Optical detection equipment coordinate keeping method based on image recognition calibration

By using image recognition calibration plates and optical cameras in AOI devices, the coordinates of the equipment are quickly restored, which solves the problems of high cost and complex maintenance in traditional methods, and efficient and stable coordinate maintenance is achieved, improving the reliability of the equipment and the continuity of the detection process.

CN120293879APending Publication Date: 2025-07-11GUANGDONG HONHOR SEMICON EQUIP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510539376.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-27
Publication Date
2025-07-11

AI Technical Summary

Technical Problem

It is difficult for existing AOI equipment to quickly recover coordinates after power outage or failure, resulting in interruption of detection processes, and traditional coordinate maintenance methods are costly, complex in maintenance, and insufficient reliability.

Method used

The calibration plate and optical camera based on image recognition are used to obtain the calibration plate image by taking pictures, build auxiliary center lines, identify punctuation coordinate characters, calculate the calculation coordinate information of the center point, and compare it with the breakpoint coordinate information to achieve rapid recovery and verification of coordinates.

Benefits of technology

It realizes low-cost and efficient coordinate recovery, reduces hardware costs and maintenance complexity, and improves the reliability of equipment and the continuity of inspection processes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120293879A_ABST
    Figure CN120293879A_ABST
Patent Text Reader

Abstract

The invention discloses an optical detection equipment coordinate keeping method based on image recognition calibration, which comprises a multi-axis AOI module with an optical camera and a calibration plate, a plurality of punctuations distributed in a grid array are arranged on the surface of the calibration plate, and each punctuation has a unique coordinate character; the power-off detection or reset initialization detection of the multi-axis AOI module is controlled through the steps of breakpoint coordinate recording, calibration plate positioning, image acquisition and auxiliary line construction, punctuation identification and coordinate acquisition, punctuation identification and coordinate acquisition, calculated coordinate calculation and coordinate consistency judgment in the steps S1-S6, and the method has the advantages of being good in stability, low in cost and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of optical detection equipment, and specifically to a coordinate holding method for an optical detection equipment based on image recognition calibration. Background Art

[0002] Automatic Optical Inspection (AOI) equipment is widely used in the field of electronic manufacturing. Its core function is to drive an optical camera through a multi-axis motion module (X / Y / Z axes) to perform high-precision positioning and inspection on workpieces. During continuous operation, if the equipment loses coordinates due to power failure, malfunction, or manual interruption, it will seriously affect the continuity of the inspection process, and even cause workpiece damage or data loss. Therefore, how to quickly restore the module coordinates after interruption has become a key technical issue for improving the reliability and efficiency of AOI equipment.

[0003] Traditional AOI equipment achieves coordinate holding in the following ways: 1) Mechanical absolute encoders / grating scales: Record the absolute position through physical scales and sensors. However, such devices are costly (accounting for 15% - 30% of the total equipment cost), and their accuracy is likely to decline due to mechanical wear during long-term use.

[0004] 2) Battery-backed encoders: Rely on built-in batteries to maintain position data. However, the battery life is limited (usually 2 - 5 years), and coordinates will still be lost when the voltage is too low, requiring frequent maintenance and replacement.

[0005] 3) Software coordinate memory method: Store the coordinates before power failure through the host computer. However, it is unable to verify the physical consistency of the restored coordinates, and there is a risk of cumulative errors caused by mechanical backlash or load disturbances.

[0006] The above methods generally have problems such as high cost, complex maintenance, and insufficient reliability, and are difficult to meet the requirements of industrial scenarios for low cost and high stability. Summary of the Invention

[0007] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an efficient and stable coordinate holding method for an optical detection equipment based on image recognition calibration.

