CN106949848B - A kind of high-precision laser 3D profile phone structural detection method - Google Patents
A kind of high-precision laser 3D profile phone structural detection method Download PDFInfo
- Publication number
- CN106949848B CN106949848B CN201710153827.XA CN201710153827A CN106949848B CN 106949848 B CN106949848 B CN 106949848B CN 201710153827 A CN201710153827 A CN 201710153827A CN 106949848 B CN106949848 B CN 106949848B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- point
- laser
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 52
- 238000005259 measurement Methods 0.000 claims abstract description 40
- 238000005070 sampling Methods 0.000 claims abstract description 24
- 238000013499 data model Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 62
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000005477 standard model Effects 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 5
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000004065 semiconductor Substances 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012952 Resampling Methods 0.000 claims description 2
- 230000003760 hair shine Effects 0.000 claims description 2
- 238000005457 optimization Methods 0.000 claims description 2
- 238000010998 test method Methods 0.000 claims description 2
- 230000007547 defect Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of high-precision laser 3D profile phone structural detection methods, comprising: step 1) uses the previously-scanned sample once of laser measuring apparatus, sets the sweep parameter of sample, including laser power, image exposuring time, acquisition range acquire the laser scanning image sample of sample；Step 2) is placed on phone structural to mobile work platform, and for fixed laser measuring head in surface, servo motor drives workbench supporting plate mobile, detection is completed, system completes surface profile detection, sampling number St=pl/sstep in 2 seconds, pl is product measurement length, and sstep is sampling step length；Step 3) establishes profile standard data model；3D MODEL C AD data are imported before step 4) measurement, data are made of thousands of to a tri patch up to ten thousand, along the slice map etc. of the path interception 3D model vertical with scanning direction.
Description
Technical Field
The invention belongs to a high-precision laser 3D contour mobile phone structural member detection method.
Background
With the development of mobile terminals (mobile phones), the quality requirements of mobile phone structural members are higher and higher. Most manufacturers for detecting the mobile phone structural parts in the past mainly adopt visual observation, and have high subjectivity and difficult quality stabilization. A few enterprises in the industry use machine vision methods to detect mobile phone structural parts, such as Huacheng technology limited company, Samsung company and the like, and some testing equipment is also deployed on a mobile phone production line, the mainstream of the testing equipment and a detection system still adopts a 2D vision technology, light sources at different angles are used for shooting outlines, and then an image processing algorithm is used for analysis and processing. The 2D vision method has large data disturbance, and the small changes of the illumination intensity, the material change and the light source angle can generate great influence on the data, so that the detection result is inaccurate. Some AOI manufacturers use a statistical machine learning method and a three-color LED multi-angle light source to manufacture image detection equipment, so that the defects can be detected to a certain extent, and the defects in the aspect of size cannot be accurately quantified.
1, the conventional 2D machine vision uses a front or side photographing method, the image characteristics are easily influenced by illumination, and a misdetection result is generated;
2. another method for processing surface detection is to use an intelligent image algorithm, for example, a machine learning method to classify image features, and to train a large number of samples to realize classification, but this method cannot quantify the size, and requires a long time to train and learn;
2. the existing laser profilometer can output surface point clouds, but the measurement width is narrow, the precision and the efficiency are not ideal, and meanwhile, the measurement on the mixed material is unreliable.
3. The depth data is used for detection, the 3D point cloud registration problem is involved, the current mainstream 3D point cloud registration algorithm is an iterative nearest neighbor algorithm (ICP) adaptive to free-form surface registration, the efficiency of 3D visual detection is low, and the online test requirement cannot be met.
