Confidence evaluation method and device for three-dimensional depth data and computer equipment

文档序号：1615060 发布日期：2020-01-10 浏览：14次中文

阅读说明：本技术 三维深度数据的置信度评估方法、装置和计算机设备 (Confidence evaluation method and device for three-dimensional depth data and computer equipment ) 是由陈海龙杨洋刘梦龙刘晓利于 2019-09-27 设计创作，主要内容包括：本申请涉及一种三维深度数据的置信度评估方法、装置、计算机设备和存储介质。所述方法包括：在通过条纹投影结构光三维成像系统确定物体表面匹配的目标点的过程中,得到点云置信度的影响因素；建立各个影响因素与点云置信度之间的关系,得出各个影响因素对点云置信度的影响程度；对各个影响因素赋予不同的影响权重,建立各个影响因素的综合结果与点云置信度间的关系,得出点云的置信度；根据点云的置信度,对点云的质量进行评估。由于是基于条纹投影结构光三维成像系统生成三维点的过程得出影响点云置信度的因素,再对点云的质量进行评估的,因此,所得出的点云的质量不会受到周围拓扑点的干扰。(The application relates to a confidence evaluation method and device for three-dimensional depth data, computer equipment and a storage medium. The method comprises the following steps: obtaining influence factors of point cloud confidence in the process of determining a target point matched with the surface of an object by a fringe projection structured light three-dimensional imaging system; establishing a relation between each influence factor and the point cloud confidence coefficient to obtain the influence degree of each influence factor on the point cloud confidence coefficient; giving different influence weights to each influence factor, establishing a relation between a comprehensive result of each influence factor and the confidence coefficient of the point cloud, and obtaining the confidence coefficient of the point cloud; and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud. Because the factors influencing the confidence coefficient of the point cloud are obtained based on the process of generating the three-dimensional points by the fringe projection structured light three-dimensional imaging system, and then the quality of the point cloud is evaluated, the quality of the obtained point cloud is not interfered by surrounding topological points.)

1. A method of confidence assessment of three-dimensional depth data, the method comprising:

obtaining influence factors of point cloud confidence in the process of determining a target point matched with the surface of an object by a fringe projection structured light three-dimensional imaging system; the influencing factors comprise at least one of modulation degree, surface reflectivity jump of the measured object, deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and effective measurement space range of the fringe projection structured light three-dimensional imaging system;

establishing a relation between each influence factor and the point cloud confidence coefficient, and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relation between each influence factor and the point cloud confidence coefficient;

according to the influence degree of each influence factor on the point cloud confidence coefficient, different influence weights are given to each influence factor, and the relation between the comprehensive result of each influence factor and the point cloud confidence coefficient is established;

obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

2. The method of claim 1, wherein the obtaining the influence factor of the point cloud confidence in determining the target point matched with the object surface in the structured light three-dimensional imaging system through fringe projection comprises:

acquiring a standard sine stripe image, projecting the standard sine stripe image to the surface of an object in the stripe projection structured light three-dimensional imaging system, acquiring a projection image of the surface of the object to obtain a deformed stripe image corresponding to the standard sine stripe image, wherein the acquisition result of the projection image of the surface of the object is influenced by the effective measurement space range in the stripe projection structured light three-dimensional imaging system;

calculating the phase of the deformed fringe image by using a phase shift method, wherein the calculation result of the phase is influenced by the modulation degree;

the modulation degree is influenced by the change of the reflectivity of the surface of the measured object;

determining corresponding phase information according to the phase, and determining a first corresponding point of a point on the surface of the object on an imaging surface of a first camera and a second corresponding point on an imaging surface of a second camera in the fringe projection structured light three-dimensional imaging system according to the phase information and the geometric relation;

determining the intersection point of the two straight lines according to the straight line connecting the first corresponding point and the central point of the first camera and the straight line connecting the second corresponding point and the central point of the second camera, wherein the intersection point of the two straight lines is a target point matched with the surface of the object;

the determining process of the target point is influenced by the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system is an included angle between the normal direction of the target point and the angle bisector of the two straight lines.

3. The method of claim 1, wherein the establishing a relationship between each influencing factor and the point cloud confidence level, and the deriving the degree of influence of each influencing factor on the point cloud confidence level according to the relationship between each influencing factor and the point cloud confidence level comprises:

establishing a relation between the modulation value and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the modulation value and the quality of the point cloud;

and obtaining the influence degree of the modulation degree on the confidence coefficient of the point cloud according to the relation between the modulation degree and the confidence coefficient of the point cloud.

4. The method of claim 1, wherein the establishing a relationship between each influencing factor and the point cloud confidence level, and the deriving the degree of influence of each influencing factor on the point cloud confidence level according to the relationship between each influencing factor and the point cloud confidence level comprises:

establishing a relation between the change of the reflectivity of the surface of the measured object and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the modulation difference value of the reflectivity change of the surface of the measured object and the quality of the point cloud;

and obtaining the influence degree of the surface reflectivity change of the measured object on the point cloud confidence degree according to the relation between the surface reflectivity change of the measured object and the point cloud confidence degree.

5. The method of claim 1, wherein the establishing a relationship between each influencing factor and the point cloud confidence level, and the deriving the degree of influence of each influencing factor on the point cloud confidence level according to the relationship between each influencing factor and the point cloud confidence level comprises:

establishing a relation between the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud by using a Gaussian kernel function according to the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the quality of the point cloud;

and obtaining the influence degree of the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud according to the relation between the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud.

6. The method of claim 1, wherein the establishing a relationship between each influencing factor and the point cloud confidence level, and the deriving the degree of influence of each influencing factor on the point cloud confidence level according to the relationship between each influencing factor and the point cloud confidence level comprises:

establishing a relation between the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the effective measurement space range of the fringe projection structured light three-dimensional imaging system and the quality of the point cloud;

and obtaining the influence degree of the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud according to the relationship between the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud.

