An automatic identification system for surface defects of machined workpieces

Abstract

The present invention discloses an automatic recognition system for surface defects of a workpiece, comprising: a scanning module, scanning a standard workpiece and a workpiece to be detected, obtaining standard data and data to be detected, establishing a coordinate system, wherein the standard data and data to be detected include reflected light intensity data and depth data, generating a standard data set and a data set to be detected, a comparison measurement module, comparing the standard data set and the data set to be detected, obtaining a comparison result, determining the defect type, calculating the position of the comparison result on the coordinate system, and a projection module, projecting the defect type and position on the workpiece to be detected. The present invention scans the standard workpiece and the workpiece to be detected and aligns them, obtains the standard data set and the data set to be detected, and compares and marks them, finds different defect types and corresponding positions, and projects the defects on the workpiece to be detected by laser, thereby achieving the purpose of defect type recognition and projection.

Description

Technical Field

[0001] The present invention relates to the field of visual inspection technology, and in particular to an automatic recognition system for surface defects of processed workpieces and a system thereof. Background Art

[0002] With the rapid development of modern manufacturing industry, the requirements for workpiece surface quality are increasing day by day. Defects on the workpiece surface, such as dents, protrusions, cracks, etc., not only affect the aesthetics of the product, but more importantly, may have a serious impact on its performance and service life. At present, workpiece surface defect detection mainly relies on manual inspection and traditional machine vision inspection methods. Although manual inspection has high accuracy, it is inefficient and easily affected by human factors. Although traditional machine vision inspection methods can realize automated inspection, their ability to recognize complex surface defects is limited, and the recognition effect will be seriously affected.

[0003] At present, the Chinese invention patent with application number 201910101499.8 discloses a method for detecting and identifying defects in workpieces in a tray, including the following steps: step 1: image acquisition of the workpieces on the tray; step 2: binarization of the acquired image; step 3: spot detection algorithm on the binarized image to identify the defects of each workpiece in the tray; step 4: converting the result of the identification judgment into readable data information output, the tray only needs to be moved to the shooting position to automatically complete the block division of the tray and image acquisition of each divided block, without the need to move to different shooting stations for multiple shots, reducing the layout of hardware equipment and reducing manufacturing costs. This method for detecting and identifying defects in workpieces in trays cannot confirm the specific location of the surface defects of the detected workpiece and mark them, and cannot project the defect location on the workpiece to be inspected. Summary of the invention

[0004] The technical problem solved by the present invention is that the prior art cannot perform surface defect comparison and oil stain comparison identification on the defects on the surface of the processed workpiece, and cannot project the defect position on the processed workpiece to be detected, which affects the defect identification effect.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An automatic recognition system for surface defects of a workpiece includes a scanning module, a comparison measurement module and a projection module:

[0007] The scanning module is used to scan the standard processed workpiece and the processed workpiece to be detected, obtain standard data and data to be detected, align the scanning surfaces of the standard processed workpiece and the processed workpiece to be detected and establish a coordinate system, wherein the standard data and the data to be detected include reflected light intensity data and depth data, respectively establish mapping relationships between the coordinate systems of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and depth data, and generate a standard data set and a data set to be detected;

[0008] The comparison measurement module is used to compare the reflected light intensity data and depth data of the standard data set and the data set to be tested, obtain the comparison result, determine the defect type, and calculate the position of the comparison result on the coordinate system;

[0009] The projection module projects the defect type and position onto the workpiece to be inspected by laser;

[0010] The scanning module includes a parameter setting unit, a data acquisition unit and a model building unit:

[0011] The parameter setting unit selects scanning parameters according to the size and surface complexity of the scanned standard workpiece and the workpiece to be inspected;

[0012] The data acquisition unit scans the standard processed workpiece and the to-be-detected processed workpiece according to the set scanning parameters to obtain a plurality of standard surfaces and standard data corresponding to the standard surfaces and surfaces to be detected and data to be detected corresponding to the surfaces to be detected;

[0013] The model building unit aligns the surface to be tested and the standard surface accordingly and establishes a coordinate system, and respectively establishes a standard data set and a data set to be tested that are mapped to the coordinate system according to the standard data and the data to be tested;

[0014] The model building unit aligns the scanned surface to be detected and the standard scanned surface respectively to obtain a reference point, establishes a coordinate system with the reference point as the coordinate origin, sets 0.5*Vmm as the horizontal axis coordinate spacing, and 0.1mm as the vertical axis coordinate spacing, establishes a mapping relationship between the coordinate system of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and the depth data, and obtains a standard data set and a data set to be detected;

[0015] The model building unit respectively registers the scan surface to be detected and the standard scan surface, including the following steps:

[0016] Step 1, the standard surface is recorded as S, and the surface to be detected is recorded as T, and they are segmented once to obtain the segmented standard surface S1 and standard surface S2, and the segmented surface to be detected T1 and surface to be detected T2;

[0017] Step 2, respectively calculating the data mean of the standard data corresponding to the surface to be tested T1 and the standard surface S1;

[0018] Step 3, respectively calculating the data mean of the to-be-tested data corresponding to the to-be-tested surface T2 and the standard surface S2;

[0019] Step 4, respectively comparing the data mean values ​​of the standard surface S1 and the surface to be tested T1, and the data mean values ​​of the standard surface S2 and the surface to be tested T2;

[0020] Step 5: If the data means of the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are equal, it means that the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are successfully registered;

[0021] Step 6: If the data means of the standard surface S1 and the surface to be tested T1 or the standard surface S2 and the surface to be tested T2 are not equal, the unequal surfaces to be tested have defects, and the surface to be tested with defects and the corresponding standard surface are segmented twice, and steps 2 to 4 are repeated until the means of the two corresponding parts after segmentation are equal. The middle value of the data in the equal area is set as the reference point, and the standard surface S and the surface to be tested T are aligned;

[0022] Step 7, if after repeating Step 2 to Step 4, the data means of the standard surface S1 and the surface to be detected T1 and the data means of the standard surface S2 and the surface to be detected T2 are not equal, then return to the state of the last segmentation, set the middle value of the data in the segmented area as the reference point, and align the standard surface S and the surface to be detected T;

[0023] The comparison measurement module includes a symbol setting unit, a defect comparison unit and a positioning unit:

[0024] The symbol setting unit sets symbols for the numerical values ​​of the results of the comparison between different to-be-detected data sets and the standard data sets:

[0025] The position smaller than the depth data in the standard data set is marked as X, the position larger than the depth data in the standard data set is marked as Y, and the position equal to the depth data in the standard data set is marked as Z;

[0026] The position with different reflected light intensity data from the standard data set is marked as A, and the position with the same reflected light intensity data as the standard data set is marked as B;

[0027] The defect comparison unit is used to compare the data set to be detected with the standard data set, traverse each row and each column of the standard data set and the data set to be detected, compare the corresponding reflected light intensity data and depth data in the standard data set and the data set to be detected, create a new data set, the new data set has the same data structure as the standard data set and the data set to be detected, and the data type is character. According to the set symbol, the comparison result is filled into the corresponding position of the new data set to obtain the defect data set;

[0028] The positioning unit calculates the defect position according to the defect data set and generates a defect position signal;

[0029] The projection module includes a defect location classification unit and a partition projection unit:

[0030] The defect location classification unit distinguishes different defect areas and generates a plurality of defect location signals of different types according to the set symbol and defect location signal;

[0031] The partition projection unit projects different types of defect areas onto the workpiece to be inspected by laser according to different types of defect position signals.

[0032] Preferably, the basic parameter settings of the parameter setting unit are:

[0033] The fixed scanning frequency is 20Hz, and the scanning interval d is divided into three levels, namely 0.1mm, 0.25mm and 0.5mm, which can be adjusted steplessly within the range of 0.1mm-0.5mm;

[0034] The scanning speed V is divided into three levels, namely 2mm / s, 5mm / s and 12.5mm / s, and can be adjusted steplessly within the range of 2mm / s-12.5mm / s, and changes proportionally with the scanning spacing d.

[0035] Preferably, the data acquisition unit uses a laser point scanner, a depth sensor and an optical sensor to scan the upper and lower surfaces, left and right surfaces and front and back surfaces of the standard processed workpiece and the workpiece to be inspected, and uses the upper and lower surfaces, left and right surfaces and front and back surfaces of the standard processed workpiece and the workpiece to be inspected as standard scanning surfaces and scanning surfaces to be inspected, respectively, to obtain the depth data and reflected light data at the scanning points of each standard scanning surface and scanning surface to be inspected, and obtain standard data and data to be inspected corresponding to multiple standard surfaces and multiple surfaces to be inspected.

[0036] Preferably, the model building unit aligns the scanning surface to be detected and the standard scanning surface respectively to obtain a reference point, establishes a coordinate system with the reference point as the coordinate origin, sets 0.5*Vmm as the horizontal axis coordinate spacing, and 0.1mm as the vertical axis coordinate spacing, establishes a mapping relationship between the coordinate system of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and depth data, and obtains a standard data set and a data set to be detected.

[0037] Preferably, the positioning unit calculates the coordinates corresponding to each symbol according to the symbol position in the defect data set, and the calculation expression is:

[0038] The horizontal coordinate of the defect = 0.05V* the horizontal order of the defect value

[0039] Defect vertical coordinate = 0.1*defect value vertical order

[0040] get:

[0041] Defect coordinates (defect horizontal coordinate, defect vertical coordinate)

[0042] Output defect location signal.

[0043] Beneficial effects of the present invention: The present invention obtains a standard data set and a data set to be detected with coordinate information, depth data, and reflected light intensity data by scanning and aligning a standard processed workpiece and a workpiece to be detected, compares and marks the standard data set and the data set to be detected, finds out different defect types and corresponding positions, and projects the defects onto the workpiece to be detected by laser, thereby achieving the purpose of defect type identification and projection. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of the basic flow of a system for automatically identifying surface defects of a workpiece provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of the basic flow of a model building unit of a system for automatically identifying surface defects of a workpiece provided by an embodiment of the present invention respectively registering a scanning surface to be inspected and a standard scanning surface. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0047] The embodiments of the present invention refer to Figure 1 and Figure 2 , provides an automatic recognition system for surface defects of processed workpieces, including a scanning module, a comparison measurement module and a projection module:

[0048] The scanning module is used to scan the standard processed workpiece and the processed workpiece to be detected, obtain standard data and data to be detected, align the scanning surfaces of the standard processed workpiece and the processed workpiece to be detected and establish a coordinate system. The standard data and the data to be detected include reflected light intensity data and depth data. A mapping relationship between the coordinate system of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and depth data is established respectively to generate a standard data set and a data set to be detected.

[0049] The comparison measurement module is used to compare the reflected light intensity data and depth data of the standard data set and the data set to be tested, obtain the comparison result, determine the defect type, and calculate the position of the comparison result on the coordinate system.

[0050] The projection module projects the defect type and location onto the workpiece to be inspected using laser.

[0051] The scanning module includes a parameter setting unit, a data acquisition unit and a model building unit:

[0052] The parameter setting unit selects scanning parameters according to the size and surface complexity of the scanned standard processed workpiece and the processed workpiece to be detected.

[0053] The data acquisition unit scans the standard processed workpiece and the processed workpiece to be detected according to the set scanning parameters to obtain multiple standard surfaces and standard data corresponding to the standard surfaces and surfaces to be detected and data to be detected corresponding to the surfaces to be detected.

[0054] The model building unit aligns the surface to be detected and the standard surface accordingly and establishes a coordinate system, and respectively establishes a standard data set and a data set to be detected that are mapped to the coordinate system according to the standard data and the data to be detected.

[0055] The comparison measurement module includes a symbol setting unit, a defect comparison unit and a positioning unit:

[0056] The symbol setting unit sets symbols for the numerical values ​​of the results of comparing different datasets to be tested with the standard dataset:

[0057] The position smaller than the depth data in the standard data set is marked as X, the position larger than the depth data in the standard data set is marked as Y, and the position equal to the depth data in the standard data set is marked as Z.

[0058] The positions with different reflected light intensity data from the standard data set are marked as A, and the positions with the same reflected light intensity data as the standard data set are marked as B.

[0059] Setting different symbols facilitates the subsequent classification of defect location coordinates.

[0060] The defect comparison unit is used to compare the data set to be tested and the standard data set, traverse each row and each column of the standard data set and the data set to be tested, compare the corresponding reflected light intensity data and depth data in the standard data set and the data set to be tested, and create a new data set. The new data set has the same data structure as the standard data set and the data set to be tested, and the data type is character. According to the set symbol, the comparison result is filled into the corresponding position of the new data set to obtain the defect data set.

[0061] The positioning unit calculates the defect position according to the defect data set and generates a defect position signal.

[0062] The projection module includes a defect location classification unit and a partition projection unit:

[0063] The defect location classification unit distinguishes different defect areas and generates a plurality of defect location signals of different types according to the set symbols and defect location signals.

[0064] The partition projection unit projects different types of defect areas onto the workpiece to be inspected by laser according to different types of defect position signals.

[0065] The basic parameter settings of the parameter setting unit are:

[0066] The fixed scanning frequency is 20 Hz, and the scanning interval d is divided into three levels, namely 0.1 mm, 0.25 mm and 0.5 mm, and can be adjusted steplessly within the range of 0.1 mm-0.5 mm.

[0067] The scanning speed V is divided into three levels, namely 2mm / s, 5mm / s and 12.5mm / s, and can be adjusted steplessly within the range of 2mm / s-12.5mm / s, and changes proportionally with the scanning spacing d. The finer the surface of the scanned standard workpiece and the workpiece to be inspected, and the smaller the size, the smaller the required scanning spacing d and scanning speed. At the same time, the basic parameters are convenient for users to make preliminary adjustments.

[0068] The data acquisition unit uses a laser point scanner, a depth sensor and an optical sensor to scan the upper and lower surfaces, left and right surfaces, and front and back surfaces of the standard processed workpiece and the workpiece to be detected. The depth sensor can measure depth data, and the optical sensor can measure reflected light data. The upper and lower surfaces, left and right surfaces, and front and back surfaces of the standard processed workpiece and the workpiece to be detected are used as standard scanning surfaces and scanning surfaces to be detected, respectively, and the depth data and reflected light data at the scanning points of each standard scanning surface and scanning surface to be detected are obtained to obtain standard data and data to be detected corresponding to multiple standard surfaces and multiple surfaces to be detected.

[0069] The model building unit aligns the scanning surface to be inspected and the standard scanning surface respectively to obtain the reference point, establishes a coordinate system with the reference point as the coordinate origin, sets 0.5*Vmm as the horizontal axis coordinate spacing, and 0.1mm as the vertical axis coordinate spacing, establishes a mapping relationship between the coordinate system of the standard processed workpiece and the processed workpiece to be inspected and the reflected light intensity data and depth data, and obtains the standard data set and the data set to be inspected.

[0070] Registration and construction of coordinate systems are used to avoid the situation where scanned surfaces of different angles and sizes are not placed on the same coordinate system, making subsequent coordinate measurement and projection more accurate.

[0071] The model building unit respectively registers the scanned surface to be detected and the standard scanned surface, including the following steps:

[0072] Step 1: record the standard surface as S and the surface to be detected as T, perform segmentation on each of them, obtain the segmented standard surface S1 and standard surface S2, and obtain the segmented surface to be detected T1 and surface to be detected T2.

[0073] Step 2, respectively calculate the data means of the standard data corresponding to the surface to be tested T1 and the standard surface S1.

[0074] Step 3, respectively calculating the data mean of the to-be-tested data corresponding to the to-be-tested surface T2 and the standard surface S2;

[0075] Step 4: Compare the data mean values ​​of the standard surface S1 and the surface to be detected T1, and the data mean values ​​of the standard surface S2 and the surface to be detected T2 respectively.

[0076] Step 5: If the data means of the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are equal, it means that the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are successfully registered.

[0077] Step 6, if the data means of the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are not equal, the unequal surfaces to be detected have defects, and the defective surface to be detected and the corresponding standard surface are segmented again, and steps 2 to 4 are repeated until the means of the two corresponding parts after segmentation are equal. The middle value of the data in the equal area is set as the reference point, and the standard surface S and the surface to be detected T are aligned.

[0078] Step 7. If after repeating steps 2 to 4, the data means of the standard surface S1 and the surface to be detected T1 and the data means of the standard surface S2 and the surface to be detected T2 are not equal, then return to the state of the last segmentation, set the middle value of the data in the segmented area as the reference point, and align the standard surface S and the surface to be detected T.

[0079] The registration method here is a method based on numerical mean comparison and iterative segmentation, so you need to use software that supports custom registration algorithms or provides open APIs (Application Programming Interface). PCL (Point Cloud Library) is used here. PCL is an open source point cloud processing library that contains many algorithms for 3D images and point cloud processing. The API provided by PCL is used to implement the registration method here in a C++ project.

[0080] The positioning unit calculates the coordinates corresponding to each symbol according to the symbol position in the defect data set. The calculation expression is:

[0081] The horizontal coordinate of the defect = 0.05V* the horizontal order of the defect value

[0082] Defect vertical coordinate = 0.1*defect value vertical order

[0083] get:

[0084] Defect coordinates (defect horizontal coordinate, defect vertical coordinate)

[0085] Output defect location signal.

[0086] The defect location classification unit classifies the defect location coordinates according to the defect location coordinates corresponding to each symbol in the defect location signal:

[0087] The coordinates at X are classified as a convex position coordinate set and output as a convex position coordinate signal.

[0088] The coordinates at Y are classified as a concave crack position coordinate set, and the output is a concave crack position coordinate signal.

[0089] The coordinates at point A are classified as an oil stain location coordinate set, and the output is an oil stain location coordinate signal.

[0090] The comparison, marking, coordinate calculation and classification of the standard data set and the data set to be tested are all performed using MeshLab. The defect locations are identified and marked in MeshLab, and the defect data set is exported. Python is used to read the file exported by MeshLab, and the coordinate data of the defect location is parsed to generate the coordinate signals of the convex position, the concave crack position, and the oil stain position.

[0091] The partition projection unit receives the coordinate signals of the convex position, the concave crack position and the oil stain position, adjusts the laser wavelength to obtain three colors of laser, and projects the convex position, concave crack position and oil stain position onto the surface of the workpiece to be processed according to the coordinate system.

[0092] After the signal format conversion is completed, the convex position coordinate signal, the concave crack position coordinate signal and the oil stain position coordinate signal are sent to the laser transmitter through the hardware interface.

[0093] After receiving the convex position coordinate signal, the concave crack position coordinate signal and the oil stain position coordinate signal, the laser transmitter emits the laser according to the instructions in the convex position coordinate signal, the concave crack position coordinate signal and the oil stain position coordinate signal, pointing to the defect position identified in MeshLab.

[0094] It should be appreciated that embodiments of the present invention may be implemented or enforced by a combination of computer hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The method may be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner - according to the methods and drawings described in the specific embodiments. Each program may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, if desired, the program may be implemented in an assembly or machine language. In any case, the language may be a compiled or interpreted language. In addition, the program may be run on a programmed ASIC for this purpose.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An automatic identification system for surface defects of processed workpieces, characterized in that: Including scanning module, contrast measurement module and projection module: The scanning module is used to scan the standard processed workpiece and the processed workpiece to be detected, obtain standard data and data to be detected, align the scanning surfaces of the standard processed workpiece and the processed workpiece to be detected and establish a coordinate system, wherein the standard data and the data to be detected include reflected light intensity data and depth data, respectively establish mapping relationships between the coordinate systems of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and depth data, and generate a standard data set and a data set to be detected; The comparison measurement module is used to compare the reflected light intensity data and depth data of the standard data set and the data set to be tested, obtain the comparison result, determine the defect type, and calculate the position of the comparison result on the coordinate system; The projection module projects the defect type and position onto the workpiece to be inspected by laser; The scanning module includes a parameter setting unit, a data acquisition unit and a model building unit: The parameter setting unit selects scanning parameters according to the size and surface complexity of the scanned standard workpiece and the workpiece to be inspected; The data acquisition unit scans the standard processed workpiece and the to-be-detected processed workpiece according to the set scanning parameters to obtain a plurality of standard surfaces and standard data corresponding to the standard surfaces and surfaces to be detected and data to be detected corresponding to the surfaces to be detected; The model building unit aligns the surface to be tested and the standard surface accordingly and establishes a coordinate system, and respectively establishes a standard data set and a data set to be tested that are mapped to the coordinate system according to the standard data and the data to be tested; The model building unit aligns the scanned surface to be detected and the standard scanned surface respectively to obtain a reference point, establishes a coordinate system with the reference point as the coordinate origin, sets 0.5*Vmm as the horizontal axis coordinate spacing, and 0.1mm as the vertical axis coordinate spacing, establishes a mapping relationship between the coordinate system of the standard processed workpiece and the processed workpiece to be detected and the reflected light intensity data and the depth data, and obtains a standard data set and a data set to be detected; The model building unit respectively registers the scan surface to be detected and the standard scan surface, including the following steps: Step 1, the standard surface is recorded as S, and the surface to be detected is recorded as T, and each is segmented once to obtain the segmented standard surface S1 and standard surface S2, and the segmented surface to be detected T1 and surface to be detected T2; Step 2, respectively calculating the data mean of the standard data corresponding to the surface to be tested T1 and the standard surface S1; Step 3, respectively calculating the data mean of the to-be-tested data corresponding to the to-be-tested surface T2 and the standard surface S2; Step 4, respectively comparing the data mean values ​​of the standard surface S1 and the surface to be tested T1, and the data mean values ​​of the standard surface S2 and the surface to be tested T2; Step 5: If the data means of the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are equal, it means that the standard surface S1 and the surface to be detected T1 or the standard surface S2 and the surface to be detected T2 are successfully registered; Step 6: If the data means of the standard surface S1 and the surface to be tested T1 or the standard surface S2 and the surface to be tested T2 are not equal, the unequal surfaces to be tested have defects, and the surface to be tested with defects and the corresponding standard surface are segmented twice, and steps 2 to 4 are repeated until the means of the two corresponding parts after segmentation are equal. The middle value of the data in the equal area is set as the reference point, and the standard surface S and the surface to be tested T are aligned; Step 7, if after repeating Step 2 to Step 4, the data means of the standard surface S1 and the surface to be detected T1 and the data means of the standard surface S2 and the surface to be detected T2 are not equal, then return to the state of the last segmentation, set the middle value of the data in the segmented area as the reference point, and align the standard surface S and the surface to be detected T; The comparison measurement module includes a symbol setting unit, a defect comparison unit and a positioning unit: The symbol setting unit sets symbols for the numerical values ​​of the results of the comparison between different to-be-detected data sets and the standard data sets: The position smaller than the depth data in the standard data set is marked as X, the position larger than the depth data in the standard data set is marked as Y, and the position equal to the depth data in the standard data set is marked as Z; The position with different reflected light intensity data from the standard data set is marked as A, and the position with the same reflected light intensity data as the standard data set is marked as B; The defect comparison unit is used to compare the data set to be detected with the standard data set, traverse each row and each column of the standard data set and the data set to be detected, compare the corresponding reflected light intensity data and depth data in the standard data set and the data set to be detected, create a new data set, the new data set has the same data structure as the standard data set and the data set to be detected, and the data type is character. According to the set symbol, the comparison result is filled into the corresponding position of the new data set to obtain the defect data set; The positioning unit calculates the defect position according to the defect data set and generates a defect position signal; The projection module includes a defect location classification unit and a partition projection unit: The defect location classification unit distinguishes different defect areas and generates a plurality of defect location signals of different types according to the set symbol and defect location signal; The partition projection unit projects different types of defect areas onto the workpiece to be inspected by laser according to different types of defect position signals; The defect position classification unit classifies the defect position coordinates according to the defect position coordinates corresponding to each symbol in the defect position signal: The coordinates at X are classified as a convex position coordinate set, and the output is a convex position coordinate signal; The coordinates at Y are classified as a concave crack position coordinate set, and the output is a concave crack position coordinate signal; The coordinates at A are classified as an oil stain location coordinate set, and the output is an oil stain location coordinate signal; The partition projection unit receives the convex position coordinate signal, the concave crack position coordinate signal and the oil stain position coordinate signal, adjusts the laser wavelength to obtain three colors of laser, and projects the convex position, the concave crack position and the oil stain position onto the surface of the workpiece to be processed according to the coordinate system.

2. The automatic identification system for surface defects of a workpiece as claimed in claim 1, characterized in that: The basic parameter settings of the parameter setting unit are: The fixed scanning frequency is 20Hz, and the scanning interval d is divided into three levels, namely 0.1mm, 0.25mm and 0.5mm, which can be adjusted steplessly within the range of 0.1mm-0.5mm; The scanning speed V is divided into three levels, namely 2mm / s, 5mm / s and 12.5mm / s, and can be adjusted steplessly within the range of 2mm / s-12.5mm / s, and changes proportionally with the scanning spacing d.

3. The automatic identification system for surface defects of a workpiece as claimed in claim 1, characterized in that: The data acquisition unit uses a laser point scanner, a depth sensor and an optical sensor to scan the upper and lower surfaces, left and right surfaces and front and back surfaces of the standard processed workpiece and the workpiece to be detected, and uses the upper and lower surfaces, left and right surfaces and front and back surfaces of the standard processed workpiece and the workpiece to be detected as standard scanning surfaces and scanning surfaces to be detected, respectively, to obtain depth data and reflected light data at scanning points of each standard scanning surface and scanning surface to be detected, and obtain standard data and data to be detected corresponding to multiple standard surfaces and multiple surfaces to be detected.

4. The automatic identification system for surface defects of a workpiece as claimed in claim 1, characterized in that: The positioning unit calculates the coordinates corresponding to each symbol according to the symbol position in the defect data set, and the calculation expression is: The horizontal coordinate of the defect = 0.05V* the horizontal order of the defect value Defect vertical coordinate = 0.1*defect value vertical order get: Defect coordinates (defect horizontal coordinate, defect vertical coordinate) Output defect location signal.

Images