Disclosure of Invention
The invention aims to solve the technical problem of providing a high-precision laser 3D contour mobile phone structural part detection method for solving the problems in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a high-precision laser 3D contour mobile phone structure detection method specifically comprises the following steps:
step 1) scanning a sample in advance by using a laser measuring instrument, setting scanning parameters of the sample, including laser power, image exposure time and acquisition range, and acquiring a laser scanning image sample of the sample;
step 2) placing a mobile phone structural member on a movable workbench, fixing a laser measuring head right above the mobile workbench, driving a workbench supporting plate to move by a servo motor to finish detection, finishing surface contour detection of the system within 2 seconds, wherein the sampling times St is pl/sstep, pl is the product measuring length, and sstep is the sampling step length;
step 3), establishing a profile standard data model;
step 4) importing 3D model CAD data before measurement, wherein the data is composed of thousands to tens of thousands of triangular surface patches, intercepting a slice image of the 3D model along a path vertical to the scanning direction, forming an intersection line between each space triangle and a slice plane of the 3D data of the structural member, calculating an intersection point of each intersection line and each side of the space triangle, judging whether the intersection point is between line segments formed by end points of the triangles, and recording intersection point coordinates between 2 line ends as slice sampling;
after the recording of the slicing sampling points is finished, sequencing the slicing sampling points to form profile data of the section, wherein key points are detected in a profile data set, and the profile contour line of the section is represented by a sequence of the key points;
and, each slice data uses (x1, pt1), (x2, pt2), (x3, pt3), (x4, pt4), (xn, ptn) sequence keypoints composition; ptn represents the data point type, and different weighting factors are adopted in the criterion of judging the flat data point and the data point with large curvature in the detection process, including:
using ptn-1 to indicate that the data point type is a flat point, and ptn-2 to indicate that the data point is an inflection point, and setting the data point as the inflection point when the curvature of the data point is larger than a certain range;
arranging a plurality of slice data according to coordinate axes to form data height data, wherein the height data is the height value on a slice curve, and the matrix-shaped elevation data can be expressed into a 2D gray false color image, wherein the gray calculation formula on the image is shown in formula (2), so that the Z-axis data of the model is converted into gray values, and the subsequent matching detection is more convenient;
Pg(x,y)＝255*(z(x,y)-z1)/(Z2-Z1) (2)
step 5)3D point cloud matching, wherein before detection, the actually measured 3D point cloud of the structural member is registered with an input standard model, and the displacement and the rotation angle of the actually measured structural member relative to the standard model are calculated, wherein the model registration algorithm is an iterative nearest neighbor ICP algorithm, and the judgment rule of the ICP algorithm is as follows:
after a product measurement height map is generated, converting the height map into a gray pseudo-color map;
step 6), after the height map of the actual product and the height map of the reference model are generated, carrying out registration processing on the height map of the actual product and the height map of the reference model, wherein a 2D contour positioning method is used as a registration method;
step 7) after product registration operation, calculating the distance difference between the model and the actual product measurement height data, comparing the actual mobile phone structural member measurement data with the model data, adopting a least square method and an Euclidean distance calculation method for the comparison method, and adopting different matching degree calculation methods according to different data point types;
step 8) feedback control and data quality monitoring of the measurement process;
wherein, include: the laser image acquisition and the laser control are fused together, the statistics of a sample image is calibrated firstly before the system measures, the image brightness mean value and the laser line width of a standard sample are recorded, whether the line width and the mean value of the laser image meet requirements or not is detected after each measurement in the measuring process is completed, if the line width and the mean value meet the requirements, a measuring result is output, otherwise, a user is informed to re-measure or optimize and output results, and then the result is output. In this way, the measurement accuracy of the output result is ensured.
Preferably, before the above test method, the method further comprises:
place test platform with cell-phone structure (4) on to laser line light source (1) that the semiconductor line structure light laser instrument that uses blue wavelength 405nm sent shines the cell-phone structure, wherein, the first CMOS that the symmetry set up makes a video recording module (2) and second CMOS make a video recording module (3) and gather the laser line image, acquire upper surface 3D point cloud data, wherein, 45 degrees have been selected to camera optical axis and laser line light plane contained angle, wherein, the formation of image is disclosed as follows:
wherein,
preferably, step 3) specifically comprises:
importing 3D model CAD data before measurement, and then carrying out normalization;
or, obtaining 3D point cloud data by using standard sample sampling, and then generating a standard three-dimensional point cloud after resampling and filtering.
Preferably, step 3) specifically comprises:
and 3D contour dimension data actually measured by using the standard sample is used as a standard template after being filtered.
Preferably, step 6) specifically comprises:
a) in a reference model diagram, the salient positioning point characteristics, such as round holes, square holes, right-angle sides and the like, are manually selected through a software interface, and the geometric quantity characteristics of the positioning characteristics are calculated, wherein the geometric quantity characteristics comprise the following steps: length, height, width, area, contour line normal vector and segmentation value characteristics, and storing the geometric characteristics of the contour as template parameters;
b) and performing binary segmentation on the product height map by using the segmentation value of the reference model, performing spot detection by using a Blob algorithm, comparing the detected spot characteristics with the positioning point characteristics of the reference model map, finding out the most similar spots as the position characteristic points of the product, and comparing the displacement relationship between the position characteristic points of the product and the positioning points of the reference model map, thereby realizing the registration of the actual product and the reference model.
