A method for checking the wiring arrangement of electronic components based on feature recognition

Through similarity calculation of feature recognition technology and standard virtual cable model, automated proofreading of electronic component cable detection is realized, solving the problems of low efficiency and high error rate in the existing technology, and improving detection accuracy and applicability.

CN116664529BActive Publication Date: 2024-12-03SHENZHEN XINLITE CIRCUIT CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310652642.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-05
Publication Date
2024-12-03
Estimated Expiration
2043-06-05

AI Technical Summary

Technical Problem

Existing electronic component wiring inspection mainly relies on manual or circuit path detection, with low efficiency and high error rate. The electrical connection detection equipment cannot be used for wiring inspection of multiple components, resulting in component life loss and detection efficiency. high.

Method used

Using the electronic component line proofreading method based on feature recognition, the electronic component line area image data is obtained and analyzed, and similarity calculation is performed using standard line virtual models to realize automated line path sniffing and proofreading, combining pixels Identification and turning angle analysis to ensure accuracy and consistency of detection results.

Benefits of technology

It improves the efficiency and accuracy of electronic component wiring detection, reduces data processing volume, and avoids component life loss. It is suitable for electronic component wiring detection of various models, ensuring the accuracy and reliability of the detection results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116664529B_ABST
    Figure CN116664529B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of electronic component assembly, and in particular to a method for calibrating the wiring of electronic components based on feature recognition, comprising: acquiring image data of a wiring area on an electronic component, storing the image data of the wiring area on the electronic component, synchronously uploading a standard wiring virtual model of the wiring area on the electronic component, selecting images in the stored image data of the wiring area on the electronic component in turn, and sniffing the wiring path image in the image of the selected image data; the method of the present invention realizes the wiring calibration function in the production process of electronic components by acquiring the image data of electronic components, and in the wiring calibration stage of electronic components, mainly by acquiring the wiring path image of the wiring surface on the electronic component image and the wiring line, and introducing the standard wiring model of the electronic component, the wiring of the electronic component is subjected to multi-level monitoring and calibration.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of electronic component assembly, and in particular to an electronic component wiring calibration method based on feature recognition. Background Art

[0002] Electronic components are basic elements in electronic circuits. They are usually individually packaged and have two or more leads or metal contacts. Electronic components must be connected to each other to form an electronic circuit with specific functions, such as amplifiers, radio receivers, oscillators, etc. One of the common ways to connect electronic components is to solder them to printed circuit boards.

[0003] The invention patent with application number 201910288579.9 discloses a method for electronic component recognition based on texture and shape features, which is characterized by comprising the following steps: image preprocessing (PP): graying and binarizing the color image, and normalizing the target by translating, rotating and scaling the target in the image;

[0004] Gabor filter (GF): Combining the idea of ​​convolutional network, it is applied to the feature extraction of electronic components. The Gabor filter is convolved with the grayscale image to generate a complex signal, and the feature information of the electronic component is initially obtained. The real part (rgg) and the imaginary part (igg) of the signal are separated, and the grayscale co-occurrence matrix of the image is calculated through the imaginary part to further process the feature information: Grayscale co-occurrence matrix (GLCM): Electronic components are rigid structures with simple and symmetrical textures. The features are calculated through the grayscale co-occurrence matrix formula to obtain three feature expressions F1, Fo, and F3. F1 is the energy of the texture, F2 is the texture contrast, and F3 is the texture correlation. It basically expresses the texture information of the rigid body;

[0005] This application aims to solve the problem that current image classification methods are mainly divided into two categories. The first category uses convolutional neural networks (CNN) to automatically learn image features for image classification, and the second category classifies images based on the image space domain or frequency domain. After years of research and accumulation, the first category of methods is relatively ideal in terms of classification accuracy, but its data volume is too large and the calculation complexity is high, which cannot meet the efficiency requirements of timely processing; the second category of methods is not accurate enough and cannot classify multiple components at the same time.

