pairs;

In one embodiment, the building module 508 is configured to, when there is an error laser line that has an intersection with the first coordinate value included in the laser line corresponding to the target point pair in the same laser map and has the same laser transformation relationship, determine an average difference value corresponding to each of the laser line corresponding to the target point pair and the error laser line; obtaining a candidate laser transformation relation corresponding to the laser line with the maximum average difference value; correcting the target point pair corresponding to the laser line with the maximum average difference value according to the candidate laser transformation relation to obtain a point pair to be constructed; and constructing according to the point pairs to be constructed and the target point pairs to obtain the three-dimensional point cloud.

In an embodiment, the constructing module 508 is further configured to transform a target point pair corresponding to a laser line with a largest average difference value according to a candidate laser transformation relationship, and determine a difference value of the transformed point pair; and taking the point pairs with the difference values meeting the difference threshold value condition as the point pairs to be constructed.

In an embodiment, the laser point obtaining module 504 is configured to, when a first coordinate value between two adjacent first laser points on the first laser line spans a preset number of positive integer values, obtain the spanned first positive integer value, and determine a second coordinate value corresponding to the first positive integer value;

For the specific definition of the device for determining the point pairs of the multi-line laser, reference may be made to the above definition of the method for determining the point pairs of the multi-line laser, and details thereof are not repeated herein. The modules in the point pair determining device of the multi-line laser can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal device, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method for testing pin connectivity of a chip. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, which includes a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the above embodiments of the testing method for pin connectivity of each chip when executing the computer program.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the above-described embodiments of the testing method for pin connectivity of each chip.

It will be understood by those skilled in the art that all or part of the processes of the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a non-volatile computer readable storage medium, and when executed, may include the processes of the above embodiments of the methods. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile asperities may include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. A method of point pair determination for a multiline laser, the method comprising:

2. The method of claim 1, wherein the performing multiple uniqueness decisions on the first set of laser points and the second set of laser points according to a set of laser transformation relationships to obtain pairs of target points comprises:

3. The method of claim 2, wherein the set of laser transformation relationships comprises a first set of laser transformation relationships and a second set of laser transformation relationships;

the method for determining the uniqueness of corresponding points of the first laser point set and the second laser point set according to the laser transformation relation set to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image includes:

for a first laser point in the first laser point set, transforming the first laser point according to the first laser transformation relation set to obtain a transformed first laser point;

under the condition that the transformed first laser point is matched with a second laser point in the second laser point set, taking the first laser point and the matched second laser point as a first candidate point pair;

for a second laser point in the second laser point set, transforming the second laser point according to the second laser transformation relation set to obtain a transformed second laser point;

under the condition that the transformed second laser point is matched with the first laser point in the first laser point set, taking the second laser point and the matched first laser point as a second candidate point pair;

and screening the first candidate point pair and the second candidate point pair to obtain a one-to-one candidate point pair corresponding to the first laser image and the second laser image.

4. The method of claim 3, wherein the first laser map and the second laser map are both maps obtained after binocular vision correction;

for first laser points with the same first coordinate value in a first laser point set, respectively transforming the first laser points with the same first coordinate value according to a first transformation relation in the first transformation relation set to obtain transformed first laser points;

when the transformed first laser point is matched with a second laser point in the second laser point set, taking the first laser point and the matched second laser point as a first candidate point pair includes:

determining a corresponding first difference value according to a second coordinate value corresponding to the transformed first laser point and a second coordinate value corresponding to a second laser point located on the same first coordinate value;

taking the first laser point and a second laser point meeting a difference threshold condition as a first candidate point pair under the condition that the first difference value meets the difference threshold condition;

for a second laser point in the second laser point set, transforming the second laser point according to the second laser transformation relation set to obtain a transformed second laser point, including:

for a second laser point with the same first coordinate value in a second laser point set, respectively transforming the second laser points with the same first coordinate value according to a second transformation relation in a second transformation relation set to obtain a transformed second laser point;

taking the second laser point and the matched first laser point as a second candidate point pair when the transformed second laser point is matched with the first laser point in the first laser point set, including:

calculating a corresponding second difference value according to the first coordinate value corresponding to the transformed second laser point and the first coordinate value corresponding to the first laser point positioned at the same first coordinate value;

and under the condition that the second difference value meets a difference threshold condition, taking the second laser point and the first laser point meeting the difference threshold condition as a second candidate point pair.