Preferably, step 7) specifically comprises:
for ptn-1 flat plane type data points, firstly fitting a plane model to a reference point set, calculating by using a least square method, then calculating the distance from an actual sampling point to a plane or a plane equation, calculating the number and the position of points exceeding the distance of judging different severity level error thresholds, and finally outputting a final result by using the patent definition decision tree classifier;
for the measuring point at the turning point with the ptn being 2, fitting an arc surface model of the turning point by using a least square method, then calculating the distance dis between the radii of the surface model of the actual surface model and the surface model of the reference point set, and taking the dis value as the basis for judgment;
in the process of calculating the parameterized model, in order to avoid interference, an iterative algorithm is used for filtering out a certain proportion of disordered data points; wherein the filtration ratio is from 10% to 60%.
The invention uses the laser profile scanning technology to measure the depth dimension of the mobile phone structural part for detection, truly and objectively measures the physical quantity embodied by the structural part authority, has high detection data precision and low false alarm and false alarm probability, and can realize 0 false alarm rate. Machine learning training is not needed when the detection model is changed, the system has high flexibility, and rapid model changing can be realized. The detection result is related to the position of the light spot and is unrelated to the intensity of the light spot, and when the brightness of the laser light source changes, the change of the center of the laser line is far smaller than the change of the line width of the laser, so that the consistency of the detection result is good.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The present invention will be described in detail below with reference to the accompanying drawings so that the above advantages of the present invention will be more apparent. Wherein,
FIG. 1 is a schematic diagram of the method for detecting the structural member of the high-precision laser 3D profile mobile phone according to the present invention;
FIG. 2 is a schematic flow chart of the method for detecting the structural member of the high-precision laser 3D profile mobile phone according to the invention;
FIG. 3 is a schematic diagram of the method for detecting the structural part of the high-precision laser 3D contour mobile phone.
Detailed Description
The following detailed description of the embodiments of the present invention will be provided with reference to the drawings and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, the embodiments and the features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.
Additionally, the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions and, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than here.
The method for testing the mobile phone structural part comprises the following steps:
1. placing a mobile phone structural part on a test platform, irradiating the mobile phone structural part by using a semiconductor line structured light laser with a blue wavelength of 405nm, acquiring a laser line image by using a double CMOS camera, acquiring 3D point cloud data of an upper surface, and selecting 45 degrees for an included angle between an optical axis of the camera and an optical plane of the laser line in order to achieve the highest measurement precision. According to the formula (1), increasing the included angle between the laser plane and the camera optical axis can improve the measurement accuracy, but is limited by the structural size, so a system structural angle of 45 degrees is used in the measurement method for the mobile phone structure. The imaging configuration is shown in figure 1.
Wherein,
FIG. 1 illustrates a laser line source 1; 2. a first CMOS camera module; 3. a second CMOS camera module; 4. a mobile phone structural member; 5. the samples were tested.
2. And scanning the sample in advance by using a laser measuring instrument, and setting scanning parameters of the sample, including laser power, image exposure time and acquisition range. A sample of a laser scan image of the sample is collected.
3. The normal detection operation process of the mobile phone structural part is as follows, the mobile phone structural part is placed on a movable workbench, a laser measuring head is fixed right above the mobile workbench, a servo motor drives a workbench supporting plate to move, detection is completed, and the system completes surface contour detection within 2 seconds. The sampling times St is pl/sstep, pl is the product measurement length, sstep is the sampling step. The system can set sampling step length according to different sizes and structural complexity of products.
4. The contour comparison needs to establish a standard data model, one of the standard model establishing methods is to introduce 3D model CAD data before measurement and then carry out normalization; another way is to obtain 3D point cloud data using standard sample sampling, then re-sample, filter and generate a standard three-dimensional point cloud. The second standard model building method is to use the 3D contour dimension data actually measured by the standard sample as a standard template after filtering processing.