[0006] However, for the existing electronic component wiring inspection and proofreading process, most of them currently adopt manual inspection or circuit path inspection as the quality inspection method. The efficiency of manual inspection is poor and there is a high error rate. The efficiency of circuit path inspection also needs to be improved. At the same time, due to the use of circuit path inspection, the electronic components need to be electrically connected, which will cause a certain life loss to the electronic components, and the electrical connection detection equipment cannot be applied to the wiring inspection and proofreading work of more than one type of electronic components. Summary of the invention

[0007] Technical issues solved

[0008] In view of the above-mentioned shortcomings of the prior art, the present invention provides an electronic component wiring proofreading method based on feature recognition, which solves the problem that the existing electronic component wiring detection and proofreading process currently mostly adopts manual detection or circuit path detection method for quality inspection, and the manual detection efficiency is poor and has a high error rate. The circuit path detection efficiency also needs to be improved. At the same time, due to the use of circuit path detection, the electronic components need to be electrically connected, which will cause a certain life loss to the electronic components, and the electrical connection detection equipment cannot be applied to the wiring detection and proofreading work of more than one electronic component.

[0009] Technical Solution

[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0011] A method for checking the wiring arrangement of electronic components based on feature recognition, comprising:

[0012] Acquire image data of the wiring area on the electronic component, store the image data of the wiring area on the electronic component, synchronously upload the standard wiring virtual model of the wiring area on the electronic component, select images from the stored image data of the wiring area on the electronic component in turn, and sniff the wiring path image in the image of the selected image data;

[0013] Perform pixel recognition on the selected image data, extract pixel blocks from the image data based on the specified color, form a wiring path image through the extracted pixel blocks to complete the sniffing of the wiring path image in the image, pick up points at equal distances at the midline position on the wiring path image, connect adjacent points with each other to form a wiring route, and synchronously perform the wiring path image and wiring route acquisition operations on the standard wiring virtual model;

[0014] The judgment threshold of the wiring of electronic components is set, and the similarity between the image data of the wiring area of ​​the electronic components and the standard wiring virtual model of the wiring area of ​​the electronic components is obtained according to the wiring path image and wiring route of the electronic components. The similarity calculation formula is:

[0015]

[0016] Where: SIM(p,q i ) is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i Image similarity of M p is the number of pixel blocks contained in the wiring path image in the standard wiring virtual model p of the wiring area; Image data of the wiring area on the electronic component q i The number of pixel blocks contained in the wiring path image; μ is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i The total similarity value of the corresponding turning angles on the two lines; γ p The number of picking points on the wiring line corresponding to the standard wiring virtual model p in the wiring area; Image data of the wiring area on the electronic component q i The number of pick-up points on the corresponding routing line; g = x + 3, x is the number of turning corners on the routing line;

[0017] Among them, electronic components whose similarity between the image data of the electronic component wiring area and the standard wiring virtual model of the electronic component wiring area is within the electronic component wiring determination threshold are determined to be qualified, otherwise, they are unqualified.

[0018] Furthermore, after the image data of the wiring area on the electronic component is stored, the similarity of the image data of the wiring area on the electronic component is compared based on the standard wiring virtual model of the wiring area, and a similarity determination threshold is set, and the dissimilar image data of the wiring area on the electronic component is discarded, and the remaining image data of the wiring area on the electronic component is used to perform the selection and wiring path image sniffing operation;

[0019] Among them, the image data of the wiring area on the electronic components is obtained by deploying cameras on the electronic component production equipment, and when each group of image data of the wiring area on the electronic components is obtained, the camera shooting angle, the placement status of the electronic components on the production equipment, and the camera shooting position are all unique. The standard wiring virtual model of the wiring area on the electronic components is drawn by the user in any mechanical drawing software.

[0020] Furthermore, the pixel identification of the image data of the wiring area on the electronic component is obtained by the following formula:

[0021]

[0022] Where: px(a,b) is the pixel of the image data; l a is the horizontal distance of the image data; μ is the area of ​​the smallest group of regions in the image data that only contain one color; l bImage data longitudinal distance;

[0023] After pixel recognition is completed for the image data of the wiring area on the electronic component, the image data is segmented at the pixel level according to the pixel recognition result to obtain pixel blocks of the image data, and pixel block extraction is further completed based on the segmented pixel blocks and the specified colors.

[0024] Furthermore, the points picked up at equal distances from the center line position on the wiring path image all fall on the pixel blocks contained in the wiring path image, the spacing of the picked up points is manually set by the user end in units of pixel blocks, and the distance between the picked up points is determined by the distance of the least pixel block path between the two points;

[0025] Among them, the electronic component production equipment includes an encoding station, and the electronic component production equipment encodes the electronic components produced and output through the encoding station. The codes carried by the electronic components are used to distinguish the corresponding wiring path images and wiring lines.

