Optical rotation scanning measuring instrument three-dimensional point cloud denoising method based on point neighborhood distance clustering

Abstract

The invention discloses an optical rotation scanning measuring instrument three-dimensional point cloud denoising method based on point neighborhood distance clustering, and the method comprises the steps: calculating the distance between each point and a neighborhood point according to the to-be-processed point cloud data, carrying out the clustering division according to the feature, obtaining a plurality of groups, keeping the group with the maximum number of points, and removing the other groups, therefore, the denoising purpose is achieved, and the target point cloud data is obtained. According to the denoising method based on the point neighborhood distance clustering, the point clouds are divided into a plurality of groups according to the neighborhood distance relation between the points, the group with the most point clouds is extracted, the other groups are all regarded as noise filtering, and the denoised target point clouds can be obtained through one-time filtering. According to the invention, the time required by the execution of the denoising algorithm is reduced, the denoising efficiency of the algorithm is effectively improved, and a denoised result can be obtained in a relatively short time even if the number of point clouds is expanded to a great extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical measurement, and in particular, relates to a three-dimensional point cloud denoising method of an optical rotating scanning measuring instrument based on point neighborhood distance clustering. Background Art

[0002] Key components such as engine blisks and sealing grooves in the aerospace industry, transmission valve bodies and engine cylinders in the automotive industry, and guns and barrels in the defense industry have the characteristics of complex internal structure, narrow space, and large aspect ratio. The current difficulties in the precise measurement of key geometric parameters of the above components are "unable to measure", "inaccurate measurement" and "slow measurement".

[0003] The optical rotary scanning measurement system is a modular, non-contact optical hole measurement system. The measurement system has a slender probe structure with optical fiber and reflector inside, which can adapt to the size and shape limitations of the narrow space of the measured part to measure the surface of the object.

[0004] When performing shape detection in a narrow space, the rotating motor drives the probe rod to rotate, scanning the object to be measured in a non-contact manner, obtaining a large number of three-dimensional point clouds located inside the narrow space, thereby calculating the geometric dimensions and shape tolerances of the object to be measured, and achieving high-precision three-dimensional measurement.

[0005] Due to installation deviations during the actual processing and assembly of parts, the actual light-emitting point and direction of the probe rod do not coincide with the ideal measurement coordinate system; at the same time, the error movement of the rotating probe causes the measurement coordinates to be unable to accurately reflect the actual measurement value. These problems lead to a lot of noise interference in the final measured point cloud data, and these noises are not simply the same type of noise, but a composite type of noise mixed by multiple types of noise, which interferes with the subsequent calculation of geometric dimensions and shape tolerances.

[0006] Traditional denoising algorithms basically filter out a certain type of noise point cloud. It is difficult to effectively remove this complex type of noise. Therefore, a denoising algorithm that can effectively remove this complex type of noise is needed.

[0007] Based on the above-mentioned deficiencies in the prior art, the present invention is proposed. Summary of the invention

[0008] In order to solve the problems raised by the background technology, the present invention proposes a 3D point cloud denoising method for an optical rotary scanning measuring instrument based on point neighborhood distance clustering. The method calculates the distance between each point and the neighborhood point according to the point cloud data to be processed, and performs clustering division according to this feature to obtain several groups, retains the group with the largest number of points, and removes the remaining groups, thereby achieving the purpose of denoising and obtaining the target point cloud data.

[0009] In order to achieve the purpose of the invention, the technical solution provided by the present invention is:

[0010] A 3D point cloud denoising method for an optical rotating scanning measuring instrument based on point neighborhood distance clustering comprises the following steps:

[0011] Step S1, the optical rotation scanning measuring instrument system obtains a large number of three-dimensional coordinates of points located inside a narrow space during the measurement process, and finally generates point cloud data;

[0012] Step S2, traverse each point in the point cloud to be processed in turn, calculate the distance value between the current point and the next point, and obtain an array distances consisting of the distance values ​​between the two points;

[0013] Step S3, traversing the distance array distances, and determining the parameter parameter for subsequent grouping according to the distance information of the entire point cloud;

[0014] Step S4, traversing the distance array distances again, and further dividing the array into a plurality of groups group according to the parameter parameter determined in step S3, each group consisting of a plurality of distance information;

[0015] Step S5, traverse all the groups in step S4 and find the group with the largest number of members;

[0016] Traverse the groups group in the group set groups, find the group with the largest number of elements, and record it as the maximum group MaxGroup;

[0017] Step S6, obtaining the subscripts of the initial element and the last element in the group with the largest number of members found in step S5;

[0018] Get the subscript BeginIndex of the initial element and the subscript EndIndex of the last element in the maximum group MaxGroup found in step S5; BeginIndex is the starting subscript of the target point cloud in the initial point cloud, recorded as minIndex; EndIndex+1 is the ending subscript of the target point cloud in the initial point cloud, recorded as maxIndex, and get the subscript range [minIndex, maxIndex] of the target point cloud in the initial point cloud;

[0019] Step S7, obtaining the subscript range of the point cloud to be retained according to the two subscripts obtained in step S6, filtering out the point cloud outside the subscript range, obtaining the target point cloud, realizing point cloud filtering, and completing point cloud denoising.

