CN109470162B - Intelligent detection system and method for micropore shape of oil nozzle based on machine vision - Google Patents

Intelligent detection system and method for micropore shape of oil nozzle based on machine vision Download PDF

Info

Publication number
CN109470162B
CN109470162B CN201811316944.4A CN201811316944A CN109470162B CN 109470162 B CN109470162 B CN 109470162B CN 201811316944 A CN201811316944 A CN 201811316944A CN 109470162 B CN109470162 B CN 109470162B
Authority
CN
China
Prior art keywords
image
micropore
oil nozzle
electron microscope
center
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
Application number
CN201811316944.4A
Other languages
Chinese (zh)
Other versions
CN109470162A (en
Inventor
杨淑珍
陈潜
李涵薇
石宵成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Second Polytechnic University
Original Assignee
Shanghai Second Polytechnic University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Second Polytechnic University filed Critical Shanghai Second Polytechnic University
Priority to CN201811316944.4A priority Critical patent/CN109470162B/en
Publication of CN109470162A publication Critical patent/CN109470162A/en
Application granted granted Critical
Publication of CN109470162B publication Critical patent/CN109470162B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00Measuring arrangements characterised by the use of optical means
    • G01B11/16Measuring arrangements characterised by the use of optical means for measuring the deformation in a solid, e.g. optical strain gauge
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B9/00Instruments as specified in the subgroups and characterised by the use of optical measuring means
    • G01B9/04Measuring microscopes

Abstract

The invention relates to an intelligent detection system and method for a micropore shape of an oil nozzle based on machine vision. The system comprises an electron microscope, an image processing system, a master control system and a rotary object stage, wherein the electron microscope, the image processing system and the rotary object stage are respectively connected with the master control system, the electron microscope is also connected with the image processing system and used for obtaining a primary observation image and a secondary observation image of the micropore of the oil nozzle, the image processing system is used for processing the primary observation image and the secondary observation image, judging whether the micropore is qualified or not and a deformation area of the micropore, displaying a result image in real time, and the master control system is used for controlling the movement of the electron microscope and the rotation of the rotary object stage. The invention realizes the full automation and intellectualization of the micropore detection of the oil nozzle and solves the problems of low manual detection efficiency, high labor intensity, uneven classification quality, high misjudgment rate and the like in the prior art.

Description

Intelligent detection system and method for micropore shape of oil nozzle based on machine vision
Technical Field
The invention relates to the technical field of automatic production, in particular to an intelligent detection system and method for a micropore shape of an oil nozzle based on machine vision.
Background
The diesel engine fuel spray nozzle is provided with a plurality of micro spray holes, and the spray holes are usually machined by firstly adopting electric spark drilling and then adopting liquid extrusion grinding machining to improve the flow coefficient, improve the atomization effect and improve the fuel efficiency. However, the nozzle hole after the press grinding may be deformed. If the orifice of the spray hole is deformed, the fuel atomization effect is greatly influenced, and the fuel efficiency is reduced. Therefore, in order to ensure the quality, the orifice deformation of the spray orifices after extrusion grinding needs to be detected, and the orifice shape of each spray orifice on the nozzle is round and is qualified if no notch is found. The existing detection method is that a digital microscope is fixed, a workpiece to be detected is placed on an object stage, then the workpiece is rotated to each spray hole by hand, the digital microscope is used for carrying out amplification imaging on the micropores on a display, and whether the micropores are qualified or not is directly judged by human eyes according to the outline condition of the spray hole. In addition, due to the fact that the heights of the holes on the spherical surface or the conical surface are different, certain barrel-shaped distortion is caused when some holes are not located in the center of the image according to the imaging principle during detection, and therefore misjudgment is caused. Therefore, the manual detection method has low efficiency, high labor intensity, uneven classification quality and relatively high false judgment rate.
Disclosure of Invention
Aiming at the defects existing in the detection process of the micropore shape of the oil nozzle in the prior art, the invention aims to provide the intelligent detection system and the intelligent detection method of the micropore shape of the oil nozzle based on machine vision. In order to realize the purpose of the invention, the invention adopts the following technical scheme:
an intelligent detection system for the pore shape of the micropore of an oil spray nozzle based on machine vision comprises an electron microscope, an image processing system, a main control system and a rotary object stage for placing the oil spray nozzle, wherein the electron microscope, the image processing system and the rotary object stage are respectively connected with the main control system, the electron microscope is also connected with the image processing system, the electron microscope is used for obtaining a primary observation image R1 and a secondary observation image R2 of the micropore of the oil spray nozzle and transmitting the primary observation image R1 and the secondary observation image R2 to the image processing system, and the image processing system is used for processing the primary observation image R1 to obtain a primary observation image center position coordinate (namely, an electron microscope vision center position coordinate) (X-ray observation image center position coordinate)a,Ya) And the center position coordinates (X) of the micro-holes of the oil nozzleh,Yh) And coordinates (X) of the center position of the primary observed imagea,Ya) And said injector micropore center position coordinate (X)h,Yh) Transmitted to the master control system, and the image processing system is further used for processing the secondary observation image to judgeDistinguishing whether the oil nozzle micropore is qualified or not and the deformation area position of the oil nozzle micropore, displaying a result image in real time, controlling the movement of the electron microscope to enable the oil nozzle micropore to enter the visual field of the electron microscope so that the electron microscope can obtain a primary observation image R1 of the oil nozzle micropore, and according to the primary observation image center position coordinate (X)a,Ya) And said injector micropore center position coordinate (X)h,Yh) And secondarily positioning the electron microscope to move along the X-axis direction and the Z-axis direction so as to adjust the center position of the microscope view field to the center of the micropore of the oil spray nozzle, so that the microscope obtains the secondary observation image R2, and the main control system is also used for controlling the rotary stage to rotate so that the next micropore of the oil spray nozzle moves into the view field of the electron microscope.
Further, the intelligent detection system further comprises a positioning mechanism, wherein the positioning mechanism is used for installing the electron microscope and is connected with the main control system so as to move the electron microscope under the control of the main control system.
Further, the system also comprises a shell, the rotary object stage and the positioning mechanism are installed on the upper surface of the shell, and the main control system is placed inside the shell.
Further, a start detection button is further installed on the side surface or the upper surface of the shell.
Further, the process of processing the primary observation image R1 by the image processing system includes the following steps:
(1) setting a variable X by obtaining the size of the primary observation imagea,YaObtaining the coordinates (X) of the central position of the primary observation imagea,Ya),Xa=A/2、YaB/2, a and B being the length and width of the image, respectively;
(2) HSV conversion is carried out on the primary observation image, the S image is adopted for next processing, a complete round hole image is screened out by adjusting the gray value within the range of 0-30, and the existing micropore of the oil nozzle is extracted by using a select-constants-xld operatorProfile, carrying out primary roundness screening, screening out a region with roundness in the range of (0.95-1), and extracting the coordinates (X) of the center position of the micropore of the oil nozzle by adopting an area _ center operatorh,Yh)。
Further, the process of the main control system for secondarily positioning the electron microscope comprises the following steps:
(1) determining coordinates (X) of the center position of the imagea,Ya) Coordinate (X) with the center position of the microporeh,Yh) Is a distance Δ X off=Xa-Xh、ΔYf=Ya-Yh
(2) Determining a motion track Tr of the electron microscope, specifically: x axis trajectoryZ-axis trajectory:rotation angle of the rotating shaft:wherein, alpha is the cone vertex angle of the oil nozzle, and R is the section radius of the micropore;
(3) and adjusting the center position of the microscope visual field to the center position of the micropore of the oil nozzle according to the motion trail Tr so that the microscope can obtain the secondary observation image R2.
Further, the process of processing the secondary observation image R2 by the image processing system includes the steps of:
(1) converting the three-channel image into an HSV (hue, saturation, value) channel by using a decomplexe operator for the secondary observation image R2, selecting the s image with the clearest value from the three images to adjust the gray value, and extracting a complete round hole when the gray value is in the range of 0-30;
(2) filling the circular hole area with red, dividing the picture area by using a connection operator, and dividing the unconnected area into independent areas to obtain independent areas; filling the image when the part is not completely filled after the segmentation and the edge part has a gap, and determining that the image is a closed graph; performing morphological processing on an image, performing corrosion and expansion operation to obtain an image R3, enabling the edge of an R3 image to become smooth, breaking off narrow discontinuities and eliminating thin protrusions, performing the operation, screening an image area by using a characteristic histogram, completely covering a round hole area after screening is completed, extracting only a circle with the roundness not lower than 0.75 by using a select-shape-xld operator, taking the center point of an edge pixel to obtain a micropore outline L1 closer to the reality, performing roundness screening, judging that the roundness is qualified in the range of (0.9-1), and fitting a perfect circle L2 by using an edge fitting segmentation method if the roundness is unqualified,
(3) after the operations are finished, filling L1 and L2 to obtain a micropore area D1 and a fitting circle area D2, and performing area subtraction operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
Further, the step of fitting the perfect circle L2 by the edge fitting segmentation method specifically includes: setting huber weight value with the formula of
The severely deformed portion is filtered to find the center point of the original micropore, and a xld circle, namely a perfect circle L2, is fitted to the preliminarily processed image R3.
The invention also provides an intelligent detection method for the micropore shape of the oil nozzle by using the intelligent detection system for the micropore shape of the oil nozzle based on machine vision, which is characterized by comprising the following steps:
s101, fixing a tested oil nozzle on a rotary objective table;
s102, starting a detection button, moving an electron microscope from an original point to the position near the oil nozzle to be detected to focus with a micropore of the oil nozzle to be detected under the control of a main control system, and shooting the micropore to obtain a primary observation image R1;
s103, processing the primary micropore observation image by the image processing system to obtain a primary observation image center position coordinate (namely, the electron microscope view center position)Coordinates) (X)a,Ya) And the coordinates (X) of the center position of the microporeh,Yh);
S104, the main control system observes the central position coordinate (X) of the image according to the first timea,Ya) And the micropore central position coordinate (X)h,Yh) Secondarily positioning the electron microscope to move the center of the field of view of the electron microscope to the center position of the micropore, and shooting the micropore again to obtain a secondary observation image R2;
s105, the image processing system processes the secondary observation image to judge whether the micropores are qualified or not and the position of a hole deformation area, and displays a result image in real time to finish the detection of one micropore;
s106, the rotary object stage rotates under the control of the master control system, the measured oil nozzle moves to enable the electron microscope and the next micropore of the measured oil nozzle to be focused, the micropore is shot to obtain an observation image, and the steps S103-S106 are repeated until all micropores on the workpiece are sequentially detected;
s107, after all micropores on the oil nozzle to be tested are detected, if all micropores are qualified, the oil nozzle to be tested is qualified, otherwise, the oil nozzle to be tested is not qualified, and the image processing system displays the result;
and S108, the electron microscope returns to the original point under the control of the main control system, and the oil nozzle to be detected is taken out to complete the detection of one oil nozzle.
Further, the intelligent detection system further comprises a positioning mechanism, wherein the positioning mechanism is used for installing the electron microscope and is connected with the main control system so as to move the electron microscope under the control of the main control system.
Further, the system also comprises a shell, the rotary object stage and the positioning mechanism are installed on the upper surface of the shell, and the main control system is placed inside the shell.
Further, the process of processing the primary observation image R1 by the image processing system includes the following steps:
(1) setting a variable X by obtaining the size of the primary observation imagea,YaObtaining the coordinates (X) of the central position of the primary observation imagea,Ya),Xa=A/2、Ya=B/2,
(2) HSV conversion is carried out on the primary observation image, S image is adopted for next processing, a complete round hole image is screened out by adjusting the gray value within the range of 0-30, the selection-constants-xld operator is used for extracting the existing outline of the micropore of the oil nozzle, the selection-shape-xld operator is used for carrying out preliminary roundness screening, the area with the roundness within the range of (0.95-1) is screened out, and the area _ center operator is adopted for extracting the central position coordinate (X) of the micropore of the oil nozzleh,Yh)。
Further, the process of secondarily positioning the electron microscope by the master control system comprises the following steps:
(1) determining coordinates (X) of the center position of the imagea,Ya) Coordinate (X) with the center position of the microporeh,Yh) Is a distance Δ X off=Xa-Xh、ΔYf=Ya-Yh
(2) Determining a motion track Tr of the electron microscope, specifically: x axis trajectoryZ-axis trajectory:rotation angle of the rotating shaft:wherein, alpha is the cone vertex angle of the oil nozzle, and R is the section radius of the micropore;
(3) and adjusting the center position of the microscope visual field to the center position of the micropore of the oil nozzle according to the motion trail Tr so that the microscope can obtain the secondary observation image R2.
Further, the process of processing the secondary observation image R2 by the image processing system includes the steps of:
(1) converting the three-channel image into an HSV (hue, saturation, value) channel by using a decomplexe operator for the secondary observation image R2, selecting the s image with the clearest value from the three images to adjust the gray value, and extracting a complete round hole when the gray value is in the range of 0-30;
(2) filling the circular hole area with red, dividing the picture area by using a connection operator, and dividing the unconnected area into independent areas to obtain independent areas; filling the image when the part is not completely filled after the segmentation and the edge part has a gap, and determining that the image is a closed graph; performing morphological processing on an image, performing corrosion and expansion operation to obtain an image R3, enabling the edge of an R3 image to become smooth, breaking off narrow discontinuities and eliminating thin protrusions, performing the operation, screening an image area by using a characteristic histogram, completely covering a round hole area after screening is completed, extracting only a circle with the roundness not lower than 0.75 by using a select-shape-xld operator, taking the center point of an edge pixel to obtain a micropore outline L1 closer to the reality, performing roundness screening, judging that the roundness is qualified in the range of (0.9-1), and fitting a perfect circle L2 by using an edge fitting segmentation method if the roundness is unqualified,
(3) after the operations are finished, filling L1 and L2 to obtain a micropore area D1 and a fitting circle area D2, and performing area subtraction operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
Further, the step of fitting the perfect circle L2 by the edge fitting segmentation method specifically includes: setting huber weight value with the formula of
The severely deformed portion is filtered to find the center point of the original micropore, and a xld circle, namely a perfect circle L2, is fitted to the preliminarily processed image R3. The edge fitting segmentation method for fitting the perfect circle is particularly suitable for the situation that the micropore has large deformation, and can solve the problem that the center position directly fitted by adopting the xld fitting mode deviates greatly from the actual situation.
The intelligent detection system and method for the shape of the micropore of the oil nozzle based on machine vision solve the problems of low efficiency, uneven classification quality and the like of manual detection of the micropore of the workpiece. Compared with the prior art, the invention achieves the following beneficial effects:
the intelligent micropore shape detection system based on machine vision adopts an electron microscope to obtain the micropore image of the oil nozzle, has large magnification and high pixel, effectively improves the imaging quality of micropores, solves the problems that in the prior art, the workpiece is required to be rotated and focused continuously by manpower, fatigue is easily generated during long-time detection, the detection efficiency is influenced, and simultaneously solves the problems that the detection effect is completely obtained by subjective identification through human eye observation, the judgment standard is different, and the error is relatively high.
The invention adopts an edge fitting and dividing method based on Huber weight aiming at the larger deformation hole, distinguishes the larger deformation area from the main profile, more accurately determines the center position of the micropore, solves the problem of larger deviation of the center position of the larger deformation micropore, and reduces the possibility of misjudgment.
The invention can also identify and mark the position of the processed micropore deformation area, compared with manual detection, the detection result is clearer and more definite, and better analysis data and basis can be provided for finding out the deformation reason.
The invention obtains the final processing image by secondary positioning alignment, and solves the problem of error judgment caused by additional distortion of hole shapes because the heights of all micropores are different and some holes are not in the center of the image on one circle of the workpiece.
The invention provides a brand-new intelligent solution for the micropore detection of the engine oil nozzle, which realizes the full automation and the intelligence of the micropore detection of a workpiece based on machine vision, reduces the labor intensity, the misjudgment rate and the production cost, and improves the production efficiency and the classification quality.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the embodiments or the drawings in the prior art, and it is obvious that the drawings described below are only some embodiments of the present invention, and that other drawings can be obtained by those skilled in the art without inventive efforts.
Fig. 1 is a schematic structural diagram of an intelligent detection system for the shape of a micro-hole of an oil nozzle based on machine vision according to an embodiment of the invention.
FIG. 2 is a single observation image R1 of a microwell obtained using the detection system of the present invention.
FIG. 3 is a secondary observation image R2 of a microwell obtained using the detection system of the present invention.
The micropore profile L1 extracted in FIG. 4 has a roundness of 0.94, which is a qualified circular hole.
Fig. 5 is a perfect circle L2 obtained after fitting.
FIG. 6 is a single observation image of a micro-hole having a severe deformation after processing obtained by the inspection system of the present invention.
FIG. 7 extracts micropore profile L1.
FIG. 8 shows that the center position of the fitting circle D2 can be accurately determined according to the edge fitting segmentation method
FIG. 9 shows the deformation region obtained by the region subtraction according to D1 and D2.
Detailed Description
In the present invention, referring to fig. 1, a plane on which an electron microscope moves in parallel along a horizontal plane is defined as a plane on which an X axis and a Y axis are located, and a plane on which the electron microscope moves up and down along a vertical horizontal plane is defined as a plane on which a Z axis is located. The position of the electron microscope before the start of the intelligent detection system of the invention or the position coordinates of the field of view of the electron microscope when the electron microscope does not start moving after the start of the intelligent detection system are set as the origin.
In a preferred embodiment of the invention, the intelligent detection system for the micropore shape of the oil nozzle based on machine vision comprises an electron microscope, an image processing system, a main control system, a rotary stage and a positioning mechanism. The oil nozzle of the workpiece to be measured is fixed on the rotary object stage; the electron microscope is fixed on the positioning mechanism and used for obtaining a micropore under the control of the master control systemThe secondary observation image R1; the image processing system is connected with the electron microscope and is used for receiving the primary observation image R1 and processing the primary observation image R1 to obtain the central position coordinate (X) of the primary observation image (namely, the central position coordinate of the field of view of the electron microscope)a,Ya) And the center position coordinates (X) of the microporesh,Yh) And coordinates (X) of the center of field of view of the primary observation image R1a,Ya) And micropore center position coordinate (X)h,Yh) Transmitting to the master control system; the main control system is connected with the image processing system and used for receiving the coordinates sent by the image processing system and calculating the offset distance delta Xf、ΔYfDetermining a motion track Tr of the electron microscope, specifically: x axis trajectoryZ-axis trajectory:rotation angle of the rotating shaft:the method comprises the following steps of firstly, carrying out secondary positioning on an electron microscope according to a movement track Tr, wherein alpha is the conical vertex angle of an oil nozzle, R is the radius of a section where a micropore is located, and adjusting the center of the micropore to the center of the view field of the electron microscope; then the electron microscope obtains a micropore secondary observation image R2 after secondary positioning, processes the micropore secondary observation image R2, judges whether the pore shape of the detected micropore is qualified or not, and identifies the position of a deformation area; and the main control system also controls the rotary object stage to move the micropore of the next oil nozzle to be detected into the field of view of the electron microscope for judgment until the holes on the workpiece to be detected are sequentially detected.
In a preferred embodiment of the present invention, the image processing system processes the primary observed image R1, and processes the observed image R1 using a watershed segmentation algorithm based on image sharpness attributes as follows: by obtaining the image size, a variable X is seta,YaObtaining the coordinates (X) of the center point of the image R1a,Ya),Xa=A/2、YaAnd (2) A, B, namely B/2, the length and the width of the picture, then performing HSV conversion on the image R1 to obtain the central position of the micropore, performing next processing by adopting an S image, screening out a complete round hole image by adjusting the gray value within the range of 0-30, and applying
Extracting the existing outline of the micropore of the oil nozzle by a select-constants-xld operator, performing primary roundness screening to screen out an area with the roundness within the range of (0.75-1) under the condition that the area identification range is large due to the influence of insufficient brightness of the edge area, and extracting the central point coordinate (X) of the micropore of the oil nozzle by an area _ center operatorh,Yh)。
In a preferred embodiment of the present invention, the processing, by the image processing system, the secondary positioned micropore observation image R2, and determining whether the measured micropore shape is qualified specifically includes: and (3) converting the three-channel image into an HSV (hue, saturation, value) channel by using a decomplexe operator for the image R2, selecting the s image with the clearest value from the three images to perform gray value adjustment range, and extracting a complete round hole when the gray value obtained by experiments is in the range of 0-30. And filling the circular hole area with red, dividing the picture area by using a connection operator, and dividing the unconnected area into independent areas to obtain the independent areas. After segmentation, the situation that the part is not completely filled and the edge part has a gap is found, so that the image needs to be filled and is determined to be a closed graph. The image was morphologically processed by erosion followed by dilation to obtain an image R3, rounding the edges of the R3 image, breaking narrow discontinuities and eliminating thin protrusions. After a large number of actual image experiments, it is found that metal residues possibly remain at the edge after the oil nozzle is machined, subsequent identification is influenced, the image area is screened by using the characteristic histogram after the operation is executed, the round hole area is completely covered after the screening is finished, the extraction of micropores is incomplete only by means of gray value adjustment, the brightness of the edge area of some pictures is low due to shooting problems, if the false judgment condition of multiple contours occurs when the subsequent contour extraction is only performed by gray value adjustment, and for accurately extracting the original micropores, a select-shape-xld operator can be used for extracting circles with the roundness not lower than 0.75. Taking the center point of the edge pixel results in a closer approximation to the actual pinhole profile L1. And (3) carrying out roundness screening, judging that the roundness is qualified within the range of (0.9-1), if the roundness is not qualified, fitting a perfect circle L2 by using an edge fitting segmentation method, filling L1 and L2 to obtain a micropore area D1 and a fitted circle area D2, and carrying out region subtraction operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
In a preferred embodiment of the present invention, the edge fitting segmentation method is used to fit a perfect circle, and the extracted micropore profile L1 is compared with the perfect circle to identify the location of the deformation region, which specifically includes: aiming at the problem that the center position fitted by directly adopting the xld fitting mode is greatly deviated from the actual center position when the micropore is greatly deformed, the method adopts edge fitting segmentation and sets huber weight value according to the formula
The severely deformed portion can be filtered to find the center point of the original micropore, and then a xld circle, namely a right circle L2, is fitted to the preliminarily processed image R3.
In a preferred embodiment of the present invention, the process of controlling the next detected micro-well to move to the field of the electron microscope for judgment by the master control system specifically includes: and when the image processing system judges whether the current micropore is qualified, the main control system controls the rotary object stage motor to rotate for a certain angle around the W shaft, the next detected micropore is moved to the field of view of the electron microscope, then a micropore image R1 is obtained and processed, secondary positioning is carried out, a secondary observation image R2 of the micropore after the secondary positioning is obtained, and whether the shape of the micropore is qualified is judged.
In a preferred embodiment of the present invention, the rotary stage is specifically configured to clamp the workpiece to be tested under the control of the main control system, and can rotate the workpiece by a predetermined angle, so that a plurality of holes evenly distributed on the circumference of the workpiece are sequentially detected.
In a preferred embodiment of the invention, the positioning mechanism comprises an X axis and a Z axis, the two axes adopt high-precision transmission structures such as a lead screw and the like, and the motor adopts an integrated stepping servo motor with encoder feedback, so that the structure is compact, the motion precision is high, and the cost performance is high. This positioning mechanism specifically is used for: at the beginning of measurement, after a workpiece is placed and clamped, controlling X, Z a shaft motor to rotate according to a motor instruction and parameters sent by a master control system, and moving the electron microscope to be close to the workpiece (an oil nozzle) from a zero position (an original point) and focus to a 1 st micropore; controlling a motor in the direction X, Z to operate according to the motion track Tr of the secondary positioning, moving the center of the field of view of the electron microscope to the center position coordinate of the round hole, and completing the secondary positioning, thereby ensuring that the hole in the hole shape discrimination image R2 obtained for the second time can be in the center position of the field of view, and solving the problem of barrel distortion error caused by that the image of the hole deviates from the center of the field of view and is even positioned at the edge of the field of view because the holes are not positioned at the same axial height; and after all micropores on the workpiece are detected, controlling the electron microscope to be far away from the workpiece and return to the zero position so as to take out the workpiece and prepare for measuring the next workpiece.
In a preferred embodiment of the invention, the electron microscope comprises an annular LED light source, with a magnification of more than 200 times and more than 200 tens of thousands of pixels.
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further explained below with the accompanying drawings.
Example 1
Referring to fig. 1, in the example shown in the figure, the intelligent detection system for the micropore shape of the oil nozzle comprises an electron microscope 1, an image processing system 2, a main control system 3, a rotary stage 4, a positioning mechanism 5 and a shell 6. The electron microscope 1 is fixed to the positioning mechanism 5. The rotary object stage 4 and the positioning mechanism 5 are fixed on the upper surface of the shell 6, the main control system 3 is placed inside the shell 6, and the image processing system 2 can be fixedly connected with the shell 6 or can be independent of each other. The positioning mechanism 5, the image processing system 2 and the rotary object stage 4 are respectively connected with the main control system 3.
The electron microscope 1 is further connected with an image processing system 2, and is configured to obtain a primary observation image R1 and a secondary observation image R2 of the fuel injector micro-hole, and send the primary observation image R1 and the secondary observation image R2 to the image processing system 2.
The image processing system 2 receives and processes the primary observed image R1 to obtain primary observed image center position coordinates (i.e., electron microscope 1 field-of-view center position coordinates) (X)a,Ya) And the center position coordinates (X) of the micro-holes of the oil nozzleh,Yh) And transmits the two center position coordinates to the main control system 3. The image processing system 2 also receives and processes the secondary observation image R2 to judge whether the oil nozzle micropore is qualified or not, the deformation area position of the oil nozzle micropore and displays the result image in real time,
the main control system 3 receives the coordinates (X) of the central position of the primary observation imagea,Ya) And the center position coordinates (X) of the micro-holes of the oil nozzleh,Yh) Determining the motion trail Tr of the electron microscope 1, and secondarily positioning the electron microscope 1 to adjust the center of the micropore of the oil nozzle to the center of the visual field of the microscope 1. The main control system 3 also controls the rotary stage 4 to rotate so that the next micropore of the oil nozzle moves to the field of the electron microscope 1 for detection.
When the device is used, a measured oil nozzle is placed on the rotary object stage 4 and fixed by a clamp, a detection button is pressed down, the main control system 3 controls the clamp to clamp the oil nozzle, and simultaneously controls the positioning mechanism 5 to enable the view field of the electron microscope 1 to start moving from a zero position (original point) along an X axis and a Z axis, so that the electron microscope 1 moves to the vicinity of the measured oil nozzle and focuses on the No. 1 micropore of the measured oil nozzle, the micropore is shot to obtain a clear pore-shaped image of the micropore, namely a primary observation image R1, and the image is transmitted to the image processing system 2; the image processing system 2 processes the primary observation image R1 of the microhole to obtain the coordinates (X) of the center position of the primary observation imagea,Ya) And micropore center position coordinate (X)h,Yh) And then transmitted to the main control system 3; the main control system 3 calculates the coordinates (X) of the center position of the image based on the primary observationa,Ya) And the center position coordinates (X) of the microporesh,Yh) Determining a motion track Tr of the electron microscope 1, then controlling the electron microscope 1 to move along the X-axis and Z-axis directions through a positioning mechanism 5 according to the motion track Tr, and moving the center of the field of view of the electron microscope 1 to the center of the detected micropore; the electron microscope 1 again photographs the micropores, obtains an image of the micropores again, called a secondary observation image R1, and transmits the image to the image processing system 2; the image processing system 2 processes the secondary observation image R2, judges whether the detected micropore is qualified or not and the position of the pore deformation region, and displays the result image R3 in real time to finish the detection of one micropore; the rotary object stage 4 rotates according to the instruction of the master control system 3, the oil spray nozzle to be detected moves, so that the next micropore enters the visual field of the microscope 1, and the steps are repeated until all micropores on the oil spray nozzle are detected in sequence; after all micropores on the tested oil nozzle are detected, if all micropores are qualified, the workpiece is qualified, otherwise, the workpiece is unqualified, and the result is displayed by the image processing system 2; and then, the main control system 3 controls the positioning mechanism 5 to move along the X-axis direction and the Z-axis direction, the positioning mechanism returns to the zero position, and finally, the workpiece is taken out and placed in different places according to the detection result, so that the detection of the oil nozzle is finished.
The acquisition of the image can be triggered by the outside, and can also be automatically acquired by the microscope 1 after the parameters are set. Considering that the micropore is not at the center of the microscope visual field during the initial image (primary observation image) acquisition, the image is acquired again by adopting a secondary positioning method, and the micropore is ensured to be at the center of the microscope visual field. For example, fig. 2 shows a primary observation image R1 of a micro-hole of the fuel injection nozzle obtained by the microscope 1, and fig. 3 shows a secondary observation image R2 of the micro-hole obtained by secondarily positioning the microscope 1.
The secondary positioning process of the microscope 1 by the main control system 3 specifically includes the following processes:
the image processing system 2 sets the variable X by acquiring the primary observation image R1a,YaObtaining the image center point (X)a,Ya),Xa=A/2、YaB/2, then obtaining the central position of the micropore, and firstly aligning the figurePerforming HSV conversion, performing next processing by adopting an S image, screening out a complete round hole image by adjusting the gray value within the range of 0-30, extracting the existing outline of the micropore of the oil nozzle by using a select-constants-xld operator, screening out the area with the roundness within the range of (0.95-1) by performing primary roundness screening under the condition that the identification range of the area is large due to the influence of insufficient brightness of the edge area, and extracting the central point (X) of the micropore of the oil nozzle by using an area _ center operatorh,Yh) And delivered to the control system 3. The control system 3 calculates the position distance delta X between the center point of the micropore and the center point of the picture (image)f=Xa-Xh,ΔYf=Ya-YhAnd determining a motion track Tr of the electron microscope 1 according to the distance so as to move a lens to the center of the micropore to take a picture, wherein the Tr specifically comprises the following steps: x axis trajectoryZ-axis trajectory:rotation angle of the rotating shaft:wherein, alpha is the cone vertex angle of the oil nozzle, and R is the section radius of the micropore; and finally, adjusting the center position of the field of view of the microscope (1) to the center position of the micropore of the oil nozzle according to the motion trail Tr so that the microscope (1) can obtain the secondary observation image R2.
The image processing system 2 processes the secondary observation image R2 to determine whether the measured micropore shape is acceptable includes the following processes:
the image processing system 2 converts the three-channel image into an HSV channel by using a decompose operator for the image R2, selects the s image with the clearest value from the three images to carry out gray value adjustment range, and can extract a complete round hole when the gray value obtained by experiments is in the range of 0-30. And filling the circular hole area with red, dividing the picture area by using a connection operator, and dividing the unconnected area into independent areas to obtain the independent areas. After segmentation, the situation that the part is not completely filled and the edge part has a gap is found, so that the image needs to be filled and is determined to be a closed graph. The image was morphologically processed by erosion followed by dilation to obtain an image R3, rounding the edges of the R3 image, breaking narrow discontinuities and eliminating thin protrusions. After a large number of actual image experiments, it is found that metal residues possibly remain at the edge after the oil nozzle is machined, subsequent identification is influenced, the image area is screened by using the characteristic histogram after the operation is executed, the round hole area is completely covered after the screening is finished, the extraction of micropores is incomplete only by means of gray value adjustment, the brightness of the edge area of some pictures is low due to shooting problems, if the false judgment condition of multiple contours occurs when the subsequent contour extraction is only performed by gray value adjustment, and for accurately extracting the original micropores, a select-shape-xld operator can be used for extracting circles with the roundness not lower than 0.75. Taking the center point of the edge pixel results in a closer approximation to the actual pinhole profile L1. And (4) carrying out roundness screening, and judging that the roundness is qualified within the range of (0.9-1). Fig. 4 shows the contour L1 of the extracted micropore, and the roundness of the micropore is 0.94, which is a qualified round hole.
If the circle is not qualified, fitting a perfect circle L2 by using an edge fitting segmentation method, which specifically comprises the following steps: aiming at the problem that the center position fitted by directly adopting an xld fitting mode is greatly deviated from the actual center position when the micropore is greatly deformed, the invention adopts an edge fitting segmentation method, sets a huber weight value and sets the huber weight value, and the formula is as follows:
the severely deformed portion can be filtered to find the center point of the original micropore, and then a xld circle, namely a right circle L2, is fitted to the preliminarily processed image R3. After the operations are finished, filling L1 and L2 to obtain a micropore area D1 and a fitting circle area D2, and performing area subtraction (21) operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
Fig. 5 shows a circular perfect circle L2 of xld fitted on the basis of fig. 4.
As shown in fig. 6, the micro-holes (primary observation images) obtained by the microscope 1 and deformed seriously after processing were rejected micro-holes.
Fig. 7 shows L1 extracted by the image processing system 2 for the defective micro-hole profile.
Fig. 8 shows that the image processing system 2 can accurately determine the center position of the fitting circle D2 according to the edge fitting segmentation method.
Fig. 9 shows the deformed regions obtained by performing the region subtraction according to D1 and D2. Filling L1 and L2 to obtain a micropore area D1 and a fitting circle area D2, and performing area subtraction operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (9)

1. An intelligent detection system for the micropore shape of an oil nozzle based on machine vision is characterized by comprising an electron microscope (1), an image processing system (2), a main control system (3) and a rotary object stage (4) for placing the oil nozzle,
the electron microscope (1), the image processing system (2) and the rotary object stage (4) are respectively connected with the main control system (3), the electron microscope (1) is also connected with the image processing system (2),
the electron microscope (1) is used for obtaining a primary observation image and a secondary observation image R2 of the micropore of the oil nozzle and transmitting the primary observation image and the secondary observation image R2 to the image processing system (2),
the image processing system (2) is used for processing the primary observation image to obtain a primary observation image center position coordinate (X)a,Ya) And the center position coordinates (X) of the micro-holes of the oil nozzleh,Yh) And coordinates (X) of the center position of the primary observed imagea,Ya) And said injector micropore center position coordinate (X)h,Yh) The secondary observation image R2 is processed by the image processing system (2) to judge whether the oil nozzle micropore is qualified or not and the deformation area position of the oil nozzle micropore, and the result image is displayed in real time,
the main control system (3) is used for controlling the movement of the electron microscope (1) to enable the oil nozzle micropore to enter the visual field of the electron microscope (1) so that the electron microscope (1) can obtain a primary observation image of the oil nozzle micropore, and the central position coordinate (X) of the primary observation image is used fora,Ya) And said injector micropore center position coordinate (X)h,Yh) Secondarily positioning the electron microscope (1) to move along the X-axis direction and the Z-axis direction so as to adjust the center position of the view field of the electron microscope (1) to the center of the micropore of the oil nozzle so that the electron microscope (1) obtains the secondary observation image R2, and the main control system (3) is also used for controlling the rotary object stage (4) to rotate so that the next micropore of the oil nozzle moves to the view field of the electron microscope (1),
the intelligent detection system for the shape of the micropore of the oil nozzle also comprises a positioning mechanism (5), wherein the positioning mechanism (5) is used for installing the electron microscope (1) and is connected with the main control system (3) so as to move the electron microscope (1) under the control of the main control system (3).
2. The intelligent detection system for the shape of the micro-hole of the oil nozzle based on the machine vision is characterized by further comprising a shell (6), wherein the rotary stage (4) and the positioning mechanism (5) are installed on the upper surface of the shell (6), and the main control system (3) is placed inside the shell (6).
3. The intelligent detection system for the micro-hole shape of the oil nozzle based on the machine vision is characterized in that the image processing system (2) processes the primary observation image and comprises the following steps:
(1) setting a variable X by acquiring the image sizea,YaObtaining the coordinates (X) of the center position of the imagea,Ya),Xa=A/2、Ya= B/2, A and B being the length and width of the image, respectively,
(2) HSV conversion is carried out on the image, S image is adopted for next processing, a complete round hole image is screened out by adjusting the gray value within the range of 0-30, the existing outline of the micropore of the oil nozzle is extracted by using a select-constants-xld operator, preliminary roundness screening is carried out, the area with the roundness within the range of 0.95-1 is screened out, and the area-center operator is adopted to extract the central position coordinate (X) of the micropore of the oil nozzleh,Yh)。
4. The intelligent detection system for the micro-hole shape of the oil nozzle based on the machine vision as claimed in claim 1, wherein the process of the main control system (3) to secondarily position the electron microscope (1) comprises the following steps: (1) finding the coordinates (X) of the center position of the imagea,Ya) And the center position coordinate (X) of the microporeh,Yh) Is a distance Δ X off=Xa-Xh、△Yf=Ya-Yh
(2) Determining a motion track Tr of the electron microscope (1), specifically: x axis trajectory(ii) a Z-axis trajectory:the rotation angle of the rotating shaft is as follows:wherein alpha is the cone apex angle of the oil nozzle, R is the section radius of the micropore,
(3) and adjusting the center position of the view field of the electron microscope (1) to the center position of the micropore of the oil nozzle according to the motion trail Tr so that the electron microscope (1) can obtain the secondary observation image R2.
5. The intelligent detection system for the micro-hole shape of the oil nozzle based on the machine vision is characterized in that the processing of the secondary observation image R2 by the image processing system (2) comprises the following steps:
(1) converting the three-channel image into an HSV (hue, saturation, value) channel by using a decomplexe operator for the secondary observation image R2, selecting the s image with the clearest value from the three images for gray value adjustment, and extracting a round hole when the gray value is in the range of 0-30;
(2) filling red in a round hole area, segmenting a picture area by using a connection operator, segmenting the unconnected area into independent areas to obtain independent areas, filling the image when the parts are not completely filled after segmentation and the edge part is notched, determining that the image is a closed graph, performing morphological processing on the image, performing expansion operation after corrosion to obtain an image R3, enabling the edge of the R3 image to become smooth, breaking narrow gaps and eliminating thin protrusions, screening the image area by using a characteristic histogram after the operation is finished, completely covering the round hole area after the screening is finished, extracting only a circle with the roundness not less than 0.75 by using a select-shape-xld operator, taking the central point of an edge pixel to obtain a micropore profile L1 which is closer to reality, performing roundness screening, and judging that the roundness is qualified within the range of 0.9-1, if not, fitting a perfect circle L2 by using an edge fitting and dividing method;
(3) after the operations are finished, filling L1 and L2 to obtain a micropore area D1 and a fitting circle area D2, and performing area subtraction operation on D1 and D2 by using a difference operator to identify the position of a deformation area.
6. The intelligent detection system for the shape of the micro-hole of the oil nozzle based on the machine vision of claim 5, wherein the specific step of fitting the perfect circle L2 by the edge fitting segmentation method comprises: setting huber weight values, filtering the severely deformed part, finding out the center point of the original micropore, and fitting a xld circle, namely a right circle L2, to the preliminarily processed image R3.
7. An intelligent detection method for the shape of a micro-hole of an oil nozzle by using an intelligent detection system for the shape of the micro-hole of the oil nozzle based on machine vision is characterized by comprising the following steps:
s101, fixing the oil nozzle to be tested on a rotary objective table (4);
s102, starting a detection button, moving an electron microscope (1) from an original point to the position near the oil nozzle to be detected to focus with a micropore of the oil nozzle to be detected under the control of a main control system (3), and shooting the micropore to obtain an observation image;
s103, the image processing system (2) processes the micropore primary observation image to obtain a primary observation image center position coordinate (X)a,Ya) And the coordinates (X) of the center position of the microporeh,Yh);
S104, the main control system (3) observes the central position coordinate (X) of the image according to the first timea,Ya) And the micropore central position coordinate (X)h,Yh) Secondarily positioning the electron microscope (1) to move the center of the field of view of the electron microscope (1) to the center position of the micropore, and shooting the micropore again to obtain a secondary observation image R2;
s105, the image processing system (2) processes the secondary observation image R2 to judge whether the micropores are qualified or not and judge the positions of the deformed areas of the micropores, and displays the result image in real time to finish the detection of one micropore;
s106, the rotary object stage (4) rotates under the control of the main control system (3), the measured oil nozzle moves to enable the electron microscope (1) to be focused with the next micropore of the measured oil nozzle, the micropore is shot to obtain an observation image, and the steps S103-S106 are repeated until all micropores on the workpiece are sequentially detected;
s107, after all micropores on the oil nozzle to be tested are detected, if all micropores are qualified, the oil nozzle to be tested is qualified, otherwise, the oil nozzle to be tested is not qualified, and the image processing system (2) displays the result;
and S108, the electron microscope (1) returns to the original point under the control of the main control system (3), and the detected oil nozzle is taken out to complete the detection of one oil nozzle.
8. The intelligent detection method for the intelligent detection system for the micro-hole shape of the oil nozzle based on the machine vision of claim 7 is characterized in that the intelligent detection system for the micro-hole shape of the oil nozzle further comprises a positioning mechanism (5), and the positioning mechanism (5) is used for installing the electron microscope (1) and is connected with the main control system (3) so as to move the electron microscope (1) under the control of the main control system (3).
9. The intelligent detection method for the intelligent detection system for the micro-hole shape of the oil nozzle based on the machine vision is characterized in that the intelligent detection system for the micro-hole shape of the oil nozzle further comprises a shell (6), the rotary stage (4) and the positioning mechanism (5) are installed on the upper surface of the shell (6), and the main control system (3) is placed inside the shell (6).
CN201811316944.4A 2018-11-07 2018-11-07 Intelligent detection system and method for micropore shape of oil nozzle based on machine vision Active CN109470162B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811316944.4A CN109470162B (en) 2018-11-07 2018-11-07 Intelligent detection system and method for micropore shape of oil nozzle based on machine vision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811316944.4A CN109470162B (en) 2018-11-07 2018-11-07 Intelligent detection system and method for micropore shape of oil nozzle based on machine vision

Publications (2)

Publication Number Publication Date
CN109470162A CN109470162A (en) 2019-03-15
CN109470162B true CN109470162B (en) 2020-11-10

Family

ID=65672050

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811316944.4A Active CN109470162B (en) 2018-11-07 2018-11-07 Intelligent detection system and method for micropore shape of oil nozzle based on machine vision

Country Status (1)

Country Link
CN (1) CN109470162B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111504193A (en) * 2020-05-07 2020-08-07 山东菏鲁电子科技有限公司 Automatic detection method for data line plug size

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102519382A (en) * 2011-12-22 2012-06-27 山东泰山钢铁集团有限公司 Fixed point marking method for deformation detection of cold-rolled steel strip
CN103868479A (en) * 2014-03-18 2014-06-18 同济大学 Automatic nozzle hole structure parameter measuring method
CN105102923A (en) * 2013-01-25 2015-11-25 沃思测量技术股份有限公司 Method and device for determining the geometry of structures by means of computer tomography
CN105844282A (en) * 2016-06-14 2016-08-10 上海贝特威自动化科技有限公司 Method for detecting defects of fuel injection nozzle O-Ring through line scanning camera
CN106895761A (en) * 2017-03-17 2017-06-27 柳州日高滤清器有限责任公司 A kind of plastic air intake manifold cubing
CN108257171A (en) * 2018-01-09 2018-07-06 江苏科技大学 Car radar assembling aperture detection method based on light vision
CN108414358A (en) * 2018-05-23 2018-08-17 中国原子能科学研究院 A kind of device measuring tensile sample elongation after fracture and the contraction percentage of area
CN209342039U (en) * 2018-11-07 2019-09-03 上海第二工业大学 A kind of atomizer micropore pile bore checking system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105143851B (en) * 2013-04-12 2019-03-01 贝克顿·迪金森公司 Automated setting for cell sorting

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102519382A (en) * 2011-12-22 2012-06-27 山东泰山钢铁集团有限公司 Fixed point marking method for deformation detection of cold-rolled steel strip
CN105102923A (en) * 2013-01-25 2015-11-25 沃思测量技术股份有限公司 Method and device for determining the geometry of structures by means of computer tomography
CN103868479A (en) * 2014-03-18 2014-06-18 同济大学 Automatic nozzle hole structure parameter measuring method
CN105844282A (en) * 2016-06-14 2016-08-10 上海贝特威自动化科技有限公司 Method for detecting defects of fuel injection nozzle O-Ring through line scanning camera
CN106895761A (en) * 2017-03-17 2017-06-27 柳州日高滤清器有限责任公司 A kind of plastic air intake manifold cubing
CN108257171A (en) * 2018-01-09 2018-07-06 江苏科技大学 Car radar assembling aperture detection method based on light vision
CN108414358A (en) * 2018-05-23 2018-08-17 中国原子能科学研究院 A kind of device measuring tensile sample elongation after fracture and the contraction percentage of area
CN209342039U (en) * 2018-11-07 2019-09-03 上海第二工业大学 A kind of atomizer micropore pile bore checking system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于机器视觉的微小零件尺寸测量技术研究;刘国阳;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150315;全文 *

Also Published As

Publication number Publication date
CN109470162A (en) 2019-03-15

Similar Documents

Publication Publication Date Title
US20150161776A1 (en) Digital Optical Comparator
CN109470162B (en) Intelligent detection system and method for micropore shape of oil nozzle based on machine vision
CN109877437B (en) Centering detection device and method for resistance welding gun
CN108269255B (en) Electric connector detection method based on machine vision
US20150012243A1 (en) Semiconductor Evaluation Device and Computer Program
CN104834081A (en) Rapid automatic focusing method for stereoscopic microscope
CN106226316A (en) A kind of single camera wide visual field vision thread detecting device and detection method thereof
CN110567965A (en) Smartphone glass cover plate edge visual defect detection method
CN107255903A (en) A kind of method that lens focus is measured based on Newton's ring
CN108257171A (en) Car radar assembling aperture detection method based on light vision
CN105511123B (en) A kind of High Precision Automatic Systems for optical inspection and method based on mechanical arm
CN209342039U (en) A kind of atomizer micropore pile bore checking system
CN109596058A (en) The size detection recognition methods of plastic workpiece
CN108844961A (en) A kind of temperature controller case vision detection system and method
CN107895362B (en) Machine vision method for detecting quality of miniature wiring terminal
Hong et al. Optical method for inspecting surface defects inside a small bore
CN112461846A (en) Workpiece defect detection method and device
CN110838149B (en) Camera light source automatic configuration method and system
CN108645867B (en) Rapid locating and batch detection method for micro defects on surface of large-caliber optical crystal
CN109343303B (en) Micro-drill visual detection method and device based on inner cone mirror scanning panoramic imaging
CN111398290A (en) Coating liquid bead defect identification method suitable for slit coating
CN108705689B (en) Automatic tool setting process method for repairing surface microdefects of large-caliber precise optical crystal
CN109631803A (en) Part angle detects adjusting method
CN110687130A (en) Automatic detection method for inner wall defects of micro-holes
CN109596625A (en) Workpiece, defect detection recognition method in charging tray

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