[0026] Furthermore, the total similarity value of the corresponding turning angles on the wiring lines of the standard wiring virtual model of the wiring area and the wiring area image data of the electronic component is obtained by the following formula:

[0027]

[0028] Where: m is the set of turning angles; h p is the turning angle of the wiring line of the standard wiring virtual model p in the wiring area; is the turning angle of the wiring line of the wiring area image data qi on the electronic component; z is the serial number of the turning angle on the wiring line.

[0029] Furthermore, after the wiring path image and wiring route are obtained for the standard wiring virtual model, a database is simultaneously constructed, and the wiring path image and wiring route of the obtained standard wiring virtual model are sent to the constructed database and stored in the database.

[0030] Furthermore, the database is uploaded to the cloud through network deployment, and the user terminal accesses the cloud through the network to read and download the wiring path image and wiring route data stored in the database.

[0031] Furthermore, after the image data of the wiring area on the electronic component is determined, the electronic component corresponding to the image data of the wiring area on the electronic component is traced according to the distinguishing mark corresponding to the image data of the wiring area on the electronic component.

[0032] Furthermore, the determination results of the electronic components within the determination threshold are transmitted to the control end of the electronic component production equipment through dielectric electrical data, stored in the control end of the electronic component production equipment, the determination result data are read by the user end and the control end of the electronic component production equipment, and the pass rate of the electronic components currently produced by the electronic component production equipment is further calculated based on the determination result data.

[0033] Beneficial Effects

[0034] Compared with the known public technology, the technical solution provided by the present invention has the following advantages:

[0035] Beneficial effects:

[0036] 1. The present invention provides an electronic component wiring calibration method based on feature recognition. The method realizes the wiring calibration function in the electronic component production process by acquiring the electronic component image data. In the electronic component wiring calibration stage, the wiring path image of the wiring surface on the electronic component image and the wiring route are acquired. Under the condition of introducing the electronic component standard wiring model, the wiring of the electronic component is monitored and calibrated at multiple levels.

[0037] 2. During the execution of the steps of the method of the present invention, in the initial monitoring and proofreading stage of the electronic component wiring, absolutely unqualified electronic components are discarded, thereby greatly reducing the amount of data processing during the execution of the method. At the same time, the electronic component image data is analyzed and processed at the pixel level, and the image of the electronic component wiring path in the electronic component image data is obtained, providing necessary data support for the electronic component wiring path and the final output result of the method.

[0038] 3. When the method of the present invention is used to calibrate and detect the wiring of electronic components, the path area image of the wiring of electronic components, the wiring route, the turning angle in the wiring route, and the number of picking points when the wiring route is generated are comprehensively considered and applied to ensure the output accuracy of the calibration and detection results of the wiring of electronic components. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A schematic diagram of the execution process of an electronic component wiring calibration method based on feature recognition;

[0041] Figure 2A schematic diagram of a flow chart of an electronic component wiring calibration method based on feature recognition;

[0042] Figure 3 It is a schematic diagram of the logic of generating the wiring path image and wiring route of the electronic component in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The present invention will be further described below in conjunction with the embodiments.

[0045] Embodiment 1

[0046] The present embodiment provides a method for calibrating the wiring arrangement of electronic components based on feature recognition, such as Figure 1-3 As shown, including:

[0047] Acquire image data of the wiring area on the electronic component, store the image data of the wiring area on the electronic component, synchronously upload the standard wiring virtual model of the wiring area on the electronic component, select images from the stored image data of the wiring area on the electronic component in turn, and sniff the wiring path image in the image of the selected image data;

[0048] Perform pixel recognition on the selected image data, extract pixel blocks from the image data based on the specified color, form a wiring path image through the extracted pixel blocks to complete the sniffing of the wiring path image in the image, pick up points at equal distances at the midline position on the wiring path image, connect adjacent points with each other to form a wiring route, and synchronously perform the wiring path image and wiring route acquisition operations on the standard wiring virtual model;

[0049] The judgment threshold of the wiring of electronic components is set, and the similarity between the image data of the wiring area of ​​the electronic components and the standard wiring virtual model of the wiring area of ​​the electronic components is obtained according to the wiring path image and wiring route of the electronic components. The similarity calculation formula is:

[0050]

[0051] Where: SIM(p,q i ) is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i Image similarity of M pis the number of pixel blocks contained in the wiring path image in the standard wiring virtual model p of the wiring area; Image data of the wiring area on the electronic component q i The number of pixel blocks contained in the wiring path image; μ is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i The total similarity value of the corresponding turning angles on the two lines; γ p The number of picking points on the wiring line corresponding to the standard wiring virtual model p in the wiring area; Image data of the wiring area on the electronic component q i The number of pick-up points on the corresponding routing line; g = x + 3, x is the number of turning corners on the routing line;

[0052] Wherein, electronic components whose similarity between the image data of the wiring area of ​​the electronic component and the standard wiring virtual model of the wiring area of ​​the electronic component is within the electronic component wiring determination threshold are determined to be qualified, otherwise, they are unqualified;

[0053] The pixel recognition of the image data of the wiring area on the electronic component is obtained by the following formula:

[0054]

[0055] Where: px(a,b) is the pixel of the image data; l a is the horizontal distance of the image data; μ is the area of ​​the smallest group of regions in the image data that only contain one color; l b Image data longitudinal distance;

[0056] After the image data of the wiring area on the electronic component is pixel-identified, the image data is segmented at the pixel level according to the pixel identification result to obtain pixel blocks of the image data, and the pixel blocks are further extracted based on the segmented pixel blocks and the specified colors;

[0057] The total similarity value of the corresponding turning angles on the wiring lines of the standard wiring virtual model of the wiring area and the wiring area image data of the electronic component is obtained by the following formula:

[0058]

[0059] Where: m is the set of turning angles; h p is the turning angle of the wiring line of the standard wiring virtual model p in the wiring area; is the turning angle of the wiring line of the wiring area image data qi on the electronic component; z is the serial number of the turning angle on the wiring line.

[0060] In this embodiment, through the above-described proofreading method, based on the electronic component production equipment in the stage of producing electronic components to output, the electronic components are first imaged, and then the wiring image on the electronic components is obtained from the electronic component image, and then the wiring route is derived from the wiring image, so that the operation from the overall to the situation and from the situation to the line is completed, so that the data in this conversion process provides the necessary data support for the multi-level proofreading monitoring of the wiring of electronic components by this method;

[0061] In view of the similarity calculation formula recorded above, the wiring of electronic components can be accurately checked and monitored. In addition, the image pixel recognition calculation formula recorded above analyzes and calculates the image data of the electronic components as a preparatory condition for image segmentation, ensuring that the image data of the electronic components is accurately segmented.

[0062] Example: See Figure 3 As shown, (a) shows the image data of the electronic component, (b) shows the image data of the wiring area of ​​the electronic component, (c) shows the image data of the wiring area and the wiring route generated by the above-mentioned recording scheme, and (d) shows a single wiring route. By using (b) and (d) as references, the wiring calibration and detection is performed on the electronic component corresponding to (a).

[0063] Embodiment 2

[0064] In terms of specific implementation, based on Example 1, this example refers to Figure 1-3 The electronic component wiring calibration method based on feature recognition in Example 1 is further described in detail:

[0065] After the image data of the wiring area on the electronic component is stored, the similarity of the image data of the wiring area on the electronic component is compared based on the standard wiring virtual model of the wiring area, and a similarity judgment threshold is set to discard the image data of the wiring area on the electronic component that are not similar, and the remaining image data of the wiring area on the electronic component is used to perform the selection and wiring path image sniffing operation;

[0066] Among them, the image data of the wiring area on the electronic components is obtained by deploying cameras on the electronic component production equipment, and when each group of image data of the wiring area on the electronic components is obtained, the camera shooting angle, the placement status of the electronic components on the production equipment, and the camera shooting position are all unique. The standard wiring virtual model of the wiring area on the electronic components is drawn by the user in any mechanical drawing software.

[0067] Through the above settings, the electronic component image data collected in the method is screened and processed, and the electronic components corresponding to the discarded electronic component image data recorded above must be unqualified. In addition, through the above record limitations, the problem collection and consistency of the electronic component image data are ensured.