5. The method comprises the steps of importing 3D model CAD data before measurement in a format stl format supported by mainstream 3D CAD design software, wherein the data is composed of thousands to tens of thousands of triangular patches, in order to realize rapid comparison and measurement of outlines, adopting a method for calculating geometry, intercepting a slice image of the 3D model along a path perpendicular to a scanning direction, forming intersecting lines on each space triangle and a slice plane of the structural part 3D data, calculating intersection points of the intersecting lines and each side of the space triangle, judging whether the intersection points are between line segments formed by end points of the triangle, and recording intersection point coordinates between 2 line ends as slice sampling points. And after the recording of the slicing sampling points is finished, sequencing the slicing sampling points to form profile data of the section. The measurement directly using the profile data of the cross section generates more misjudgment and has larger data volume, so that key points are detected in the profile data set, and the profile line of the cross section is represented by a sequence of the key points. For each type of structural part, multiple groups of section data are created according to fixed step length, and section model data are created by using five step lengths of 0.025mm, 0.05 mm, 0.075mm,0.1mm and 0.2 mm. Each slice data was composed using (x1, pt1), (x2, pt2), (x3, pt3), (x4, pt4),. ·, (xn, ptn) sequence keypoints. ptn represents a data point type, different weighting factors are adopted in the criterion of judging a flat data point and a data point with large curvature in the detection process, the data point type is a flat point by using ptn as 1, the data point is an inflection point by using ptn as 2, and the inflection point is set when the curvature of the data point is larger than a certain range. And arranging a plurality of slice data according to a coordinate axis to form data height data, wherein the height data is the height value on a slice curve, and the matrix-shaped height data can be expressed into a 2D gray false color image. The gray level calculation formula on the image is shown as the formula (2), so that the Z-axis data of the model is converted into the gray level value, and the subsequent matching detection is more convenient.
Pg(x,y)＝255*(z(x,y)-z1)/(Z2-Z1) (2)
6. And 3D point cloud matching, namely registering the actually measured 3D point cloud of the structural member and an input standard model before detection, and calculating the displacement and the rotation angle of the actually measured structural member relative to the standard model. The currently common model registration algorithm is the iterative nearest neighbor ICP algorithm. The ICP algorithm has the judgment rule that:
the ICP algorithm is mainly used in the field of free-form surface three-dimensional splicing, and the matching precision and efficiency in the field of regular structure detection are not high. In fact, the point cloud of the mobile phone structural part can be represented by digital height on a reference plane, so that the height map is used for representing the measurement result of the mobile phone structural part. Since the test process may be slightly skewed, the tilt angle correction is performed using the ICP algorithm prior to height map generation, the angle search range is limited to +/-n degrees, the value of n is set using software, and the amount of computation is greatly reduced as the search range is reduced.
After the height map is generated, the height map is converted into a gray pseudo-color map, wherein light color in the image represents convex structural features, and dark color represents concave structural features. The gray false color image is processed according to a gray image processing method, and a 2D image detection method is used for post-processing.
7. After the actual product height map and the reference model height map are generated, the actual product height map and the reference model height map are subjected to registration processing, a 2D contour positioning method is used in the registration method, the method is fast in speed, and positioning calculation can be completed within tens of milliseconds. The registration process is as follows:
a) in a reference model diagram, the characteristics of a significant positioning point, such as a round hole, a square hole, a right-angle side and the like, are manually selected through a software interface, the geometric quantity characteristics (length, height, width, area and contour line normal vector) and the segmentation value characteristics of the positioning characteristics are calculated, and the geometric characteristics of the contour are stored as template parameters;
b) and performing binary segmentation on the product height map by using the segmentation value of the reference model, performing spot detection by using a Blob algorithm, comparing the detected spot characteristics with the positioning point characteristics of the reference model map, finding out the most similar spots as the position characteristic points of the product, and comparing the displacement relationship between the position characteristic points of the product and the positioning points of the reference model map, thereby realizing the registration of the actual product and the reference model.
8. After the product registration operation, the distance difference between the model and the actual product measurement height data needs to be calculated, the actual mobile phone structural member measurement data is compared with the model data, and a least square method and an Euclidean distance calculation method are adopted for the comparison method. Different matching degree calculation methods are adopted according to different data point types. For ptn-1 flat plane type data points, a plane model is fitted to a reference point set, least square method calculation is used, then the distance from an actual sampling point to a plane or a plane equation is calculated, the number and the position of points exceeding the distance of error threshold values for judging different severity levels are calculated, and finally a final result is output by using the patent definition decision tree classifier. And for the measuring point with the ptn being 2, fitting an arc surface model of the inflection point by using a least square method, then calculating the distance dis between the radii of the actual surface model and the surface model of the reference point set, and taking the dis value as the basis for judgment. In order to avoid interference in the process of calculating the parameterized model, an iterative algorithm is used for filtering a certain proportion of disordered data points, the filtering proportion used in the method is from 10% to 60%, and the method is adjusted according to different product type systems.