7. The method of claim 1, further comprising:

acquiring a point cloud set under each visual angle, which is obtained by carrying out multi-directional and multi-angle scanning on a measured object by a three-dimensional scanner;

screening target point clouds corresponding to all the visual angles according to the point cloud set under each visual angle and the contact ratio of the point clouds under all the visual angles; the average value of the confidence degrees of the target point clouds is highest;

determining the position relation between each piece of target point cloud according to the target point cloud;

obtaining the closest point in the overlapping area of each target point cloud by using a forward projection method according to the position relation between each target point cloud;

performing global closest point iterative optimization according to the closest point in each point cloud overlapping region, and taking the point cloud confidence as a weight factor of global matching optimization to participate in the global closest point iterative optimization; the weight factor of the global matching optimization is the average value of the sum of the confidence degrees of each pair of nearest points;

and obtaining a point cloud matching result according to the global closest point iterative optimization result.

8. An apparatus for confidence evaluation of three-dimensional depth data, the apparatus comprising:

the acquisition module is used for acquiring influence factors of point cloud confidence in the process of determining a target point matched with the surface of an object by a fringe projection structured light three-dimensional imaging system; the influencing factors comprise at least one of modulation degree, surface reflectivity jump of the measured object, deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and effective measurement space range of the fringe projection structured light three-dimensional imaging system;

the influence degree calculation module is used for establishing the relationship between each influence factor and the point cloud confidence coefficient and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relationship between each influence factor and the point cloud confidence coefficient;

the relation establishing module is used for endowing different influence weights to the influence factors according to the influence degrees of the influence factors on the point cloud confidence coefficient, and establishing the relation between the comprehensive result of the influence factors and the point cloud confidence coefficient;

the confidence coefficient calculation module is used for obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and the quality evaluation module is used for evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the computer program is executed by the processor.

10. 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 7.

Technical Field

The present application relates to the field of three-dimensional reconstruction technology, and in particular, to a method and an apparatus for confidence evaluation of three-dimensional depth data, a computer device, and a storage medium.

Background

With the continuous development of modern technologies and the continuous improvement of manufacturing processes, the demand of each industry for three-dimensional digital modeling is continuously increasing. As a generally adopted practical technology, the three-dimensional reconstruction technology based on computer vision effectively solves the problem of low precision of a purely manual modeling method, thereby being widely applied. The three-dimensional point cloud matching and fusing technology is an important data processing step in a three-dimensional reconstruction technology and plays an extremely important role in acquiring the three-dimensional point cloud in the surface space of a complete measured object. In the actual process, the number of the three-dimensional points is huge, and the three-dimensional point cloud needs to be sampled. Since the quality of each three-dimensional point is different, it is necessary to have a certain guidance in sampling, that is, to evaluate the quality of the three-dimensional points.

In order to solve the above problems, the conventional method evaluates the quality of the point cloud by establishing a topological relation between the point clouds and a three-dimensional point grid structure constructed according to the topological relation of the point clouds after the three-dimensional point cloud is obtained. However, the quality of each point cloud is affected by the surrounding topological points, which interferes with the quality evaluation of the point cloud.

Disclosure of Invention

In view of the above, it is necessary to provide a confidence evaluation method, an apparatus, a computer device, and a storage medium capable of reducing the disturbing three-dimensional depth data.

A method of confidence assessment of three-dimensional depth data, the method comprising:

obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

In one embodiment, the obtaining of the influence factor of the point cloud confidence in the process of determining the target point matched with the surface of the object in the structured light three-dimensional imaging system through fringe projection includes:

calculating the phase of the deformed fringe image by using a phase shift method, wherein the calculation result of the phase is influenced by the modulation degree;

the modulation degree is influenced by the change of the reflectivity of the surface of the measured object;

In one embodiment, the establishing a relationship between each influence factor and the point cloud confidence, and obtaining the influence degree of each influence factor on the point cloud confidence according to the relationship between each influence factor and the point cloud confidence includes:

In one embodiment, the method further comprises:

acquiring a point cloud set under each visual angle, which is obtained by carrying out multi-directional and multi-angle scanning on a measured object by a three-dimensional scanner;

determining the position relation between each piece of target point cloud according to the target point cloud;

obtaining the closest point in the overlapping area of each target point cloud by using a forward projection method according to the position relation between each target point cloud;

performing global closest point iterative optimization according to the closest point in each point cloud overlapping region, and taking the point cloud confidence as a weight factor of global matching optimization to participate in the global closest point iterative optimization; the weight factor of the global matching optimization is the average of the sum of the confidences of each pair of closest points.

And obtaining a point cloud matching result according to the global closest point iterative optimization result.

An apparatus for confidence evaluation of three-dimensional depth data, the apparatus comprising:

and the quality evaluation module is used for evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:

obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

According to the confidence evaluation method and device of the three-dimensional depth data, the computer equipment and the storage medium, influence factors of the confidence of the point cloud are obtained in the process of determining the target point matched with the surface of the object by the fringe projection structured light three-dimensional imaging system. Establishing a relation between each influence factor and the point cloud confidence coefficient to obtain the influence degree of each factor on the point cloud confidence coefficient, endowing different weights to each influence factor, establishing a relation between the comprehensive result of each influence factor and the point cloud confidence coefficient to obtain the point cloud confidence coefficient. And evaluating the quality of the point cloud according to the obtained point cloud confidence coefficient to obtain a point cloud quality evaluation result. Because the factors influencing the confidence coefficient of the point cloud are obtained based on the process of generating the three-dimensional points by the fringe projection structured light three-dimensional imaging system, and then the quality of the point cloud is evaluated, the quality of the obtained point cloud is not interfered by surrounding topological points.