[0068] Preferably, the points picked up at equal distances at the midline position on the wiring path image all fall on the pixel blocks contained in the wiring path image, the spacing of the picked up points is manually set by the user end in units of pixel blocks, and the distance between the picked up points is determined according to the distance of the least pixel block path between the two points;

[0069] The electronic component production equipment includes an encoding station, through which the electronic component production equipment encodes the electronic components produced and output, and the codes carried by the electronic components are used as distinguishing marks of the corresponding wiring path images and wiring lines;

[0070] After the image data of the wiring area on the electronic component is determined, the electronic component corresponding to the image data of the wiring area on the electronic component is traced according to the distinguishing mark corresponding to the image data of the wiring area on the electronic component.

[0071] Through the above settings, necessary data support is provided for the subsequent tracing of unqualified electronic components during the execution of the method.

[0072] Embodiment 3

[0073] In terms of specific implementation, based on Example 1, this example refers to Figure 1-3 The electronic component wiring calibration method based on feature recognition in Example 1 is further described in detail:

[0074] After the operation of obtaining the wiring path image and the wiring route of the standard wiring virtual model is performed, a database is simultaneously constructed, and the obtained wiring path image and wiring route of the standard wiring virtual model are sent to the constructed database and stored in the database;

[0075] The database is uploaded to the cloud through network deployment, and the user terminal accesses the cloud through the network to read and download the wiring path images and wiring route data stored in the database.

[0076] Through the above settings, the wiring path image and wiring route of the standard wiring virtual model used in the execution of the method can be stored for subsequent use by the user end, which also enables the method to be applied to the wiring calibration and detection work of any model of electronic components.

[0077] Preferably, the determination results of the electronic components within the determination threshold are transmitted to the control end of the electronic component production equipment via dielectric electrical data, stored in the control end of the electronic component production equipment, the determination result data are read by the user end and the control end of the electronic component production equipment, and the pass rate of the electronic components currently produced by the electronic component production equipment is further calculated based on the determination result data.

[0078] Through the above settings, the user end can calculate the qualified rate of electronic components produced on the electronic component production equipment according to the result data output by the method, thereby providing the user end with more data references for further inspection and maintenance of the electronic component production equipment.

[0079] In summary, in the above-mentioned embodiment, by acquiring the image data of the electronic component, the wiring calibration function is realized in the production process of the electronic component. In the wiring calibration stage of the electronic component, the wiring path image of the wiring surface on the electronic component image and the wiring route are mainly acquired. Under the condition of introducing the standard wiring model of the electronic component, the wiring of the electronic component is monitored and calibrated at multiple levels; at the same time, in the process of executing its steps, the method discards the absolutely unqualified electronic components in the first monitoring and calibration stage of the wiring of the electronic component, thereby greatly reducing the amount of data processing in the execution process of the method, and at the same time, the electronic component image data is analyzed and processed at the pixel level, and the wiring path image of the electronic component in the electronic component image data is acquired, which provides necessary data support for the wiring path of the electronic component and the final output result of the method; in addition, when the method calibrates and detects the wiring of the electronic component, the path area image of the wiring of the electronic component, the wiring route, the turning angle in the wiring route, and the number of picking points when the wiring route is generated are comprehensively considered and applied, so as to ensure the output accuracy of the wiring calibration detection result of the electronic component.