9. Feedback control of the measurement process and data quality monitoring. The stability of the light source is important in structured light measurement systems. This patent fuses laser image collection and laser control together, and the statistics of at first demarcation sample image before the system measurement records image brightness mean value, the laser line width of standard sample, and at the measurement in-process at every turn after accomplishing, whether the line width, the mean value of detection laser image satisfy the requirement, if satisfy the requirement then output measuring result, otherwise inform the user to measure again or output the result again after the result carries out optimization processing more. In this way, the measurement accuracy of the output result is ensured. The control of the laser is completed by manufacturing a laser control panel, and the control signal adopts a voltage signal in the range of 0-3 v.
The depth dimension of the mobile phone structural part is measured by using a laser profile scanning technology for detection, the physical quantity reflected by the structural part authority is measured truly and objectively, the detection data precision is high, the false alarm and false alarm probability is low, and the false alarm rate of 0 can be realized. Machine learning training is not needed when the detection model is changed, the system has high flexibility, and rapid model changing can be realized. The detection result is related to the position of the light spot and is unrelated to the intensity of the light spot, and when the brightness of the laser light source changes, the change of the center of the laser line is far smaller than the change of the line width of the laser, so that the consistency of the detection result is good. And converting the 3D point cloud data of the theoretical model into digital elevation data, and representing the digital elevation data by using a 2D gray level image. This allows 3D data to be processed with 2D image algorithms.
The key points of the scheme which need protection are summarized as follows:
1. the laser plane and the binocular CMOS sensor large-angle imaging structure are innovatively provided for detecting a precise structural part, the designed angle is 45 degrees, and the measurement precision of the structural part of the mobile phone is optimized in the mode.
2. In order to adapt to the measurement of mobile phone structural parts made of different materials, the laser intensity is monitored in the measurement process, the laser energy is automatically adjusted according to the line width and the light spot scattering degree detected by the image, and the high-stability measurement process is realized.
3. And converting the 3D contour data into a 2D gray image or a false color image, and realizing rapid registration and rapid detection of the contour data by using a 2D visual Blob positioning algorithm and a gray comparison algorithm.
4. And a detection template STL data processing method is used for converting template 3D point cloud data into a normalized 2D image.
5. The actual measurement point cloud data is represented by matrixing digital height data, converted into a measurement depth image, positioned by calling a 2D contour detection algorithm, subtracted by the positioned data and model height data, a threshold value is set for a residual image after subtraction for segmentation, and the area, length and width characteristics of effective defects of a segmentation map are detected and counted by using Blob.
6. Different methods are adopted for comparing the distance between the reference model and the actual product, the flat surface is compared by using a plane linear model, and the turning part is compared by using an arc model.
7. A tree-shaped classifier is designed for defect judgment, firstly, a weak classifier is used for distinguishing significant defects and products with obvious quality, and then suspected defects are further distinguished by combining defect feature distribution until the false detection rate and the missed detection rate meet the requirements of customers.
It should be noted that for simplicity of description, the above method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A high-precision laser 3D contour mobile phone structure detection method is characterized by comprising the following steps:
step 1) scanning a sample in advance by using a laser measuring instrument, setting scanning parameters of the sample, including laser power, image exposure time and acquisition range, and acquiring a laser scanning image sample of the sample;
step 2) placing a mobile phone structural member on a movable workbench, fixing a laser measuring head right above the mobile workbench, driving a workbench supporting plate to move by a servo motor to finish detection, finishing surface contour detection of the system within 2 seconds, wherein the sampling times St is pl/sstep, pl is the product measuring length, and sstep is the sampling step length;
step 3), establishing a profile standard data model;
step 4) importing 3D model CAD data before measurement, wherein the data is composed of thousands to tens of thousands of triangular surface patches, intercepting a slice image of the 3D model along a path vertical to a scanning direction, forming an intersection line between each space triangle and a slice plane of the 3D data of the structural member, calculating an intersection point of each intersection line and each side of the space triangle, judging whether the intersection point is between line segments formed by end points of the triangles, and recording intersection point coordinates as slice sampling points if the intersection point is between 2 line segments;
after the recording of the slicing sampling points is finished, sequencing the slicing sampling points to form profile data of the section, wherein key points are detected in a profile data set, and the profile contour line of the section is represented by a sequence of the key points;