Drawings

FIG. 1 is a diagram illustrating an exemplary embodiment of a confidence evaluation method for three-dimensional depth data;

FIG. 2 is a schematic flow chart illustrating a confidence evaluation method for three-dimensional depth data according to an embodiment;

FIG. 3 is a schematic diagram of a fringe projection structured light three-dimensional imaging system in one embodiment;

FIG. 4 is a graph of a confidence function relationship in one embodiment;

FIG. 5 is a schematic flow chart illustrating the derivation of point cloud confidence impact factors in one embodiment;

FIG. 6 is a schematic diagram of an effective measurement space of a fringe projection structured light three-dimensional imaging system in one embodiment;

FIG. 7 is a graphical illustration of the effect of the degree of deflection of a surface of an object under measurement on the degree of confidence of the measurement device in one embodiment;

FIG. 8 is a diagram illustrating modulation of an object under test in one embodiment;

FIG. 9 is a graph of modulation versus confidence results for one embodiment;

FIG. 10 is a graph of the effect of changes in surface reflectance of an object under test on confidence in one embodiment;

FIG. 11 is a graph of the effect of the degree of deflection versus confidence for the surface of the object under test and a fringe projection structured light three-dimensional imaging system in one embodiment;

FIG. 12 is a graphical illustration of confidence levels for a combination of four influencing factors in one embodiment;

FIG. 13 is a schematic flow chart illustrating participation of point cloud confidence in global optimization according to an embodiment;

FIG. 14 is a diagram illustrating comparison of results of point cloud confidence in global optimization and direct optimization according to an embodiment;

FIG. 15 is a block diagram of an apparatus for confidence evaluation of three-dimensional depth data according to an embodiment;

FIG. 16 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The confidence evaluation method of the three-dimensional depth data provided by the application can be applied to the application environment shown in fig. 1. The terminal 110 is connected to the three-dimensional scanner 120 through a USB. The three-dimensional scanner 120 forms a fringe projection structured light three-dimensional imaging system by projecting a standard sinusoidal fringe image onto a measured object. The terminal 110 obtains image data collected by the three-dimensional scanner 120, and obtains influence factors of point cloud confidence based on a process of generating three-dimensional points by a fringe projection structured light three-dimensional imaging system. And establishing a relational expression according to the influence factors of the point cloud confidence coefficient to obtain the point cloud confidence coefficient under each influence factor and the point cloud confidence coefficient under the comprehensive action of all the influence factors, and evaluating the quality of the point cloud of the image. The terminal 110 may be, but is not limited to, various personal computers, notebook computers, and tablet computers.

In an embodiment, as shown in fig. 2, a confidence evaluation method for three-dimensional depth data is provided, which is described by taking the method as an example applied to the terminal in fig. 1, and includes the following steps:

202, in the process of determining a target point matched with the object surface through the fringe projection structured light three-dimensional imaging system, obtaining influence factors of point cloud confidence, including at least one of a modulation degree, a jump of the reflectivity of the object surface to be measured, a deflection degree of the object surface to be measured and the fringe projection structured light three-dimensional imaging system, and an effective measurement space range of the fringe projection structured light three-dimensional imaging system.

The point cloud confidence coefficient refers to the precision and reliability of the point cloud, namely the quality of the point cloud. The fringe projection structured light three-dimensional imaging system consists of two cameras and a projector, as shown in fig. 3. The projector is used to project the computer-generated pattern onto the surface of the object. The pattern is modulated by the surface of the object, and the two cameras are used for collecting the pattern modulated by the surface of the object. And by combining the process of generating the three-dimensional points by the fringe projection structured light three-dimensional imaging system, the factors influencing the generation of the three-dimensional points are obtained.

Specifically, the factors influencing the generation of the three-dimensional point include at least one of a modulation degree, a surface reflectivity jump of the measured object, a deflection degree between the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and an effective measurement space range of the fringe projection structured light three-dimensional imaging system. When the generation process of the three-dimensional point is affected, the quality of the finally generated point cloud is also affected. Therefore, factors affecting the generation of the three-dimensional point are also factors affecting the confidence of the point cloud. The influence factors of the point cloud confidence include at least one of a modulation degree, the surface reflectivity jump of the measured object, the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and the effective measurement space range of the fringe projection structured light three-dimensional imaging system.

And 204, establishing a relation between each influence factor and the point cloud confidence coefficient, and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relation between each influence factor and the point cloud confidence coefficient.

The relationship between each influencing factor and the confidence coefficient of the point cloud is established through a Gaussian kernel function. Because the gaussian kernel function has the property of defining a threshold, the influence coefficient of each factor on the confidence of the point cloud can be defined from 0 to 1. Specifically, as shown in fig. 4, the confidence coefficient influence function constructed by using the gaussian kernel function is mainly divided into three action regions: [ X1, X4], [ X4, X9] and [ X9, X12 ]. In the first interval, the confidence corresponding to the influence factor is close to 0. In the second range, the confidence level is increased with increasing influence factors. In the third range, the confidence corresponding to the influencing factor is close to 1, and the confidence is relatively high. Because the action range and the action degree of each factor on the point cloud confidence coefficient are different, the values of the influence degree factor and the modulation degree factor in the corresponding relational expression are different.

And step 206, according to the influence degree of each influence factor on the point cloud confidence coefficient, giving different influence weights to each influence factor, and establishing the relation between the comprehensive result of each influence factor and the point cloud confidence coefficient.

After the influence degree of each influence factor on the point cloud confidence is obtained, the influence degree values of each factor on the point cloud confidence are added, and different influence weights are given to each factor according to the influence degree. Wherein the factor with the greatest influence degree is given the greatest weight; and the values of all weights add up to 1.

And 208, obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud.