[0080] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for calibrating the wiring of electronic components based on feature recognition, characterized in that: include: Acquire image data of the wiring area on the electronic component, store the image data of the wiring area on the electronic component, synchronously upload the standard wiring virtual model of the wiring area on the electronic component, select images from the stored image data of the wiring area on the electronic component in turn, and sniff the wiring path image in the image of the selected image data; Perform pixel recognition on the selected image data, extract pixel blocks from the image data based on the specified color, form a wiring path image through the extracted pixel blocks to complete the sniffing of the wiring path image in the image, pick up points at equal distances at the midline position on the wiring path image, connect adjacent points with each other to form a wiring route, and synchronously perform the wiring path image and wiring route acquisition operations on the standard wiring virtual model; The judgment threshold of the wiring of electronic components is set, and the similarity between the image data of the wiring area of ​​the electronic components and the standard wiring virtual model of the wiring area of ​​the electronic components is obtained according to the wiring path image and wiring route of the electronic components. The similarity calculation formula is: Where: SIM(p,q i ) is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i Image similarity of M p is the number of pixel blocks contained in the wiring path image in the standard wiring virtual model p of the wiring area; Image data of the wiring area on the electronic component q i The number of pixel blocks contained in the wiring path image; μ is the standard wiring virtual model p of the wiring area and the image data q of the wiring area on the electronic component i The total similarity value of the corresponding turning angles on the two lines; γ p The number of picking points on the wiring line corresponding to the standard wiring virtual model p in the wiring area; Image data of the wiring area on the electronic component q i The number of pick-up points on the corresponding routing line; g = x + 3, x is the number of turning corners on the routing line; Wherein, electronic components whose similarity between the image data of the wiring area of ​​the electronic component and the standard wiring virtual model of the wiring area of ​​the electronic component is within the electronic component wiring determination threshold are determined to be qualified, otherwise, they are unqualified; The pixel identification of the image data of the wiring area on the electronic component is obtained by the following formula: Where: px(a,b) is the pixel of the image data; l a is the horizontal distance of the image data; μ is the area of ​​the smallest group of regions in the image data that only contain one color; l b Image data longitudinal distance; After the image data of the wiring area on the electronic component is pixel-identified, the image data is segmented at the pixel level according to the pixel identification result to obtain pixel blocks of the image data, and the pixel blocks are further extracted based on the segmented pixel blocks and the specified colors; The points picked up at equal distances from the center line position on the wiring path image all fall on the pixel blocks contained in the wiring path image, the spacing of the picked up points is manually set by the user end in units of pixel blocks, and the distance between the picked up points is determined by the distance of the least pixel block path between the two points; The electronic component production equipment includes an encoding station, through which the electronic component production equipment encodes the electronic components produced and output, and the codes carried by the electronic components are used as distinguishing marks of the corresponding wiring path images and wiring lines; The total similarity value of the corresponding turning angles on the wiring lines of the standard wiring virtual model of the wiring area and the wiring area image data of the electronic component is obtained by the following formula: Where: m is the set of turning angles; h p is the turning angle of the wiring line of the standard wiring virtual model p in the wiring area; is the turning angle of the wiring line of the wiring area image data qi on the electronic component; z is the serial number of the turning angle on the wiring line.

2. The electronic component wiring calibration method based on feature recognition according to claim 1, characterized in that: After the image data of the wiring area on the electronic component is stored, the similarity of the image data of the wiring area on the electronic component is compared based on the standard wiring virtual model of the wiring area, and a similarity judgment threshold is set to discard the image data of the wiring area on the electronic component that are not similar, and the remaining image data of the wiring area on the electronic component is used to perform the selection and wiring path image sniffing operation; Among them, the image data of the wiring area on the electronic components is obtained by deploying cameras on the electronic component production equipment, and when each group of image data of the wiring area on the electronic components is obtained, the camera shooting angle, the placement status of the electronic components on the production equipment, and the camera shooting position are all unique. The standard wiring virtual model of the wiring area on the electronic components is drawn by the user in any mechanical drawing software.

3. The electronic component wiring calibration method based on feature recognition according to claim 1, characterized in that: After the wiring path image and wiring route are obtained for the standard wiring virtual model, a database is simultaneously constructed, and the wiring path image and wiring route of the obtained standard wiring virtual model are sent to the constructed database and stored in the database.

4. The electronic component wiring calibration method based on feature recognition according to claim 1, characterized in that: The database is uploaded to the cloud through network deployment, and the user terminal accesses the cloud through the network to read and download the wiring path image and wiring route data stored in the database.

5. The electronic component wiring calibration method based on feature recognition according to claim 1, characterized in that: After the image data of the wiring area on the electronic component is determined, the electronic component corresponding to the image data of the wiring area on the electronic component is traced according to the distinguishing mark corresponding to the image data of the wiring area on the electronic component.

6. The electronic component wiring calibration method based on feature recognition according to claim 1, characterized in that: The determination results of the electronic components within the determination threshold are transmitted to the control end of the electronic component production equipment through dielectric electrical data, stored in the control end of the electronic component production equipment, and the determination result data are read by the user end and the control end of the electronic component production equipment, and the qualified rate of the electronic components currently produced by the electronic component production equipment is further calculated based on the determination result data.

Citation Information

Patent Citations

  • Electronic element identification method based on texture and shape features

    CN110147570A

  • Method and device for detecting optical fiber winding displacement defects

    CN111413350A