and, each slice data uses (x1, pt1), (x2, pt2), (x3, pt3), (x4, pt4), (xn, ptn) sequence keypoints composition; ptn represents the data point type, and different weighting factors are adopted in the criterion of judging the flat data point and the data point with large curvature in the detection process, including:
using ptn-1 to indicate that the data point type is a flat point, and ptn-2 to indicate that the data point is an inflection point, and setting the data point as the inflection point when the curvature of the data point is larger than a certain range;
arranging a plurality of slice data according to coordinate axes to form data height data, wherein the height data is the height value on a slice curve, and the matrix-shaped height data is expressed into a 2D gray false color image, wherein the gray calculation formula on the image is shown in formula (2), so that the Z-axis data of the model is converted into gray values, and the subsequent matching detection is more convenient;
Pg(x,y)＝255*(z(x,y)-z1)/(z2-z1) (2)
step 5)3D point cloud matching, namely registering the actually measured 3D point cloud of the structural part with an input standard model before detection, and calculating the displacement and the rotation angle of the actually measured structural part relative to the standard model, wherein the model registration algorithm is an iterative nearest neighbor ICP algorithm, and the judgment rule of the ICP algorithm is as follows:
after a product measurement height map is generated, converting the height map into a gray pseudo-color map;
step 6), after the height map of the actual product and the height map of the reference model are generated, carrying out registration processing on the height map of the actual product and the height map of the reference model, wherein a 2D contour positioning method is used as a registration method;
step 7) after product registration operation, calculating the distance difference between the model and the actual product measurement height data, comparing the actual mobile phone structural member measurement data with the model data, adopting a least square method and an Euclidean distance calculation method for the comparison method, and adopting different matching degree calculation methods according to different data point types;
step 8) feedback control and data quality monitoring of the measurement process;
wherein, include: the laser image acquisition and the laser control are integrated together, the statistics of a sample image is calibrated firstly before the system is measured, the image brightness mean value and the laser line width of a standard sample are recorded, whether the line width and the mean value of the laser image meet requirements or not is detected after each measurement is completed in the measuring process, if the line width and the mean value meet the requirements, a measuring result is output, otherwise, a user is informed to re-measure or output results for optimization processing, and then the result is output, and the measuring precision of the output result is ensured through the mode.
2. The method for detecting the structural part of the high-precision laser 3D profile mobile phone according to claim 1, wherein before the above test method, the method further comprises:
place the test platform with the cell phone structure spare on to the laser line light source that uses blue wavelength 405 nm's semiconductor line structure light laser instrument to send shines the cell phone structure spare, wherein, the first CMOS that the symmetry set up makes a video recording module and second CMOS make a video recording module and gather the laser line image, acquires upper surface 3D point cloud data, wherein, 45 degrees have been selected to camera optical axis and laser line light plane contained angle, and wherein, the imaging formula is as follows:
wherein,
3. the method for detecting the structural part of the high-precision laser 3D profile mobile phone according to claim 1, wherein the step 3) specifically comprises the following steps: importing 3D model CAD data before measurement, and then carrying out normalization;
or, obtaining 3D point cloud data by using standard sample sampling, and then generating a standard three-dimensional point cloud after resampling and filtering.
4. The method for detecting the structural part of the high-precision laser 3D profile mobile phone according to claim 1, wherein the step 3) specifically comprises the following steps: and 3D contour dimension data actually measured by using the standard sample is used as a standard template after being filtered.
5. The method for detecting the structural part of the high-precision laser 3D contour mobile phone according to claim 1, wherein the step 6) specifically comprises the following steps:
a) in a reference model diagram, a significant positioning point feature is manually selected through a software interface, and the geometric quantity feature of the positioning feature is calculated, wherein the method comprises the following steps: length, height, width, area, contour line normal vector and segmentation value characteristics, and storing the geometric characteristics of the contour as template parameters;
b) and performing binary segmentation on the product height map by using the segmentation value of the reference model, performing spot detection by using a Blob algorithm, comparing the detected spot characteristics with the positioning point characteristics of the reference model map, finding out the most similar spots as the position characteristic points of the product, and comparing the displacement relationship between the position characteristic points of the product and the positioning points of the reference model map, thereby realizing the registration of the actual product and the reference model.