And after the relation between the comprehensive result of each influence factor and the confidence coefficient of the point cloud is obtained, calculating to obtain the confidence coefficient of the point cloud under the comprehensive action.

And step 210, evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

And after the confidence degrees of the point clouds are obtained, evaluating the quality of the point clouds according to the confidence degree value of each point. Specifically, when the confidence of a certain point is closer to 0, the confidence of the point is lower; when the confidence of a certain point is closer to 1, the higher the confidence of the point is. And when the confidence degrees of all the points in a piece of point cloud are higher, the confidence degree of the point cloud is correspondingly higher.

According to the confidence evaluation method of the three-dimensional depth data, influence factors of the point cloud confidence are obtained in the process of determining the target point matched with the surface of the object by the fringe projection structured light three-dimensional imaging system. Establishing a relation between each influence factor and the point cloud confidence coefficient to obtain the influence degree of each factor on the point cloud confidence coefficient, endowing different weights to each influence factor, establishing a relation between the comprehensive result of each influence factor and the point cloud confidence coefficient to obtain the point cloud confidence coefficient. And evaluating the quality of the point cloud according to the obtained point cloud confidence coefficient to obtain a point cloud quality evaluation result. Because the factors influencing the confidence coefficient of the point cloud are obtained based on the process of generating the three-dimensional points by the fringe projection structured light three-dimensional imaging system, and then the quality of the point cloud is evaluated, the obtained point cloud quality is not interfered by surrounding topological points, and the method is more accurate.

step S502, a standard sine stripe image is obtained and projected to the surface of an object in the stripe projection structured light three-dimensional imaging system, a projection image of the surface of the object is obtained to obtain a deformed stripe image corresponding to the standard sine stripe image, wherein the obtaining result of the projection image of the surface of the object is influenced by the effective measurement space range in the stripe projection structured light three-dimensional imaging system.

As shown in fig. 3, the projector in the fringe projection structured light three-dimensional imaging system acquires a standard sinusoidal fringe image generated by a computer and projects the standard sinusoidal fringe image onto the surface of an object. The object surface modulates a standard sinusoidal fringe image. And the two cameras collect the deformed stripe images modulated by the surface of the object.

Because the fringe projection structured light three-dimensional imaging system is composed of a projector and two cameras, the cameras and the projector both have depth of field, and the optical axes of the three devices have a certain included angle, the fringe projection structured light three-dimensional imaging system has an effective measurement space, as shown in fig. 6. The effective measurement space range in the fringe projection structured light three-dimensional imaging system can affect the projection of a projector and the acquisition of two cameras, thereby affecting the acquisition result of a projection image on the surface of an object. In contrast, point clouds within the effective measurement space are more confident than point clouds outside the effective measurement space, and the farther away from the effective measurement space the lower the confidence level.

Step S504 is to calculate the phase of the deformed fringe image by using a phase shift method, wherein the calculation result of the phase is affected by the modulation degree.

And after obtaining the deformed fringe image, obtaining the phase of the deformed fringe image by using a phase shift method. According to the principle and formula of phase calculation, the modulation degree of the fringe image directly affects the accuracy and noise resistance of phase demodulation. Specifically, the magnitude of the modulation degree of the image affects the calculation result of the phase. The larger the modulation degree is, the higher the corresponding phase precision is; the smaller the modulation degree, the lower the corresponding phase accuracy. And the accuracy of the phase directly affects the accuracy of the matching points on the surface of the object.

In step S506, the modulation degree is affected by the change of the reflectivity of the surface of the object to be measured.

When the reflectivity of the surface of the object to be measured jumps, the corresponding modulation degree jumps accordingly. Further, when the modulation difference between adjacent pixels is large, the interference of phase calculation is enhanced under the influence of the modulation difference between the adjacent pixels, and the positioning accuracy of the object surface matching point is low.

Step S508, determining corresponding phase information according to the phase, and determining a first corresponding point of a point on the object surface on an imaging plane of a first camera and a second corresponding point on an imaging plane of a second camera in the fringe projection structured light three-dimensional imaging system according to the phase information and the geometric relationship.

When the two cameras collect points on the object to be measured, the points of the object to be measured form two corresponding points on the imaging surfaces of the two cameras. From the phase information and the geometrical optics principle, the two corresponding points are obtained, e.g. m in fig. 3_lAnd m_r。

Step S510, determining an intersection point of two straight lines according to the straight line connecting the first corresponding point and the center point of the first camera and the straight line connecting the second corresponding point and the center point of the second camera, wherein the intersection point of the two straight lines is a target point matched with the surface of the object.

Using triangulation, the first corresponding point m is shown in FIG. 3_lAnd a center point O of the first camera_lThe second corresponding point m_rAnd a second camera m_lThe central points of the two straight lines are respectively connected to obtain two crossed straight lines. According to the principle of geometric optics, the intersection point P of the two intersecting straight lines is the target point matched with the surface of the measured object.

Step S512, the determining process of the object surface matching target point is influenced by the deflection degree of the measured object surface and the fringe projection structured light three-dimensional imaging system, and the deflection degree of the measured object surface and the fringe projection structured light three-dimensional imaging system is an included angle between the normal direction of the target point and an angular bisector of two straight lines.

In step S510, the target point matched with the surface of the object to be measured is obtained by intersecting straight lines formed by corresponding points on the imaging surfaces of the two cameras and the center points of the cameras. As shown in fig. 7, when the included angle between the normal direction of the target point and the bisector of the two straight lines is large, the image information collected by the camera is poor, so that the obtained phase precision is low, and the confidence of the point cloud is low. That is, the higher the degree of deflection between the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system, the lower the confidence of the point cloud.

The embodiment is based on the principle that a fringe projection structured light three-dimensional imaging system generates three-dimensional points, obtains influence factors of point cloud confidence, and evaluates the quality of the point cloud. The quality of the point cloud obtained by the method is not interfered and limited by surrounding topological points.