[0008] To achieve the above purpose, a coordinate holding method for an optical detection equipment based on image recognition calibration provided by the present invention includes a multi-axis AOI module with an optical camera and a calibration plate. The calibration plate surface is provided with a number of punctuation marks distributed in a grid array, and each punctuation mark has a unique coordinate character. This coordinate holding method includes the following steps: S1. When the multi-axis AOI module loses power and interrupts, obtain and record the breakpoint coordinate information of the optical camera at this moment; S2. When the multi-axis AOI module is powered on, place the calibration plate at the rated position directly below the field of view of the optical camera; S3. The optical camera obtains an image with the local position of the calibration plate through photographing, and constructs a horizontal auxiliary median line and a vertical auxiliary median line perpendicular to each other on the image. Among them, the intersection point between the horizontal auxiliary median line and the vertical auxiliary median line is used as the center point A; S4. Based on the image, determine the four punctuation marks closest to the periphery of the center point A, and obtain the coordinate characters of the four punctuation marks through optical character recognition; S5. Based on the coordinate characters of the four punctuation marks and the pixel occupancy of the center point A within the pixel area enclosed by the four punctuation marks, calculate the calculated coordinate information of the center point A; S6. Compare whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information. Among them, if the information is consistent, the multi-axis AOI module continues to execute the detection process before the interruption; otherwise, if the information is inconsistent, the multi-axis AOI module resets and initializes and starts to execute the detection process again.

[0009] Further, in step S5, the calculation of the calculated coordinate information of the center point A specifically includes: determining its absolute coordinate value based on the coordinate characters of the four punctuation marks, and through the horizontal pixel occupancy and vertical pixel occupancy of the center point A within the pixel area enclosed by the four punctuation marks, add the absolute coordinate values of the closest punctuation marks at equal interval ratios to obtain the calculated coordinate information of the center point A.

[0010] Further, in step S6, compare whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information or within the error threshold. Among them, by calculating the difference between the calculated coordinate information of the center point A and the breakpoint coordinate information, if the difference exceeds the preset error threshold, the multi-axis AOI module resets and starts to execute the detection process again; otherwise, if the difference does not exceed the preset error threshold, the multi-axis AOI module continues to execute the detection process after the interruption.

[0011] Further, the punctuation marks of the calibration plate are arranged at equal intervals according to the rated length, and the coordinate characters of each punctuation mark strictly correspond to its actual physical position.

[0012] The present invention adopts the above-mentioned scheme, and its beneficial effects are as follows: 1) Low cost: Using a reusable high-precision calibration plate to replace the mechanical absolute encoder, the hardware cost is reduced to 5%-10% of the traditional scheme. The calibration plate is mass-produced based on alumina substrate and laser etching process, and the single-piece cost is less than 1 / 20 of the absolute grating scale, and it does not require battery power supply or mechanical maintenance throughout the life cycle, and the comprehensive operation and maintenance cost is reduced by more than 83%; 2) High efficiency: By constructing a trinity coordinate guarantee system of "visual calibration - data persistence - dynamic verification", efficient determination and breakpoint detection can be realized. Description of the Drawings

[0013] Figure 1 It is a schematic diagram of a multi-axis AOI module.

[0014] Figure 2 It is a schematic diagram of the use of a multi-axis AOI module and a calibration board.

[0015] Figure 3 It is a schematic diagram of the calibration board.

[0016] Figure 4 It is a schematic diagram of image acquisition and auxiliary line construction in step S3.

[0017] Figure 5 It is a flowchart of the coordinate holding method of an optical detection device.

[0018] Figure 6 It is a schematic diagram of the manufacturing of an optical detection control system.

[0019] Among them, 100 - base, 101 - Y-axis drive module, 102 - workpiece to be detected, 103 - optical camera, 104 - Z-axis drive module, 105 - X-axis drive module, 106 - positioning slot, 107 - accommodation slot, 110 - calibration board, 111 - punctuation mark. Specific embodiments

[0020] For the convenience of understanding the present invention, the present invention will be described more comprehensively below with reference to the accompanying drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0021] See the attached Figures 1-6 As shown, in this embodiment, a coordinate holding method for an optical detection device based on image recognition calibration includes a multi-axis AOI module with an optical camera 103 and a calibration board 110. Among them, see the attached Figure 1As shown in FIG. 2, the multi-axis AOI module includes a base 100, an X-axis drive module 105, a Y-axis drive module 101, a Z-axis drive module 104, and an optical camera 103. The Y-axis drive module 101 is mounted on the base 100, the X-axis drive module 105 is mounted on the Y-axis drive module 101, the Z-axis drive module 104 is mounted on the X-axis drive module 105, and the optical camera 103 is mounted on the Z-axis drive module 104. Further, the X-axis drive module 105, the Y-axis drive module 101, and the Z-axis drive module 104 can adopt a drive assembly solution with a track base, a motor, and a lead screw in linkage. Specifically, the track bases of the Y-axis drive module 101 are fixedly mounted on both sides of the base 100. The track base of the X-axis drive module 105 spans above the base 100, and both ends of the track base of the X-axis drive module 105 are slidably connected to the track bases of the Y-axis drive module 101. Thus, driven by the motor and the lead screw of the Y-axis drive module 101, the track base of the X-axis drive module 105 translates along the track base of the Y-axis drive module 101 accordingly. The track base of the Z-axis drive module 104 is arranged vertically and is slidably connected to the track base of the X-axis drive module 105. Thus, driven by the motor and the lead screw of the X-axis drive module 105, the track base of the Z-axis drive module 104 translates along the track base of the X-axis drive module 105 accordingly. The optical camera 103 is slidably connected to the track base of the Z-axis drive module 104. Driven by the motor and the lead screw of the Z-axis drive module 104, the optical camera 103 moves up and down along the track base of the Z-axis drive module 104 accordingly. In summary, through the driving actions of the X-axis drive module 105, the Y-axis drive module 101, and the Z-axis drive module 104, the optical camera 103 is driven to perform multi-axis movement in the X-axis, Y-axis, and Z-axis directions.

[0022] See the appendix Figures 1-2As shown, in this embodiment, the top surface of the base 100 serves as a platform for placing the workpiece 102 to be detected, and the optical camera 103 is correspondingly located above the top surface of the base 100. Secondly, a receiving slot 107 for storing the calibration plate 110 is provided at the lower position of the base 100. Support portions are formed by protruding on both sides of the top surface of the base 100 (the support portions are used as the track bases for the Y-axis driving module 101 to be fixedly installed), and positioning slots 106 that are parallel to each other and into which the calibration plate 110 is slidably inserted are provided on the opposite surfaces of the two support portions. Specifically, when the calibration plate 110 needs to be used, after the calibration plate 110 is taken out from the receiving slot 107, the calibration plate 110 is slid into the positioning slot 106 until it is completely in place (which means that the calibration plate 110 reaches the predetermined position. Positioning structures such as relevant positioning points and positioning blocks can be added inside the positioning slot 106 to further ensure the accuracy and precision of each sliding insertion of the calibration plate 110. Specific limitations are not made here, and those skilled in the art can select appropriate structural designs as needed). At this time, the calibration plate 110 is located directly above the top surface of the base 100. Conversely, when the calibration plate 110 does not need to be used, the calibration plate 110 is taken out from the positioning slot 106 and then slid into the receiving slot 107 to complete the storage.

[0023] See the appendix Figure 3 As shown, in this embodiment, a plurality of punctuation marks 111 distributed in a grid array are provided on the surface of the calibration plate 110, and each punctuation mark 111 has a unique coordinate character. Specifically, The calibration plate 110 is made of alumina material, has no reflection, high precision and is maintenance-free, and has high stability. Punctuation marks 111 are etched on the surface of the calibration plate 110 in an array arranged at equal intervals of 10 mm. Each punctuation mark 111 contains a unique coordinate character (for example Figure 4 the coordinate character (20, 30) corresponding to the punctuation mark 111Bc in the figure). The punctuation marks 111 of the calibration plate 110 are arranged at equal intervals according to the rated length, and the coordinate characters of each punctuation mark 111 strictly correspond to its actual physical position. In this embodiment, see the appendix Figure 5 As shown, the coordinate holding method includes the following steps: Step S1. Breakpoint coordinate recording: When the multi-axis AOI module loses power and is interrupted (the interruption can occur due to reasons such as power failure, shutdown and suspension, and fault interruption), obtain and record the breakpoint coordinate information of the optical camera 103 at this moment. Among them, the power-off coordinate information here is the coordinate of the current field of view center position of the optical camera 103 at the interruption moment. For example, the breakpoint coordinate information at this moment is (24, 23); Step S2. Calibration plate 110 positioning: When the multi-axis AOI module is powered on, place the calibration plate 110 at the rated position directly below the field of view of the optical camera 103; Step S3. Image acquisition and auxiliary line construction: The optical camera 103 takes a picture to obtain an image with the local position of the calibration board 110, and constructs a horizontal auxiliary median line (along the X-axis direction) and a vertical auxiliary median line (along the Y-axis direction) that are perpendicular to each other on the image. The intersection point between the horizontal auxiliary median line and the vertical auxiliary median line is used as the center point A; Step S4. Punctuation 111 recognition and coordinate acquisition: Based on the image, determine the four punctuation marks 111 closest to the center point A, and obtain the coordinate characters of the four punctuation marks 111 through optical character recognition. For example: Refer to the appendix Figure 4 As shown, according to the image obtained by taking pictures in step S3, the four punctuation marks 111 closest to the center point A are Bc, Cc, Bd, and Cd, and the coordinate characters corresponding to the four punctuation marks 111 are obtained through optical character recognition (OCR image recognition) as (20, 30), (30, 30), (20, 40), and (30, 40).

[0024] Step S5. Calculation of the calculated coordinates: Based on the coordinate characters of the four punctuation marks 111 and the pixel occupancy of the center point A in the pixel area enclosed by the four punctuation marks 111, the calculated coordinate information of the center point A is calculated; specifically, in step S5, the calculation of the calculated coordinate information of the center point A specifically includes: determining the absolute coordinate values based on the coordinate characters of the four punctuation marks 111, and through the horizontal pixel occupancy and vertical pixel occupancy of the center point A in the pixel area enclosed by the four punctuation marks 111, adding the absolute coordinate values of the closest punctuation mark 111 at equal interval ratios to obtain the calculated coordinate information of the center point A. For example: The pixel area enclosed by the four punctuation marks 111 is 800 * 800 pixels. By calculating the horizontal pixel occupancy of the center point A in the X-axis direction (as Figure 4 shown, the pixel distance between the center point A and the left punctuation marks 111 Bc and Bd is 320, and the horizontal pixel occupancy is 320 / 800 = 40%), and the vertical pixel occupancy in the Y-axis direction (as Figure 4 shown, the pixel distance between the center point A and the upper coordinate points 111 Bc and Bd is 240, and the vertical pixel occupancy is 240 / 800 = 30%). Therefore, the calculation of the calculated coordinate information is: The X-axis coordinate of the center point A = 20 (the X-axis coordinate of the left punctuation mark 111) + 10 (the difference between the left punctuation mark 111 and the right punctuation mark 111) * 40% (horizontal pixel occupancy) = 24; The Y-axis coordinate of the center point A = 30 (the X-axis coordinate of the left punctuation mark 111) + 10 (the difference between the left punctuation mark 111 and the right punctuation mark 111) * 30% (horizontal pixel occupancy) = 33. From this, it can be obtained that the calculated coordinate information of the center point A is (24, 23).

[0025] Step S6. Coordinate consistency judgment: Compare whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information. Specifically, if the information is consistent, the multi-axis AOI module will continue to execute the detection process before the interruption; otherwise, if the information is inconsistent, the multi-axis AOI module will be reset and initialized and start the detection process again. Specifically, the numerical comparison of the X-axis direction coordinate and the Y-axis direction coordinate is respectively performed according to the breakpoint coordinate information determined in step S1 and the calculated coordinate information of the center point A determined in step S5, so as to determine the consistency. For example, if the breakpoint coordinate information is (24, 23) and the calculated coordinate information is (24, 23), it is determined as consistent at this time, and the multi-axis AOI module continues to execute the detection process before the interruption; if the breakpoint coordinate information is (23, 25) and the calculated coordinate information is (24, 23), it is determined as inconsistent at this time, and the multi-axis AOI module performs reset initialization and starts the detection process again.

[0026] In summary, through the above steps S1 - S6, when the multi-axis AOI module recovers after a power failure, it can complete the coordinate calibration judgment in a short time, avoid the detection failure caused by coordinate offset, improve the reliability of continuous operation of the equipment, and further enable the multi-axis AOI module to stop arbitrarily during the detection process, and continue the breakpoint detection or reset and re-detection according to the actual situation. Secondly, the method of using the external calibration plate 110 in cooperation with character recognition has a lower cost and is easier to maintain compared with the traditional absolute encoder or absolute grating scale. In addition, the form of using the calibration plate 110 avoids the traditional battery maintenance method, thus solving the problem of position loss caused by low voltage in the traditional absolute encoder using the battery form.

[0027] In this embodiment, considering the actual position error and image recognition error, an error threshold judgment mechanism is introduced in step S6. Specifically, it is compared whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information or within the error threshold. Specifically, by calculating the difference between the calculated coordinate information of the center point A and the breakpoint coordinate information, if the difference exceeds the preset error threshold, the multi-axis AOI module will be reset and start the detection process again; otherwise, if the difference does not exceed the preset error threshold, the multi-axis AOI module will continue to execute the detection process after the interruption. For example: The error threshold is set to 10% of the distance between adjacent punctuation marks 111 (i.e., ±1 mm). If the difference is within the threshold, the multi-axis AOI module continues to execute the original detection process; if it exceeds the tolerance, the reset mechanism is triggered, and the multi-axis AOI module returns to the origin and re-initializes the calibration.

[0028] See the appendix Figure 5As shown, in this embodiment, the optical detection control system specifically includes a PLC controller and a host computer for image optical character recognition processing, wherein the PLC controller includes a processor, a power-off register, a comparator, and a position controller. Specifically, the input end of the host computer is electrically connected to the optical camera 103, so that the image with the local position of the calibration plate 110 obtained by the optical camera 103 is transmitted to the host computer, and the image is subjected to image optical character recognition processing through the relevant OCR controls preset in the host computer, thereby obtaining the coordinate characters on the image. The output end of the host computer is electrically connected to the processor, and is used to transmit the recognized coordinate characters to the processor for data processing, so as to obtain the required calculation coordinate information of the center point A after being processed by the processor. The processor is electrically connected to the power-off register, the comparator and the position controller respectively; the input end of the power-off register is electrically connected to the position controller and the output end of the power-off register is electrically connected to the comparator; the position controller is electrically connected to the multi-axis AOI module, and the position controller is used to issue position instructions to each drive module of the multi-axis AOI module and obtain the position information of the optical camera 103 in real time, so that when the multi-axis AOI module loses power, the position controller can transmit the current position information of the optical camera 103 to the power-off register as breakpoint coordinate information for storage; the comparator is used to compare the calculated coordinate information processed by the processor and the breakpoint coordinate information stored in the power-off register, and after the comparator transfers the compared information as data to the processor for processing, the position controller issues corresponding position instructions to each drive module of the multi-axis AOI module.

[0029] In this embodiment, refer to the flowchart of the optical detection control system shown in the accompanying drawings: When the multi-axis AOI module is in normal operation, the position controller of the PLC controller sends position commands to each driving module of the multi-axis AOI module, thereby driving the optical camera 103 to perform high-precision movement to realize the visual inspection process; When the multi-axis AOI module loses power, the current position of the optical camera 103 is transferred to the power-off register of the PLC controller as the breakpoint coordinate information via the position controller; When the multi-axis AOI module is powered on, first, the optical camera 103 takes a photo to obtain an image with the local position of the calibration plate 110 and transfers it to the host computer to construct the horizontal auxiliary center line and the vertical auxiliary center line and to obtain the coordinate characters of the adjacent punctuation points 111 through optical character recognition. Then, the obtained coordinate characters are processed by the processor of the PLC controller to obtain the calculated coordinate information which is transferred to the comparator, so that the calculated coordinate information is compared with the breakpoint coordinate information from the power-off register in the comparator. Finally, the comparison result is transferred to the position controller of the PLC controller, so that the position controller of the PLC controller issues a breakpoint detection or reset detection command.

[0030] The embodiments described above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Any person skilled in the art, without departing from the scope of the technical solution of the present invention, makes more possible changes, modifications or retouches to the technical solution of the present invention by using the technical content disclosed above, or such modifications are equivalent embodiments of the present invention. Therefore, all equivalent changes made according to the idea of the present invention without departing from the content of the technical solution of the present invention shall be covered by the protection scope of the present invention.

Claims

1. A method for maintaining the coordinates of an optical detection device based on image recognition calibration, characterized in that: It includes a multi-axis AOI module with an optical camera (103) and a calibration board (110). Among them, several punctuation marks (111) distributed in a grid array are provided on the board surface of the calibration board (110), and each punctuation mark (111) has a unique coordinate character; This coordinate maintaining method includes the following steps: S1. When the multi-axis AOI module loses power and interrupts, obtain and record the breakpoint coordinate information of the optical camera (103) at this moment; S2. When the multi-axis AOI module is powered on, place the calibration board (110) at the rated position directly below the field of view of the optical camera (103); S3. The optical camera (103) obtains an image with a partial position of the calibration board (110) by taking a picture, and constructs a horizontal auxiliary median line and a vertical auxiliary median line perpendicular to each other on the image. Among them, the intersection point between the horizontal auxiliary median line and the vertical auxiliary median line is used as the center point A; S4. Based on the image, determine the four nearest punctuation marks (111) around the center point A, and obtain the coordinate characters of the four punctuation marks (111) through optical character recognition; S5. Based on the coordinate characters of the four punctuation marks (111) and the pixel proportion of the center point A in the pixel area enclosed by the four punctuation marks (111), calculate the calculated coordinate information of the center point A; S6. Compare whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information. Among them, if the information is consistent, the multi-axis AOI module continues to execute the detection process before the interruption; otherwise, if the information is inconsistent, the multi-axis AOI module resets and initializes and starts to execute the detection process again.

2. A coordinate maintaining method for an optical detection device based on image recognition calibration according to claim 1, characterized in that: In step S5, the calculation of the calculated coordinate information of the center point A specifically includes: determining its absolute coordinate value based on the coordinate characters of the four punctuation marks (111), and according to the horizontal pixel proportion and the vertical pixel proportion of the center point A in the pixel area enclosed by the four punctuation marks (111), superimpose the absolute coordinate values of the nearest punctuation marks (111) at equal interval ratios to obtain the calculated coordinate information of the center point A.

3. A method for maintaining coordinates of an optical detection device based on image recognition calibration according to claim 1, characterized in that: In step S6, compare whether the calculated coordinate information of the center point A is consistent with the breakpoint coordinate information or within the error threshold. Among them, by calculating the difference between the calculated coordinate information of the center point A and the breakpoint coordinate information, if the difference exceeds the preset error threshold, the multi-axis AOI module resets and starts to execute the detection process again; otherwise, if the difference does not exceed the preset error threshold, the multi-axis AOI module continues to execute the detection process after the interruption.

4. A method for maintaining the coordinates of an optical detection device based on image recognition calibration according to claim 1, characterized in that: The punctuation marks (111) of the calibration board (110) are arranged at equal intervals according to the rated length, and the coordinate character of each punctuation mark (111) strictly corresponds to its actual physical position.