5. The method according to claim 2, wherein the performing laser transform relationship uniqueness judgment on the one-to-one point pair according to the laser transform relationship corresponding to the one-to-one candidate point pair to obtain the target point pair comprises:

and screening out a target point pair corresponding to the unique corresponding laser transformation relation on the same laser line from the reference point pair according to the laser transformation relation corresponding to the reference point pair.

6. The method according to claim 5, wherein the step of screening out a reference point pair, for which the laser transformation relationship does not overlap if the first coordinate values are the same, from the one-to-one candidate point pair according to the laser transformation relationship corresponding to the one-to-one candidate point pair comprises:

and combining the unique point pairs with one-to-one candidate point pairs except for the point pairs which have the same first coordinate values and the same corresponding laser transformation relations to obtain reference point pairs with non-repeated laser transformation relations under the condition that the first coordinate values are the same.

7. The method according to claim 5, wherein the screening out, from the reference point pairs, a target point pair corresponding to a unique corresponding laser transformation relation on the same laser line according to the laser transformation relation corresponding to the reference point pair comprises:

determining the laser transformation relation corresponding to the reference point pairs on the same laser line according to the laser transformation relation corresponding to the reference point pairs;

determining the laser transformation relation of the maximum number on the same laser line according to the laser transformation relation corresponding to the reference point pairs on the same laser line;

and determining a target point pair according to the reference point pair corresponding to the maximum number of laser transformation relations.

8. The method of claim 7, wherein the determining the target point pairs according to the reference point pairs corresponding to the maximum number of laser transformation relationships comprises:

determining the relation proportion of the laser transformation relation of the maximum number in the laser transformation relations on the same laser line;

and when the relation proportion meets a relation proportion threshold condition and the maximum number meets a number threshold condition, taking the point pairs corresponding to the laser conversion relations with the maximum number as target point pairs.

9. The method of claim 7, wherein the determining the target point pairs according to the reference point pairs corresponding to the maximum number of laser transformation relationships comprises:

when the maximum number meets the number threshold condition, transforming the point pairs on the same laser line according to the maximum number of laser transformation relations, and determining the difference value corresponding to the transformed point pairs;

and taking the reference point pairs which are positioned on the same laser line and the difference values of which meet the difference threshold value condition as target point pairs.

10. The method according to any one of claims 1 to 9, wherein constructing from the target point pairs, obtaining a three-dimensional point cloud comprises:

when an error laser line which has intersection with a first coordinate value contained in a laser line corresponding to a target point pair in the same laser image and has the same laser transformation relation exists, determining an average difference value corresponding to the laser line corresponding to the target point pair and the error laser line respectively;

obtaining a candidate laser transformation relation corresponding to the laser line with the maximum average difference value;

according to the candidate laser transformation relation, correcting the target point pair corresponding to the laser line with the maximum average difference value to obtain a point pair to be constructed;

and constructing according to the point pairs to be constructed and the target point pairs to obtain the three-dimensional point cloud.

11. The method according to claim 10, wherein the modifying the target point pair corresponding to the laser line with the largest average difference value according to the candidate laser transformation relation to obtain the point pair to be constructed comprises:

and taking the point pairs with the difference values meeting the difference threshold value condition as point pairs to be constructed.

12. The method of any of claims 1 to 9, wherein said acquiring a first set of laser spots located on a first laser line comprises:

when a first coordinate value between two adjacent first laser points on a first laser line spans a preset number of positive integer values, acquiring the spanned first positive integer values, and determining a second coordinate value corresponding to the first positive integer values;

correspondingly storing the plurality of first positive integer values and the second coordinate values corresponding to the first positive integer values to obtain a first laser point set;

the acquiring a second set of laser points located on a second laser line includes:

when a first coordinate value between two adjacent second laser points on a second laser line spans a preset number of positive integer values, acquiring the spanned second positive integer values, and determining a second coordinate value corresponding to the second positive integer values;

and correspondingly storing the plurality of second positive integer values and the second coordinate values corresponding to the second positive integer values to obtain a second laser point set.

13. A device for point pair determination for a multiline laser, the device comprising:

14. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 12.

15. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 12.