The magnitude of the modulation affects the phase calculation. The larger the modulation degree is, the higher the corresponding phase precision is; the smaller the modulation degree, the lower the corresponding phase accuracy. And the accuracy of the phase directly affects the accuracy of the matching points on the surface of the object. Therefore, the modulation degree is used as an influence factor of the point cloud confidence degree, and in order to obtain the relationship between the modulation degree and the point cloud confidence degree, a Gaussian kernel function is used to establish the following relational expression:

wherein, B (x, y) is a modulation value at the coordinate (x, y), a is an influence degree factor, B is a modulation factor, and the values of a and B are set according to measuring equipment such as a fringe projection structured light three-dimensional imaging system and the like.

Fig. 8 is a schematic diagram of the modulation degree of the object to be measured, in which the higher the brightness, the higher the modulation degree corresponding to the region. Substituting the modulation value at each point into the above formula and calculating to obtain a result graph of the effect of the modulation degree on the confidence of the point cloud, as shown in fig. 9. The horizontal and numerical axes of the coordinates shown in fig. 9 represent the coordinate locations of the points on the measured object, and the paraxial axes are the confidence value references.

In this embodiment, a relationship between the modulation degree and the confidence of the point cloud is established by using a gaussian kernel function, so as to obtain the degree of influence of the modulation degree on the confidence of the point cloud. Compared with the traditional scheme, the method has the advantages that the high-reliability three-dimensional point and the low-reliability three-dimensional point are obtained by setting the modulation threshold value in half, and the influence of the modulation on the confidence coefficient of the point cloud can be more accurately expressed.

The modulation degree at the surface reflectivity change position of the measured object also jumps, namely if the modulation degree difference between adjacent pixels in the modulation degree image is large, the interference of phase calculation is enhanced under the influence of the modulation degree difference between the adjacent pixels, so that the positioning accuracy of the corresponding point is unreliable, and the obtained three-dimensional point has low quality and relatively small confidence coefficient. Therefore, the change of the surface reflectivity of the measured object is used as an influence factor of the confidence coefficient of the point cloud, namely, the difference value of the modulation degree of the three-dimensional point and the eight-neighborhood modulation degree of the three-dimensional point is used as the influence factor of the confidence coefficient of the point cloud. In order to obtain the relationship between the modulation degree value of the three-dimensional point and the difference value of the eight-neighborhood modulation degree and the confidence coefficient of the point cloud, a Gaussian kernel function is used for establishing the following relational expression:

v (x, y) is a difference value between a modulation degree value at a coordinate (x, y) and an eight-neighborhood modulation degree value of the coordinate (x, y), k is an influence degree factor, c is a modulation factor, and values of k and c are set according to measuring equipment such as a fringe projection structured light three-dimensional imaging system.

Substituting the modulation value at each point and the difference value between the eight-neighborhood modulation values into the formula and calculating to obtain a result graph of the effect of the surface reflectivity change of the measured object on the confidence coefficient of the point cloud, as shown in fig. 10.

In the embodiment, the confidence coefficient of the point cloud can be more accurately expressed by calculating the influence of the change of the reflectivity of the surface of the measured object on the modulation degree of the point cloud.

As shown in fig. 7, there are two groups of corresponding points on the left and right camera imaging planes: m is_l1、m _r1 and m _l2、m _r2, p1 and p2 are points on the measured object obj, and the calculation of the coordinates of p1 and p2 needs O_lm_l1、O_rm_r1 and O_lm_l2、O_rm_r2 the two rays intersect. As can be seen from FIG. 7, the tangent plane of point p1 is aligned with the fringe projection structured light three-dimensional imaging system, and ray O_lm_l1、O_rm_r1 can acquire better fringe information, and further can acquire more accurate phase value, and the tangent plane of the point p2 and the fringe projection structured light three-dimensional imaging system have larger deflection angle, so that the light ray O_rm_r2, the fringe information of the surface at the point p2 cannot be acquired well, so that the phase accuracy and reliability obtained by calculation are low, and the confidence of generating the p2 three-dimensional point coordinate is influenced. Therefore, the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system is used as an influence factor of the confidence coefficient of the point cloud, namely, the included angle between the normal direction of the three-dimensional point and the angular bisector of the two intersected rays is used as the influence factor of the confidence coefficient of the point cloud. In order to obtain the relationship between the included angle between the normal direction of the three-dimensional point and the angular bisector of the two intersecting rays and the confidence coefficient of the point cloud, a Gaussian kernel function is used to establish the following relational expression:

wherein the content of the first and second substances,

the method is characterized in that the method is an included angle between the normal direction of a corresponding three-dimensional point on an imaging surface (x, y) of a left camera (or a right camera) and an angular bisector of two intersected rays, m is an influence degree factor, n is a modulation factor, and the values of m and n are set according to measuring equipment such as a fringe projection structured light three-dimensional imaging system.

Substituting the included angle between the normal direction of each point and the angular bisector of the two intersecting rays into the formula and calculating to obtain an effect result graph of the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud, as shown in fig. 11.

In the embodiment, the confidence of the point cloud can be more accurately expressed by calculating the influence of the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the modulation degree of the point cloud.

The fringe projection structured light three-dimensional imaging system consists of two cameras and a projector, the cameras and the projector have field depth, and optical axes of the three devices have certain included angles. Therefore, the fringe projection structured light three-dimensional imaging system has an effective measurement space, and as shown in fig. 6, a hexagonal wire frame is an effective measurement space of the three-dimensional measurement device. Wherein CCD1 and CCD2 are respectively a left camera and a right camera, O_CLAnd O_CRIs the center point of the left and right cameras, X_CLAnd X_CLFor imaging planes corresponding to two cameras, Z_CLAnd Z_CRIs the optical axis of two cameras,/₆And l₈Forming the imaging angle of the left camera,/₅And l₇Forming an imaging angle, Z, of the right camera₀Is 1₅And l₈A, B is the intersection point of the optical axes of the two cameras and the surface of the object to be measured, theta is the half-imaging angle of the two cameras,

is the angle between the optical axis of the two-phase machine and the Z axis₁And l₂、l₃And l₄、l₉And l₁₀Are respectively two opposite sides of the upper surface of the hexagon, and Delta L is L₁And l₂、l₃And l₄L is L₉And l₁₀The distance between two opposite edges of the hexagonal side edge is W, and the effective measurement space of the cuboid on the right side of the figure 6 can be deduced by utilizing the geometric principle. The confidence of the point cloud in the effective measurement space is higher than that of the point cloud outside the effective measurement space, and the closer to the effective measurement space, the higher the confidence. Therefore, whether the point cloud is in the effective measurement space range of the fringe projection structured light three-dimensional imaging system or not is determined, namely, the distance between a measured object and the effective measurement space is used as an influence factor of the confidence coefficient of the point cloud. In order to obtain the relationship between the distance from the measured object to the effective measurement space and the confidence coefficient of the point cloud, a Gaussian kernel function is applied to establish the following relational expression:

d (x, y) is the distance between a three-dimensional point corresponding to the position (x, y) on the imaging surface of the left camera (or the right camera) and the center of an effective measurement space, h is an influence degree factor, l is a modulation factor, and the values of h and l are set according to measuring equipment such as a fringe projection structured light three-dimensional imaging system.

In practice, due to the low confidence of the points outside the effective measurement space, the points outside the effective measurement space are usually directly ignored, i.e., the points outside the effective measurement space are not considered in the three-dimensional reconstruction process.

In the embodiment, the influence of the partial effective measurement space of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the modulation degree of the point cloud is calculated, so that the confidence coefficient of the point cloud can be more accurately expressed.

In one embodiment, the influence degrees of the above influence factors on the point cloud confidence are combined, different weights are respectively given according to the influence degrees, and the following comprehensive action relational expression is established:

where C (x, y) is the confidence of the corresponding point at the (x, y) position on the left (right) camera imaging plane, ω_B、ω_v、

ω_DRespectively the action weights of the modulation factor, the modulation change factor, the included angle factor and the distance factor on the point cloud modulation; and the number of the first and second electrodes,

wherein, ω is_B、ω_v、

ω_DIs determined empirically from actual measurements. Substituting the influence degrees of the influence factors on the point cloud confidence degrees into the relational expression to obtain a final comprehensive action result, as shown in fig. 12.

In the embodiment, the degree of influence of each influence factor on the confidence coefficient of the point cloud is comprehensively considered, and different weights are given according to the degree of influence, so that the confidence coefficient of the point cloud can be more accurately expressed.

It should be noted that the comprehensive relational expression in the present embodiment is exemplified by four influencing factors. In other embodiments, the relationship may be a comprehensive relationship formed by combining two or three influencing factors. But the point cloud confidence obtained by comprehensively considering all the influencing factors is the most accurate.

In one embodiment, the resulting point cloud confidence is participated in a global closest point iterative optimization, comprising:

step S1302, a point cloud set under each visual angle is obtained by the three-dimensional scanner performing multi-directional multi-angle scanning on the measured object.

Because the actual point cloud data acquisition process is influenced by factors such as the size of the measured object, the self-shielding problem of the object surface, the view field range and the like, all geometric information of the three-dimensional object cannot be obtained by one-time scanning. Therefore, the measured object needs to be scanned in multiple directions and multiple angles, three-dimensional data of the measured object is collected, and a set of complete three-dimensional models is further spliced.

Step S1304, according to the coincidence degree of the point clouds at all the visual angles, screening the target point clouds corresponding to all the visual angles from the point cloud set at all the visual angles, wherein the average value of the confidence degrees of the target point clouds is the highest.

When the measured object is scanned in multiple directions and angles, point clouds of the measured object under multiple visual angles can be obtained, and the point clouds are highly overlapped. However, in the process of point cloud matching and stitching, all point clouds do not need to be stitched, so point clouds with high confidence coefficient need to be selected from a plurality of point clouds for matching and fusing.

Step 1306, according to the coincidence degree of the point clouds at all the visual angles, screening the target point clouds corresponding to all the visual angles from the point cloud set at all the visual angles, wherein the average value of the confidence degrees of the target point clouds is the highest.

The target point cloud is a point cloud with high confidence. Before matching and fusing the point clouds with high reliability, the position relation between each piece of point cloud needs to be determined, and then the splicing can be better realized.

Step S1308, obtaining the closest point in the overlap region of each target point cloud by using a forward projection method according to the position relationship between each target point cloud.

And projecting the point clouds after the position relation is determined on the same plane by using a forward projection method, and obtaining the closest point in the overlapping area of each target point cloud according to the projection result.

Step S1310, performing global closest point iterative optimization according to the closest point in each point cloud overlapping region, and using the point cloud confidence as a weight factor of global matching optimization to participate in the global closest point iterative optimization, where the weight factor of global matching optimization is an average of the sums of the confidences of each pair of closest points.

Substituting all the closest points into a closest point iterative optimization formula, taking the average value of the confidence sum of each pair of the closest points as a weight factor to participate in the formula, and distinguishing and processing the influence of each pair of the closest points on the global optimization result, so that the global optimization function of the points with low confidence is reduced, and the optimized acting force of the closest points with high confidence is enhanced. The final optimization formula is as follows:

wherein R is_ij、T_ijIs a rigid body transformation relation between the ith point cloud and the jth point cloud, p_i、q_jAs a set of matching points for two point clouds, W_pq＝(C_p+C_q)/2，C_p、C_qThe confidence levels of points p, q, respectively.

And step S1312, obtaining a point cloud matching result according to the global closest point iterative optimization result.

In general, when the average value of the distances of all the closest points reaches 0.02mm (millimeter), the iteration is stopped, and the point cloud matching result is output. As shown in fig. 14, the left side of the graph is a matching result graph after the point cloud confidence coefficient participates in the global optimization, and the right side of the graph is a matching result graph without the point cloud confidence coefficient participating in the global optimization. It can be seen that the accuracy of the left side of the figure is higher.

In the embodiment, the point cloud confidence is participated in global optimization, and the influence of each pair of closest points on the global optimization result is processed in a distinguishing manner, so that the global optimization effect of the points with low confidence is reduced, and the optimization acting force of the closest points with high confidence is enhanced. And finally, a matching result with higher precision is obtained, as shown in fig. 14.

In the embodiment, only the obtained point cloud confidence coefficient is used for evaluating the quality of the point cloud, and the point cloud confidence coefficient participates in global optimization to improve the point cloud matching precision. It should be noted that the applications of the point cloud confidence include, but are not limited to, the above two.

It should be understood that, although the steps in the flowcharts of fig. 2, 5 and 13 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. 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, 5, and 13 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 15, there is provided a confidence evaluation apparatus 1500 of three-dimensional depth data, including: an obtaining module 1502, an influence degree calculating module 1504, a relationship establishing module 1506, a confidence degree calculating module 1508, and a quality evaluating module 1510, wherein:

the obtaining module 1502 is configured to obtain influence factors of the point cloud confidence including at least one of a modulation degree, a jump in reflectivity of a surface of an object to be measured, a deflection degree between the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system, and an effective measurement space range of the fringe projection structured light three-dimensional imaging system in a process of determining a target point matched with the surface of the object by the fringe projection structured light three-dimensional imaging system;

the influence degree calculation module 1504 is used for establishing the relationship between each influence factor and the point cloud confidence coefficient, and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relationship between each influence factor and the point cloud confidence coefficient;

the relationship establishing module 1506 is used for endowing different influence weights to the influence factors according to the influence degrees of the influence factors on the point cloud confidence degrees, and establishing the relationship between the comprehensive result of the influence factors and the point cloud confidence degrees;

the confidence coefficient calculation module 1508 is configured to obtain a confidence coefficient of the point cloud according to a relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud;

and the quality evaluation module 1510 is configured to evaluate the quality of the point cloud according to the confidence of the point cloud to obtain a point cloud quality evaluation result.

In one embodiment, acquisition module 1502 includes: acquiring a standard sine stripe image, and projecting the standard sine stripe image to the surface of an object in a stripe projection structured light three-dimensional imaging system; acquiring a projection image of the surface of an object to obtain a corresponding deformed fringe image, wherein the acquisition result of the projection image of the surface of the object is influenced by an effective measurement space range in a fringe projection structured light three-dimensional imaging system; calculating the phase of the deformed fringe image by using a phase shift method, wherein the calculation result of the phase is influenced by the modulation degree, and the modulation degree is influenced by the change of the surface reflectivity of the measured object; determining corresponding phase information according to the phase, and determining a first corresponding point of a point on the surface of the object on an imaging surface of a first camera and a second corresponding point on an imaging surface of a second camera in the fringe projection structured light three-dimensional imaging system according to the phase information and the geometric relation; determining the intersection point of the two straight lines according to the straight line connecting the first corresponding point and the central point of the first camera and the straight line connecting the second corresponding point and the central point of the second camera, wherein the intersection point of the two straight lines is a target point matched with the surface of the object; the determining process of the target point is influenced by the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system, and the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system is an included angle between the normal direction of the target point and an angular bisector of two straight lines.

In one embodiment, the influence degree calculation module 1504 includes: establishing a relation between the modulation value and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the modulation value and the quality of the point cloud; and obtaining the influence degree of the modulation degree on the confidence coefficient of the point cloud according to the relation between the modulation degree and the confidence coefficient of the point cloud.

In one embodiment, the influence degree calculation module 1504 includes: establishing a relation between the change of the reflectivity of the surface of the measured object and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the modulation difference value of the reflectivity change of the surface of the measured object and the quality of the point cloud; and obtaining the influence degree of the change of the surface reflectivity of the measured object on the confidence coefficient of the point cloud according to the relation between the change of the surface reflectivity of the measured object and the confidence coefficient of the point cloud.

In one embodiment, the influence degree calculation module 1504 includes: establishing a relation between the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud by using a Gaussian kernel function according to the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the quality of the point cloud; and obtaining the influence degree of the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud according to the relation between the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud.

In one embodiment, the influence degree calculation module 1504 includes: establishing a relation between the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud by applying a Gaussian kernel function according to the effective measurement space range of the fringe projection structured light three-dimensional imaging system and the quality of the point cloud; and obtaining the influence degree of the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud according to the relationship between the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud.

In one embodiment, the confidence evaluation apparatus 1500 of three-dimensional depth data further includes: acquiring a point cloud set under each visual angle, which is obtained by carrying out multi-directional and multi-angle scanning on a measured object by a three-dimensional scanner; screening target point clouds corresponding to all the visual angles from the point cloud set under all the visual angles according to the coincidence degree of the point clouds under all the visual angles, wherein the average value of confidence degrees of the target point clouds is the highest; determining the position relation between each piece of target point cloud according to the target point cloud; obtaining the closest point in the overlapping area of each target point cloud by using a forward projection method; performing global closest point iterative optimization on the closest point in each point cloud overlapping region, and taking the point cloud confidence coefficient as a weight factor of global matching optimization to participate in the global closest point iterative optimization, wherein the weight factor of the global matching optimization is an average value of the sum of the confidence coefficients of each pair of closest points; and obtaining a point cloud matching result according to the global closest point iterative optimization result.

For the specific definition of the confidence evaluation device for three-dimensional depth data, reference may be made to the above definition of the confidence evaluation method for three-dimensional depth data, and details are not repeated here. The modules in the confidence evaluation device for three-dimensional depth data 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, and its internal structure diagram may be as shown in fig. 16. The computer device includes a processor, a memory, a network 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 network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a point cloud confidence evaluation method. 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. 16 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, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program: in the process of determining a target point matched with the surface of an object by a fringe projection structured light three-dimensional imaging system, obtaining influence factors of point cloud confidence, including at least one of modulation degree, jump of reflectivity of the surface of the object to be measured, deflection degree of the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system, and effective measurement space range of the fringe projection structured light three-dimensional imaging system; establishing a relation between each influence factor and the point cloud confidence coefficient, and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relation between each influence factor and the point cloud confidence coefficient; according to the influence degree of each influence factor on the point cloud confidence coefficient, different influence weights are given to each influence factor, and the relation between the comprehensive result of each influence factor and the point cloud confidence coefficient is established; obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud; and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

In one embodiment, in the process of determining a target point matched with the surface of an object in the structured light three-dimensional imaging system through fringe projection, the influence factor of the point cloud confidence coefficient is obtained by: acquiring a standard sine stripe image, projecting the standard sine stripe image to the surface of an object in a stripe projection structured light three-dimensional imaging system, and acquiring a projection image of the surface of the object to obtain a deformed stripe image corresponding to the standard sine stripe image, wherein the acquisition result of the projection image of the surface of the object is influenced by the effective measurement space range in the stripe projection structured light three-dimensional imaging system; calculating the phase of the deformed fringe image by using a phase shift method, wherein the calculation result of the phase is influenced by the modulation degree; the modulation degree is influenced by the change of the reflectivity of the surface of the measured object; determining corresponding phase information according to the phase, and determining a first corresponding point of a point on the surface of the object on an imaging surface of a first camera and a second corresponding point on an imaging surface of a second camera in the fringe projection structured light three-dimensional imaging system according to the phase information and the geometric relation; determining the intersection point of the two straight lines according to the straight line connecting the first corresponding point and the central point of the first camera and the straight line connecting the second corresponding point and the central point of the second camera, wherein the intersection point of the two straight lines is a target point matched with the surface of the object; the determining process of the object surface matching target point is influenced by the deflection degree of the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system, and the deflection degree of the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system is an included angle between the normal direction of the target point and the angular bisector of the two straight lines.

In one embodiment, establishing a relationship between each influence factor and the point cloud confidence, and obtaining the influence degree of each influence factor on the point cloud confidence according to the relationship between each influence factor and the point cloud confidence comprises: establishing a relation between the modulation degree and the point cloud confidence coefficient by using a Gaussian kernel function according to the modulation degree and the quality of the point cloud to obtain the influence degree of the modulation degree on the point cloud confidence coefficient; establishing a relation between the change of the surface reflectivity of the measured object and the confidence coefficient of the point cloud by using a Gaussian kernel function according to the modulation difference value of the surface reflectivity change of the measured object and the quality of the point cloud, and obtaining the influence degree of the change of the surface reflectivity of the measured object on the confidence coefficient of the point cloud; according to the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the quality of the point cloud, establishing a relation between the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud by using a Gaussian kernel function, and obtaining the influence degree of the deflection degree of the surface of the measured object and the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud; according to the effective measurement space range of the fringe projection structured light three-dimensional imaging system and the quality of the point cloud, a Gaussian kernel function is used to establish the relationship between the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system and the confidence coefficient of the point cloud, and the influence degree of the distance from the point cloud to the effective measurement space center of the fringe projection structured light three-dimensional imaging system on the confidence coefficient of the point cloud is obtained.

In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a point cloud set under each visual angle, which is obtained by carrying out multi-directional and multi-angle scanning on a measured object by a three-dimensional scanner; screening target point clouds corresponding to all the visual angles from the point cloud set under all the visual angles according to the coincidence degree of the point clouds under all the visual angles, wherein the average value of confidence degrees of the target point clouds is the highest; obtaining the closest point in the overlapping area of each target point cloud by using a forward projection method according to the position relation between each target point cloud; and performing global closest point iterative optimization according to the closest point in each point cloud overlapping region, and taking the point cloud confidence as a weight factor of global matching optimization to participate in the global closest point iterative optimization, wherein the weight factor of the global matching optimization is the average value of the sum of the confidence of each pair of closest points.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: in the process of determining a target point matched with the surface of an object by a fringe projection structured light three-dimensional imaging system, obtaining influence factors of point cloud confidence, including at least one of modulation degree, jump of reflectivity of the surface of the object to be measured, deflection degree of the surface of the object to be measured and the fringe projection structured light three-dimensional imaging system, and effective measurement space range of the fringe projection structured light three-dimensional imaging system; establishing a relation between each influence factor and the point cloud confidence coefficient, and obtaining the influence degree of each influence factor on the point cloud confidence coefficient according to the relation between each influence factor and the point cloud confidence coefficient; according to the influence degree of each influence factor on the point cloud confidence coefficient, different influence weights are given to each influence factor, and the relation between the comprehensive result of each influence factor and the point cloud confidence coefficient is established; obtaining the confidence coefficient of the point cloud according to the relationship between the comprehensive result of each influence factor and the confidence coefficient of the point cloud; and evaluating the quality of the point cloud according to the confidence coefficient of the point cloud to obtain a point cloud quality evaluation result.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

29页详细技术资料下载

上一篇：一种医用注射器针头装配设备

下一篇：基于激光点云的输电线路导地线建模提取方法

Confidence evaluation method and device for three-dimensional depth data and computer equipment

相关技术

网友询问留言