6. The method for detecting the structural part of the high-precision laser 3D contour mobile phone according to claim 1, wherein the step 7) specifically comprises the following steps:
for ptn-1 flat plane type data points, firstly fitting a plane model to a reference point set, calculating by using a least square method, then calculating the distance from an actual sampling point to a plane or a plane equation, calculating the number and the position of points exceeding the distance of judging different severity level error thresholds, and finally outputting a final result by using a decision tree classifier;
for the measuring point at the turning point with the ptn being 2, fitting an arc surface model of the turning point by using a least square method, then calculating the distance dis between the radii of the surface model of the actual surface model and the surface model of the reference point set, and taking the dis value as the basis for judgment;
in the process of calculating the parameterized model, in order to avoid interference, an iterative algorithm is used for filtering out a certain proportion of disordered data points; wherein the filtration ratio is from 10% to 60%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153827.XA CN106949848B (en) | 2017-03-15 | 2017-03-15 | A kind of high-precision laser 3D profile phone structural detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153827.XA CN106949848B (en) | 2017-03-15 | 2017-03-15 | A kind of high-precision laser 3D profile phone structural detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106949848A CN106949848A (en) | 2017-07-14 |
| CN106949848B true CN106949848B (en) | 2019-11-29 |
Family
ID=59471964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710153827.XA Active CN106949848B (en) | 2017-03-15 | 2017-03-15 | A kind of high-precision laser 3D profile phone structural detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106949848B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107621235B (en) * | 2017-08-26 | 2024-02-27 | 东莞市三姆森光电科技有限公司 | Method and equipment for measuring contour of curved surface shell of mobile phone based on spectral confocal technology |
| CN109579765B (en) * | 2017-09-29 | 2020-09-18 | 上海时元互联网科技有限公司 | Three-dimensional point cloud data matching method and system based on servo motor and controller |
| CN108195289A (en) * | 2017-12-29 | 2018-06-22 | 上海华太信息技术有限公司 | A kind of method measured using line laser to workpiece specific position parameter |
| CN108596966A (en) * | 2018-03-14 | 2018-09-28 | 浙江大学山东工业技术研究院 | The measurement method at the refractory brick surface inclination angle based on fit Plane normal vector |
| CN108682012B (en) * | 2018-05-15 | 2021-07-27 | 佛山市南海区广工大数控装备协同创新研究院 | 3D curved surface glass surface flatness defect detection method based on line scanning laser |
| CN108592847A (en) * | 2018-07-27 | 2018-09-28 | Oppo（重庆）智能科技有限公司 | The method that the groove depth of electronic device is measured using measuring device |
| CN109739243B (en) * | 2019-01-30 | 2022-07-08 | 东软睿驰汽车技术（沈阳）有限公司 | Vehicle positioning method, automatic driving control method and related system |
| CN109596058A (en) * | 2019-02-01 | 2019-04-09 | 东莞中科蓝海智能视觉科技有限公司 | The size detection recognition methods of plastic workpiece |
| CN110049243B (en) * | 2019-04-19 | 2021-09-10 | 博众精工科技股份有限公司 | Image acquisition method, apparatus, device, and medium |
| CN110276371B (en) * | 2019-05-05 | 2021-05-07 | 杭州电子科技大学 | Container corner fitting identification method based on deep learning |
| CN112102223B (en) * | 2019-06-18 | 2024-05-14 | 通用电气精准医疗有限责任公司 | Method and system for automatically setting scan range |
| CN111336947A (en) * | 2020-03-02 | 2020-06-26 | 南昌航空大学 | Mirror surface object line laser scanning method based on binocular point cloud fusion |
| CN111193855B (en) * | 2020-03-24 | 2021-09-03 | 无锡创驰电气有限公司 | Binocular 3D camera |
| CN111412859B (en) * | 2020-03-24 | 2021-10-15 | 无锡创驰电气有限公司 | Monocular 3D camera |
| CN111504223B (en) * | 2020-04-22 | 2022-05-31 | 荆亮 | Blade profile measuring method, device and system based on line laser sensor |
| CN113267122B (en) * | 2021-05-12 | 2023-01-31 | 温州理工学院 | Industrial part size measurement method based on 3D vision sensor |
| CN114399507B (en) * | 2022-03-25 | 2022-06-17 | 季华实验室 | Mobile phone appearance quality detection method and device, electronic equipment and storage medium |
| CN115046477B (en) * | 2022-06-16 | 2024-10-11 | 广东博智林机器人有限公司 | Laser detection device |
| CN116051542B (en) * | 2023-03-06 | 2023-07-14 | 深圳市深视智能科技有限公司 | Defect detection method and defect detection device |
| CN117053718B (en) * | 2023-10-11 | 2023-12-12 | 贵州黔程弘景工程咨询有限责任公司 | Beam bottom linear model generation method based on beam bottom linear measurement |
| CN117299583B (en) * | 2023-11-25 | 2024-05-03 | 东莞市欧宇精密科技有限公司 | Control method of optical screening machine |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5546189A (en) * | 1994-05-19 | 1996-08-13 | View Engineering, Inc. | Triangulation-based 3D imaging and processing method and system |