[0020] A point cloud index object is established according to the subscript range [minIndex, maxIndex] determined in step S6, and the extractor object is applied to retain the point cloud in the index object and filter out the remaining noise point clouds to achieve a denoising effect.

[0021] The preferred technical solution provided by the present invention is:

[0022] The step S2 specifically includes:

[0023] Step S21: Calculate the distance between the current point and the next point, as follows:

[0024]

[0025] Where x i ,y i 、z i is the three-dimensional coordinate of the current traversal point, x i+1 ,y i+1 、z i+1 is the three-dimensional coordinate of the subsequent adjacent point of the current traversal point; dis is the calculated distance between the two points;

[0026] Step S22: The distance values ​​obtained in step S21 are sequentially stored in an array to obtain an array distances consisting of the distance values ​​between two points. The subscript i in the array is consistent with the subscript i of the point during traversal.

[0027] The further preferred technical solution provided by the present invention is:

[0028] The step S3 is specifically as follows:

[0029] Step S31: traverse the distances array and calculate the expected distance values ​​in the array, as follows:

[0030]

[0031] average is the desired expectation, n is the total number of elements in the array, i is the i-th distance value in the array;

[0032] Step S32: The expected multiple is used as a parameter parameter to measure the change range of the distance value, as follows:

[0033] parameter = k × average (3)

[0034] Where k is a multiple.

[0035] The more preferred technical solution provided by the present invention is:

[0036] In the step S4, specifically:

[0037] Step S41: traverse the distances array again, and form a set of key-value pairs with the distance value of each traversed element and its current index, as follows:

[0038] pair=[i,dis i ] (4)

[0039] i is the subscript of the current element, dis i Represents the distance value of the current element, and pair is the key-value pair they consist of.

[0040] Step S42: put the current key-value pair into the group group, and compare the difference between the current distance value and the next distance value, as follows:

[0041] difference=|dis i -dis i+1 | (5)

[0042] dis i Represents the distance value of the i-th element in the array, and difference represents the absolute value of the difference between two adjacent distances;

[0043] Step S43: compare the difference with the parameter parameter. If the difference is greater than the parameter, store the current group into the group set groups; otherwise, do not execute;

[0044] Step S44: repeat steps S41 to S43, and finally divide the data into several groups according to the distance values, and store them in the group set groups.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. Existing point cloud denoising algorithms mostly denoise single-type noise points. When there are complex types of noise points, it is difficult to effectively remove all of these complex types of noise, and denoising is often incomplete. The denoising method based on point neighborhood distance clustering proposed in the present invention divides the point cloud into several groups according to the neighborhood distance relationship between points, extracts the group with the largest number of point clouds, and the remaining groups are treated as noise filters. After one filtering, the denoised target point cloud can be obtained.

[0047] 2. The time complexity of common point cloud denoising algorithms is mostly between O(n*logN) and O(n 2), when the number of point clouds is large to a certain extent, the time required for the algorithm to execute is very long. The present invention fully considers the point-by-point generation characteristics of the point cloud data obtained by the rotating scanning side head system. According to this characteristic, the point cloud denoising process is customized to control the time complexity of the algorithm to O(n), reducing the time required for the execution of the denoising algorithm and effectively improving the algorithm denoising efficiency. Even when the number of point clouds expands to a large extent, the denoised result can be obtained in a relatively fast time. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a schematic diagram of the structure of the rotary scanning probe system used in the present invention;

[0049] Figure 2 for Figure 1 Schematic diagram of the point cloud data to be processed finally obtained by the scanning system;

[0050] Figure 3 It is a flowchart of the denoising method based on point neighborhood distance clustering of the present invention;

[0051] Figure 4 Schematic diagram of the group division process based on distance parameters. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] The point cloud data targeted by the present invention is Figure 1 The optical rotation scanning measurement system shown is measured. Figure 1 The structure diagram of the rotary scanning probe system used in the present invention is as follows: Figure 1As shown, the optical rotation scanning measurement system used in the present invention includes but is not limited to a one-dimensional displacement stage, a rotating probe and a grating sensor. When performing three-dimensional morphology detection inside a narrow space, the measuring light beam is emitted by the absolute distance measuring device through the optical fiber probe, and is emitted along the radial direction of the probe rod through the reflector and contacts the surface of the object to be measured. Then the light beam returns along the same path, passes through the reflector and the optical fiber probe and finally returns to the absolute distance measuring device. Through this process, the system can measure the distance from the light-emitting point of the probe rod to the surface of the object to be measured (denoted as d). At the same time, the one-dimensional displacement of the rotating probe is measured by the long grating sensor (denoted as l), and the angular rotation is measured by the angle sensor (denoted as θ). Based on these data, the system can obtain a large number of three-dimensional coordinates of points located inside the narrow space during the measurement process, and finally generate point cloud data. The point cloud data obtained by the system has the characteristic of point-by-point generation, and the present invention is proposed based on this characteristic.

[0054] In order to solve the problems raised by the background technology, the present invention proposes a 3D point cloud denoising method for an optical rotary scanning measuring instrument based on point neighborhood distance clustering. The method first calculates the distance between each point and the neighborhood point according to the point cloud data to be processed, and performs clustering division according to this feature to obtain several groups, retains the group with the largest number of points, and removes the remaining groups, thereby achieving the purpose of denoising and obtaining the target point cloud data.

[0055] The specific steps include:

[0056] Step S1, using Figure 1 The optical rotary scanning measuring instrument system shown obtains the three-dimensional coordinates of a large number of points located inside a narrow space during the measurement process, and finally generates point cloud data;

[0057] The point cloud data to be processed finally obtained by the scanning system is as follows: Figure 2 As shown, there are complex types of noise points, one is sparse noise points, and the other is dense noise points. The target point cloud is surrounded by these complex types of noise points. For sparse noise points, traditional denoising algorithms can remove them better. However, the density and other characteristics of dense noise points are similar to those of the target point cloud and are difficult to be effectively removed. The denoising algorithm based on neighborhood distance clustering of the present invention can remove all these complex types of noise around the target point cloud at one time to obtain a complete target point cloud.

[0058] Figure 3 This is a flowchart of the denoising method based on point neighborhood distance clustering of the present invention; Figure 3 As shown,

[0059] Step S2, traverse each point in the point cloud to be processed in turn, calculate the distance value between the current point and the next point, and obtain an array distances consisting of the distance values ​​between the two points;

[0060] Step S21: Calculate the distance between the current point and the next point, as follows:

[0061]

[0062] Where x i ,y i 、z i is the three-dimensional coordinate of the current traversal point, x i+1 ,y i+1 、z i+1 is the three-dimensional coordinate of the subsequent adjacent point of the current traversal point; dis is the calculated distance between the two points;

[0063] Step S22: The distance values ​​obtained in step S21 are sequentially stored in an array to obtain an array distances consisting of the distance values ​​between two points. The subscript i in the array is consistent with the subscript i of the point during traversal.

[0064] Step S3, traversing the distance array distances, and determining the parameter parameter for subsequent grouping according to the distance information of the entire point cloud;

[0065] Step S31: traverse the distances array and calculate the expected distance values ​​in the array, as follows:

[0066]

[0067] average is the desired expectation, n is the total number of elements in the array, i is the i-th distance value in the array;

[0068] Step S32: The expected multiple is used as a parameter parameter to measure the change range of the distance value, as follows:

[0069] parameter = k × average (3)

[0070] Where k is a multiple.

[0071] Step S4, traversing the distance array distances again, and further dividing the array into a plurality of groups group according to the parameter parameter determined in step S3, each group consisting of a plurality of distance information;

[0072] Step S41: traverse the distances array again, and form a set of key-value pairs with the distance value of each traversed element and its current index, as follows:

[0073] pair=[i,dis i] (4)

[0074] i is the subscript of the current element, dis i Represents the distance value of the current element, and pair is the key-value pair they consist of.

[0075] Step S42: put the current key-value pair into the group group, and compare the difference between the current distance value and the next distance value, as follows:

[0076] difference=|dis i -dis i+1 | (5)

[0077] dis i Represents the distance value of the i-th element in the array, and difference represents the absolute value of the difference between two adjacent distances;

[0078] Step S43: compare the difference with the parameter parameter. If the difference is greater than the parameter, store the current group into the group set groups; otherwise, do not execute;

[0079] Step S44: repeat steps S41 to S43, and finally divide the data into several groups according to the distance values, and store them in the group set groups.

[0080] Step S5, traverse all the groups in step S4 and find the group with the largest number of members;

[0081] Traverse the groups group in the group set groups, find the group with the largest number of elements, and record it as the maximum group MaxGroup;

[0082] Step S6, obtaining the subscripts of the initial element and the last element in the group with the largest number of members found in step S5;

[0083] Get the subscript BeginIndex of the initial element and the subscript EndIndex of the last element in the maximum group MaxGroup found in step S5; BeginIndex is the starting subscript of the target point cloud in the initial point cloud, recorded as minIndex; EndIndex+1 is the ending subscript of the target point cloud in the initial point cloud, recorded as maxIndex, and get the subscript range [minIndex, maxIndex] of the target point cloud in the initial point cloud;

[0084] Step S7, obtaining the subscript range of the point cloud to be retained according to the two subscripts obtained in step S6, filtering out the point cloud outside the subscript range, obtaining the target point cloud, realizing point cloud filtering, and completing point cloud denoising.

[0085] A point cloud index object is established according to the subscript range [minIndex, maxIndex] determined in step S6, and the extractor object is applied to retain the point cloud in the index object and filter out the remaining noise point clouds to achieve a denoising effect.

[0086] Although the functions and working processes of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific functions and working processes. The above-mentioned specific implementation methods are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all within the protection of the present invention.

Claims

1. A 3D point cloud denoising method for an optical rotation scanning measuring instrument based on point neighborhood distance clustering, characterized in that: The following steps are included: Step S1, the optical rotation scanning measuring instrument system obtains a large number of three-dimensional coordinates of points located inside a narrow space during the measurement process, and finally generates point cloud data; Step S2, traverse each point in the point cloud to be processed in turn, calculate the distance value between the current point and the next point, and obtain an array distances consisting of the distance values ​​between the two points; Step S3, traversing the distance array distances, and determining the parameter parameter for subsequent grouping according to the distance information of the entire point cloud; Step S4, traversing the distance array distances again, and further dividing the array into a plurality of groups group according to the parameter parameter determined in step S3, each group consisting of a plurality of distance information; Step S5, traverse all the groups in step S4 and find the group with the largest number of members; Traverse the groups group in the group set groups, find the group with the largest number of elements, and record it as the maximum group MaxGroup; Step S6, obtaining the subscripts of the initial element and the last element in the group with the largest number of members found in step S5; Get the subscript BeginIndex of the initial element and the subscript EndIndex of the last element in the maximum group MaxGroup found in step S5; BeginIndex is the starting subscript of the target point cloud in the initial point cloud, recorded as minIndex; EndIndex+1 is the ending subscript of the target point cloud in the initial point cloud, recorded as maxIndex, and get the subscript range [minIndex, maxIndex] of the target point cloud in the initial point cloud; Step S7, obtaining the subscript range of the point cloud to be retained according to the two subscripts obtained in step S6, filtering out the point cloud outside the subscript range, obtaining the target point cloud, implementing point cloud filtering, and completing point cloud denoising; A point cloud index object is established according to the subscript range [minIndex, maxIndex] determined in step S6, and the extractor object is applied to retain the point cloud in the index object and filter out the remaining noise point clouds to achieve a denoising effect.

2. The method for denoising a three-dimensional point cloud of an optical rotating scanning measuring instrument based on point neighborhood distance clustering according to claim 1, characterized in that: The step S2 specifically includes: Step S21: Calculate the distance between the current point and the next point, as follows: Where x i ,y i 、z i is the three-dimensional coordinate of the current traversal point, x i+1 ,y i+1 、z i+1 is the three-dimensional coordinate of the subsequent adjacent point of the current traversal point; dis is the calculated distance between the two points; Step S22: The distance values ​​obtained in step S21 are sequentially stored in an array to obtain an array distances consisting of the distance values ​​between two points. The subscript i in the array is consistent with the subscript i of the point during traversal.

3. The method for denoising a three-dimensional point cloud of an optical rotation scanning measuring instrument based on point neighborhood distance clustering according to claim 1, characterized in that: The step S3 is specifically as follows: Step S31: traverse the distances array and calculate the expected distance values ​​in the array, as follows: average is the desired expectation, n is the total number of elements in the array, i is the i-th distance value in the array; Step S32: The expected multiple is used as a parameter parameter to measure the change range of the distance value, as follows: parameter = k × average (3) Where k is a multiple.

4. The method for denoising a three-dimensional point cloud of an optical rotation scanning measuring instrument based on point neighborhood distance clustering according to claim 1, characterized in that: In the step S4, specifically: Step S41: traverse the distances array again, and form a set of key-value pairs with the distance value of each traversed element and its current index, as follows: pair=[i,dis i ] (4) i is the subscript of the current element, dis i Represents the distance value of the current element, and pair is the key-value pair they consist of. Step S42: put the current key-value pair into the group group, and compare the difference between the current distance value and the next distance value, as follows: difference=|dis i -dis i+1 | (5) dis i Represents the distance value of the i-th element in the array, and difference represents the absolute value of the difference between two adjacent distances; Step S43: compare the difference with the parameter parameter. If the difference is greater than the parameter, store the current group into the group set groups; otherwise, do not execute; Step S44: repeat steps S41 to S43, and finally divide the data into several groups according to the distance values, and store them in the group set groups.

Images