| US6915004B2 (en) * | 2002-02-28 | 2005-07-05 | Cti Pet Systems, Inc. | Continuous tomography bed motion data processing apparatus and method |
| CN101639452A (en) * | 2009-09-11 | 2010-02-03 | 北京科技大学 | 3D detection method for rail surface defects |
| CN101907439A (en) * | 2010-03-17 | 2010-12-08 | 中国二十二冶集团有限公司 | Stimulated measurement and detection method in architectural steel structure fabrication |
| CN102592284A (en) * | 2012-02-27 | 2012-07-18 | 上海交通大学 | Method for transforming part surface appearance three-dimensional high-density point cloud data into grayscale image |
| CN105526882A (en) * | 2015-12-28 | 2016-04-27 | 西南交通大学 | Turnout wear detection system and detection method based on structured light measurement |
-
2017
- 2017-03-15 CN CN201710153827.XA patent/CN106949848B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5546189A (en) * | 1994-05-19 | 1996-08-13 | View Engineering, Inc. | Triangulation-based 3D imaging and processing method and system |
| US6915004B2 (en) * | 2002-02-28 | 2005-07-05 | Cti Pet Systems, Inc. | Continuous tomography bed motion data processing apparatus and method |
| CN101639452A (en) * | 2009-09-11 | 2010-02-03 | 北京科技大学 | 3D detection method for rail surface defects |
| CN101907439A (en) * | 2010-03-17 | 2010-12-08 | 中国二十二冶集团有限公司 | Stimulated measurement and detection method in architectural steel structure fabrication |
| CN102592284A (en) * | 2012-02-27 | 2012-07-18 | 上海交通大学 | Method for transforming part surface appearance three-dimensional high-density point cloud data into grayscale image |
| CN105526882A (en) * | 2015-12-28 | 2016-04-27 | 西南交通大学 | Turnout wear detection system and detection method based on structured light measurement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106949848A (en) | 2017-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106949848B (en) | A kind of high-precision laser 3D profile phone structural detection method | |
| CN107064170B (en) | Method for detecting profile defect of mobile phone shell | |
| CN109520436B (en) | Butterfly spring three-dimensional size automatic measurement system based on machine vision and measurement method thereof | |
| CN105842885B (en) | A kind of liquid crystal display defect Hierarchical Location method and device | |
| CN114041168A (en) | Automated 360-degree dense point object inspection | |
| CN105021628A (en) | Detection method for surface defects of optical fiber image inverter | |
| WO2013061976A1 (en) | Shape inspection method and device | |
| JP2008292365A (en) | Shape evaluation method, shape evaluation device, and three-dimensional inspection device | |
| JP2017032548A (en) | Using 3d vision for automated industrial inspection | |
| CN112734662B (en) | Machine vision detection method and system for bevel gear abrasion | |
| CN101793840A (en) | Diamond cutting parameter measurement method and measuring device | |
| CN113920081A (en) | Cutter wear degree detection method | |
| CN109283182A (en) | A kind of detection method of battery welding point defect, apparatus and system | |
| US20210156669A1 (en) | Measuring Device for Examining a Specimen and Method for Determining a Topographic Map of a Specimen | |
| JP5913903B2 (en) | Shape inspection method and apparatus | |
| CN114280075A (en) | Online visual inspection system and method for surface defects of pipe parts | |
| CN117649404A (en) | Medicine packaging box quality detection method and system based on image data analysis | |
| CN116678826A (en) | Appearance defect detection system and method based on rapid three-dimensional reconstruction | |
| CN115656182A (en) | Sheet material point cloud defect detection method based on tensor voting principal component analysis | |
| CN109622404B (en) | Automatic sorting system and method for micro-workpieces based on machine vision | |
| CN108844469B (en) | Method and system for testing workpiece step height based on laser | |
| CN103472064B (en) | A kind of method of justifying the qualification of bright cut jewel cut | |
| Yu et al. | Visual texture-based 3-D roughness measurement for additive manufacturing surfaces | |
| García et al. | Rail surface inspection system using differential topographic images | |
| TW201317587A (en) | Dimension measuring device and method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: 518000 room 501-504, building 2, Chongwen Park, Nanshan wisdom Park, 3370 Liuxian Avenue, Fuguang community, Taoyuan Street, Nanshan District, Shenzhen City, Guangdong Province Patentee after: SHENZHEN SINCEVISION TECHNOLOGY Co.,Ltd. Address before: 518000 room 519, block a, Shenzhen Zhihui innovation center, Taoyuanju south of Qianjin 2nd Road, Bao'an District, Shenzhen City, Guangdong Province Patentee before: SHENZHEN SINCEVISION TECHNOLOGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |