Intelligent Industrial Control Method, Device, Electronic Device and Storage Medium Based on 5G Technology

Abstract

The present invention relates to intelligent industrial control technology for circuit boards, and discloses an intelligent industrial control method based on 5G technology, including: receiving the circuit image and identity information of the circuit board to be detected, identifying the components and circuit traces in the circuit image, calculating the position coordinate data of each component in the circuit image and calculating the line coordinate data of the circuit traces in the circuit image, generating a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data, obtaining a preset standard simulated circuit diagram corresponding to the identity information of the circuit board to be detected, performing a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram, and when the comparison is inconsistent, sending a pause instruction and an abnormal warning instruction by using a 5G communication module. The present invention also proposes an intelligent industrial control device, an electronic device, and a medium based on 5G technology. The present invention can improve the efficiency of circuit board mounting detection and reduce the cost of circuit board mounting detection.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent industrial control of circuit boards, and particularly to an intelligent industrial control method, device, electronic device and computer-readable storage medium based on 5G technology. Background Art

[0002] During the production process of integrated circuit boards in the production line, there are certain quality problems such as component mounting offset, less welding, and missed welding. In order to ensure the mounting quality of electronic components on the circuit board, it is necessary to detect the mounting situation of the circuit board.

[0003] Traditional circuit board mounting detection mainly relies on manual inspection. However, there are usually many solder joints and small parts on the circuit board. In this complex background, manual detection has a large labor intensity, low detection efficiency, and is easily affected by personal subjective factors, which greatly limits the improvement of enterprise production efficiency and product quality.

[0004] The currently more effective method for circuit board mounting detection is to use X-ray detection technology, that is, after the X-ray penetrates the components mounted on the circuit board to be detected, it is received by an image intensifier. The image intensifier converts the invisible X-ray detection signal into an optical image, and then uses a high-definition camera to capture the optical image, input it into a computer for A / D conversion into a digital image, and then the computer performs digital processing and mounting quality analysis on the image.

[0005] The above detection method requires the installation of detection tooling equipment such as X-ray machines, image intensifiers, and optical lenses, resulting in a relatively high detection cost. Summary of the Invention

[0006] The present invention provides an intelligent industrial control method, device and computer-readable storage medium based on 5G technology, and its main purpose is to improve the efficiency of circuit board mounting detection and reduce the cost of circuit board mounting detection.

[0007] To achieve the above object, an intelligent industrial control method based on 5G technology provided by the present invention includes:

[0008] Receiving the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module;

[0009] Using a pre-trained circuit image recognition model to identify the components and circuit traces in the circuit image;

[0010] Calculating the position coordinate data of each component in the circuit image and calculating the line coordinate data of the circuit traces in the circuit image;

[0011] Generate a simulated circuit diagram of the circuit board to be detected based on the position coordinate data and the circuit line coordinate data;

[0012] Obtain a preset standard simulated circuit diagram corresponding to the identity - associated data of the circuit board to be detected;

[0013] Perform a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, send a pause instruction to a preset production line control terminal through a 5G communication module and send an abnormal warning instruction to the automatic image acquisition terminal.

[0014] Optionally, the using the pre - trained circuit image recognition model to identify the components and circuit traces in the circuit image includes:

[0015] Perform a vector conversion operation on the circuit image to obtain a vector matrix corresponding to the circuit image;

[0016] Use the pre - trained circuit image recognition model to extract the component features and circuit trace features of the vector matrix;

[0017] Use a pre - trained activation function to match a preset component label corresponding to each component feature, and label the corresponding component according to the matched component label;

[0018] Mark the circuit trace pixel points in the circuit image according to the circuit trace features, and connect the circuit trace pixel points to obtain the circuit trace.

[0019] Optionally, the calculating the position coordinate data of each component in the circuit image includes:

[0020] Identify the reference point area corresponding to the preset reference point in the circuit board to be detected in the circuit image;

[0021] Obtain the center of each reference point area, and generate the coordinate origin of the circuit image according to the spatial correspondence relationship between the centers of each reference point area;

[0022] Perform edge detection processing on the labeled components to obtain a graphic border corresponding to each component;

[0023] Randomly select a preset number of pixel points from each graphic border as measurement points in turn, calculate the spatial distance between each measurement point and the coordinate origin, and obtain the coordinate value of each measurement point;

[0024] Collect the coordinate values of each measurement point to obtain the position coordinate data of the corresponding component.

[0025] Optionally, calculating the line coordinate data of the circuit traces in the circuit image includes:

[0026] Segmenting the circuit trace according to the shape of the circuit trace;

[0027] Sequentially calculating the starting coordinate value of the starting point of each segment relative to the coordinate origin and the ending coordinate value of the ending point of each segment relative to the coordinate origin;

[0028] Calculating the azimuth angle of the corresponding segment according to the starting coordinate value and the ending coordinate value;

[0029] When the segment is in the shape of a circular curve, calculating the radius of the circular curve of the corresponding segment;

[0030] Collecting the starting coordinate value, ending coordinate value, azimuth angle and circular curve radius of each segment to obtain the line coordinate data.

[0031] Optionally, obtaining the center of the circle of each reference point area includes:

[0032] Separating the reference point area from the background of the circuit image;

[0033] Calculating the edge points of the separated reference point area to obtain the image contour corresponding to the reference point area;

[0034] Performing ellipse fitting on the image contour and generating a circumscribed rectangle corresponding to the image contour after ellipse fitting;

[0035] Calculating the ratio of the difference between the length and width of the circumscribed rectangle to the width of the circumscribed rectangle;

[0036] When the ratio is less than a preset ratio threshold, taking the center point of the image contour after fitting as the center of the corresponding reference area.

[0037] Optionally, comparing the consistency between the simulated circuit diagram and the preset standard simulated circuit diagram includes:

[0038] Respectively extracting the corner point features of the simulated circuit diagram and the corner point features of the preset standard simulated circuit;

[0039] Performing binary encoding on the corner point features of the simulated circuit diagram to obtain a comparison corner point feature vector, and performing binary encoding on the corner point features of the preset standard simulated circuit to obtain a reference corner point feature vector;

[0040] Calculating the distance between the comparison corner point feature vector and the reference corner point feature vector;

[0041] When the distance is less than or equal to a preset distance threshold, output result information indicating consistent comparison;

[0042] When the distance is not less than the preset distance threshold, output result information indicating inconsistent comparison.

[0043] Optionally, before using the pre-trained circuit image recognition model to recognize components and circuit traces in the circuit image, the method further includes:

[0044] Performing pixel correction on the circuit image using a flat-field correction algorithm;

[0045] Performing binarization processing on the image after pixel correction to obtain a binarized image;

[0046] Performing hole area filling processing on the binarized image using a closing operation algorithm; and

[0047] Performing edge smoothing processing on the binarized image using an opening operation algorithm.

[0048] To solve the above problems, the present invention further provides an intelligent industrial control device based on 5G technology, the device includes:

[0049] An image and identity information receiving module, configured to receive the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by an automatic image acquisition terminal through a 5G communication module;

[0050] An image recognition module, configured to use a pre-trained circuit image recognition model to recognize components and circuit traces in the circuit image;

[0051] A simulated circuit diagram generation module, configured to calculate the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image, and generate a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data;

[0052] A standard circuit diagram acquisition module, configured to acquire a preset standard simulated circuit diagram corresponding to the identity association data of the circuit board to be detected;

[0053] An industrial control module, configured to perform consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result, and when the consistency comparison result is abnormal, send a pause instruction to a preset production line control terminal through a 5G communication module and send an abnormal warning instruction to the automatic image acquisition terminal.

[0054] To solve the above problems, the present invention further provides an electronic device, the electronic device includes:

[0055] A memory that stores at least one instruction; and

[0056] A processor that executes the instructions stored in the memory to implement the intelligent industrial control method based on 5G technology described above.

[0057] To solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the intelligent industrial control method based on 5G technology described above.

[0058] In an embodiment of the present invention, a circuit image of a circuit board to be detected is collected. By identifying the components and circuit traces in the circuit image, the separation of the target to be detected from the pad background is achieved, and the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image are calculated. Further, according to the position coordinate data and the line coordinate data, a simulated circuit diagram of the circuit board to be detected is generated, and the actual mounting result of the circuit board to be detected is mapped into the simulated circuit diagram. Finally, through the graphic consistency comparison between the simulated circuit diagram and the standard simulated circuit diagram, the detection of the mounting quality of the circuit board is realized. Therefore, the intelligent industrial control method based on 5G technology proposed by the present invention improves the detection efficiency of circuit board mounting and reduces the cost compared with manual detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic structural diagram of an intelligent industrial control system based on 5G technology provided by an embodiment of the present invention;

[0060] Figure 2 It is a schematic flowchart of an intelligent industrial control method based on 5G technology provided by an embodiment of the present invention;

[0061] Figure 3 It is a detailed implementation flowchart of one of the steps in the intelligent industrial control method based on 5G technology provided by an embodiment of the present invention;

[0062] Figure 4 It is a detailed implementation flowchart of another one of the steps in the intelligent industrial control method based on 5G technology provided by an embodiment of the present invention;

[0063] Figure 5 It is a detailed implementation flowchart of another one of the steps in the intelligent industrial control method based on 5G technology provided by an embodiment of the present invention;

[0064] Figure 6 It is a detailed implementation flowchart of another one of the steps in the intelligent industrial control method based on 5G technology provided by an embodiment of the present invention;

[0065] Figure 7 Function module diagram of an intelligent industrial control device based on 5G technology provided by an embodiment of the present invention;

[0066] Figure 8 Schematic structural diagram of an electronic device for implementing the intelligent industrial control method based on 5G technology provided by an embodiment of the present invention.

[0067] The realization, functional characteristics and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] An embodiment of the present application provides an intelligent industrial control method based on 5G technology. The execution subject of the intelligent industrial control method based on 5G technology includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the intelligent industrial control method based on 5G technology can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0070] Referring to Figure 1 As shown, it is a schematic structural diagram of an intelligent industrial control system based on 5G technology provided by an embodiment of the present invention. In this embodiment, the intelligent industrial control system 00 based on 5G technology includes: an automatic image acquisition terminal 001, a server terminal 000, and a production line control terminal 002. Among them, the automatic image acquisition terminal 001 is mainly used to acquire the circuit image and identity association data of the circuit board to be detected. The server terminal 000 is mainly used to identify the circuit image of the circuit board to be detected, generate a simulated circuit diagram of the circuit board to be detected, and perform a consistency comparison between the simulated circuit diagram and the corresponding standard simulated circuit diagram, and issue corresponding control instructions and warning instructions. The production line control terminal 002 is used to control the production of the production line according to the control instructions.

[0071] Specifically, the automatic image acquisition terminal 001 can be deployed between the work station to be detected and the next work station. For example, in the task of industrial control of the soldering process quality of a circuit board using the intelligent industrial control system 00 based on 5G technology, the automatic image acquisition terminal 001 can be deployed between the mounter and the corresponding next work station.

[0072] In this embodiment, the automatic image acquisition terminal 001 includes an image acquisition device 0011, a warning indication device 0013, and a 5G communication module 0012. Among them, the image acquisition device 0011 is used to acquire the circuit image of the circuit board to be detected and the identity-related data of the circuit board to be detected. Then, the acquired circuit image and identity-related data are sent to the server terminal 000 by using the 5G communication module 0012. The warning indication device 0013 is used to issue an abnormal warning according to the warning instruction sent by the server terminal 000 received by the 5G communication module 0012. For example, the abnormal warning can be a light warning of a preset color or a voice warning of a preset broadcast message.

[0073] In this embodiment, the server terminal 000 includes a 5G communication module 0001, an image recognition module 0002, and an image comparison module 0003. Among them, the server terminal 000 receives the circuit image and identity-related data sent by the automatic image acquisition terminal 001 through the 5G communication module 0001, and then uses the image recognition module 0002 to perform preprocessing and image recognition operations on the circuit image.

[0074] Preferably, the pre-trained circuit image recognition model can be deployed in the image recognition module 0002. By using the circuit image recognition model, the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image are recognized and calculated, and a simulated circuit diagram of the circuit board to be detected is generated according to the position coordinate data and the line coordinate data.

[0075] In this embodiment, the image comparison module 0003 is used to compare the simulated circuit diagram with the preset standard simulated circuit diagram for consistency. When the consistency comparison result is abnormal, a pause instruction is sent to the production line control terminal 002 through the 5G communication module 0001 and an abnormal warning instruction is sent to the automatic image acquisition terminal 001.

[0076] In this embodiment, the production line control terminal 002 is provided with a 5G communication module 0021, which receives the pause instruction sent by the server terminal 000 by using the 5G communication module 0021 and controls the production of the production line where the circuit board to be detected is located according to the pause instruction.

[0077] Refer to Figure 2 As shown, it is a schematic flowchart of an intelligent industrial control method based on 5G technology provided by an embodiment of the present invention. In this embodiment, the intelligent industrial control method based on 5G technology includes:

[0078] S1. Receive the circuit image of the circuit board to be detected and the identity - related data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module;

[0079] In the embodiment of the present invention, taking a PCB (Printed Circuit Board) as an example, an intelligent industrial control method based on 5G technology is described. It should be noted that the intelligent industrial control method based on 5G technology provided by this application of the present invention is also applicable to the mounting detection of circuit boards such as ceramic circuit boards, alumina ceramic circuit boards, circuit boards, and aluminum substrates.

[0080] In the embodiment of the present invention, in order to be able to collect the circuit board image and identity information of the circuit board to be detected, an automatic image acquisition terminal can be assembled between the pick - and - place machine and the corresponding next workstation. Among them, the selection of the camera, lens, and light source involved in the automatic image acquisition terminal can be set according to the actual situation, and the acquisition route and photographing frequency of the automatic image acquisition terminal can be set according to the movement path and speed of the actual PCB board to be detected.

[0081] Preferably, an early warning signal indicating device and the preset 5G communication module can be set in the automatic image acquisition terminal. By using the preset 5G communication module, the circuit image of the circuit board to be detected collected can be quickly transmitted at a high rate and with low latency.

[0082] In the embodiment of the present invention, the identity information of the circuit board to be detected can be realized by pre - spraying a two - dimensional code with the main board identity information at a preset position on the circuit board to be detected. The two - dimensional code information includes, but is not limited to, information such as the model and production batch of the corresponding circuit board.

[0083] S2. Use the pre - trained circuit image recognition model to identify the components and circuit traces in the circuit image;

[0084] It can be understood that in the circuit image, each patch area, that is, the area where a single component or circuit trace is located, is a very small area relative to the entire circuit image area. In order to accurately obtain the patch area and exclude the interference area, before using the pre - trained circuit image recognition model to identify the components and circuit traces in the circuit image, pre - processing operations need to be performed on the circuit image.

[0085] Exemplarily, the pre - processing operations on the circuit image include, but are not limited to: performing pixel correction on the circuit image using a flat - field correction algorithm, performing binarization processing on the image after pixel correction to obtain a binarized image, filling the hole areas in the binarized image using a closing operation algorithm, and performing edge smoothing processing on the binarized image using an opening operation algorithm.

[0086] In the embodiments of the present invention, performing flat-field calibration on the circuit image can eliminate the problem of non-uniform pixel response in the circuit image, ensuring that the gray values of all pixel points in the circuit image are relatively uniform. The circuit image is binarized because the pixel values of a grayscale image are 256 levels, and performing data operations on a grayscale image is very time-consuming. If the grayscale image is converted into a binary image with pixel values of 2 levels, a large amount of data operations can be reduced. At the same time, the binary image can reflect the geometric features of the image, which is beneficial to the separation of the patch area and the background area. Filling the hole areas and smoothing the edges of the circuit image can make the shape features of the pads in the circuit image unified.

[0087] In the embodiments of the present invention, the pre-trained circuit image recognition model can be an image recognition model based on a neural network. Using a large amount of circuit board image samples, the circuit image recognition model is trained. During this training process, the circuit image recognition model continuously learns the image features in the circuit board image samples and then maps the features of the image to the neural network for image recognition and classification.

[0088] Specifically, referring to Figure 3 as shown, the recognition of components and circuit traces in the circuit image using the pre-trained circuit image recognition model includes:

[0089] S21. Perform a vector conversion operation on the circuit image to obtain a vector matrix corresponding to the circuit image;

[0090] S22. Use the pre-trained circuit image recognition model to extract the component features and circuit trace features of the vector matrix;

[0091] S23. Use a pre-trained activation function to match the preset component labels corresponding to each component feature, and label the corresponding components according to the matched component labels;

[0092] S24. Label the circuit trace pixel points in the circuit image according to the circuit trace features, and connect the circuit trace pixel points to obtain the circuit traces.

[0093] In the embodiments of the present invention, the pre-trained circuit image recognition model includes an input layer, a convolutional layer, and an output layer. Among them, through the input layer, vector encoding is performed on the circuit image, and the circuit image information is converted into digital information recognizable by a computer. In the convolutional layer, convolutional calculations are performed on the vector matrix to extract the component features and circuit trace features of the circuit image. In the output layer, by connecting with a pre-trained activation function, the relative probability value between the component features and the preset component labels is calculated, and the component label with the highest probability value is selected as the classification of the corresponding component.

[0094] In the embodiments of the present invention, the pre-trained activation functions include but are not limited to the softmax activation function, the sigmoid activation function, and the relu activation function. The preset component labels include but are not limited to resistors, capacitors, diodes, voltage regulators, and transistors, etc.

[0095] In one of the embodiments of the present invention, the following activation function can be used to calculate the relative probability value:

[0096]

[0097] Among them, p(a|x) is the relative probability value between the component feature x and the component label a, w a is the weight vector of the component label a, T is the transpose operation symbol, exp is the expectation operation symbol, and A is the number of preset component labels.

[0098] In the embodiments of the present invention, with the aid of a pre-trained circuit image recognition model, the components and circuit traces in the circuit image are recognized, realizing the separation of the patch area and the pad background area.

[0099] S3. Calculate the position coordinate data of each component in the circuit image and calculate the line coordinate data of the circuit traces in the circuit image;

[0100] It can be understood that usually in the circuit board mounting process, multiple optical positioning points, also called fiducial points or Mark points, are set at preset positions on the circuit board. The fiducial points are used as common measurable points in all steps of the mounting process. Therefore, in the embodiments of the present invention, the preset fiducial points in the circuit board to be detected are used to locate each component and the circuit traces.

[0101] Specifically, referring to Figure 4 as shown, the calculation of the position coordinate data of each component in the circuit image includes:

[0102] S31. Identify the fiducial point area corresponding to the preset fiducial point in the circuit board to be detected in the circuit image;

[0103] S32. Obtain the center of each of the reference point regions, and generate the coordinate origin of the circuit image according to the spatial correspondence relationship between the centers of each of the reference point regions;

[0104] S33. Perform edge detection processing on the labeled components to obtain the graphic border corresponding to each of the components;

[0105] S34. Randomly select a preset number of pixel points from each of the graphic borders as measurement points in sequence, calculate the spatial distance between each of the measurement points and the coordinate origin, and obtain the coordinate values of each of the measurement points;

[0106] S35. Aggregate the coordinate values of each of the measurement points to obtain the position coordinate data of the corresponding component.

[0107] In the embodiment of the present invention, the pixel features of the preset reference points can be extracted by using the pre-trained circuit image recognition model, and the pixel features of the preset reference points are mapped to the corresponding reference point regions.

[0108] It can be understood that generally, there can be multiple reference points on a circuit board, and each reference point is mostly circular. The accurate position of the corresponding reference point can be determined by obtaining the center of each of the reference point regions. In practical applications, since the reference point regions sometimes deform, which affects the accurate positioning of the reference points, in the embodiment of the present invention, the following method is used to perform ellipse fitting on the reference point regions to ensure the accuracy of reference point positioning:

[0109] Separate the reference point region from the background of the circuit image;

[0110] Calculate the edge points of the separated reference point region to obtain the image contour corresponding to the reference point region;

[0111] Perform ellipse fitting on the image contour, and generate the circumscribed rectangle corresponding to the image contour after ellipse fitting;

[0112] Calculate the ratio of the difference between the length and the width of the circumscribed rectangle to the width of the circumscribed rectangle;

[0113] When the ratio is less than the preset ratio threshold, the center point of the image contour after fitting is used as the center of the corresponding reference region.

[0114] In the embodiment of the present invention, the preset ratio threshold can be determined according to the actual debugging data. For example, the preset ratio threshold can be 0.1.

[0115] In an embodiment of the present invention, a point with the smallest average distance from the centers of each reference point region can be selected through calculation as the coordinate origin of the circuit image.

[0116] It can be understood that the circuit traces on a circuit board usually include shapes such as straight lines and curves, and corresponding coordinate data can be collected according to the different shapes of the circuit traces.

[0117] Specifically, referring to Figure 5 as shown, calculating the line coordinate data of the circuit traces in the circuit image includes:

[0118] S36. Segment the circuit trace according to the shape of the circuit trace;

[0119] S37. Calculate the starting coordinate value of each segment relative to the coordinate origin and the ending coordinate value of each segment relative to the coordinate origin in sequence;

[0120] S38. Calculate the azimuth angle of the corresponding segment according to the starting coordinate value and the ending coordinate value;

[0121] S39. When the segment is in the shape of a circular curve, calculate the radius of the circular curve of the corresponding segment;

[0122] S40. Collect the starting coordinate value, ending coordinate value, azimuth angle, and radius of the circular curve of each segment to obtain the line coordinate data.

[0123] In an embodiment of the present invention, the coordinate data of each segment can be calculated using a line coordinate calculation formula.

[0124] Exemplarily, when the segment is a straight line, the coordinate value of the segment can be calculated using the following line coordinate formula.

[0125] X = X0 + L * cosα1

[0126] Y = Y0 + L * sinα1

[0127] Wherein, X and Y are the abscissa and ordinate of the segment relative to the coordinate origin respectively, X0 is the starting coordinate value of the segment, Y0 is the ending coordinate value of the segment, α1 is the azimuth angle of the segment, and L is the distance from the coordinate origin to the starting point of the segment.

[0128] In the embodiments of the present invention, by calculating the position coordinate data of each component in the circuit image and calculating the line coordinate data of the circuit traces in the circuit image, it is beneficial to form the spatial layout relationship of each component and circuit trace in the to-be-detected circuit board under the background of the circuit board pads with the help of the position coordinate data and line coordinate data, which facilitates subsequent investigation of the problem of component offset.

[0129] S4. Generate a simulated circuit diagram of the to-be-detected circuit board according to the position coordinate data and the line coordinate data, and obtain a preset standard simulated circuit diagram corresponding to the identity information of the to-be-detected circuit board;

[0130] In the embodiments of the present invention, the spatial position relationship of each component in the pads, the spatial relative relationship between each component, and the spatial position relationship of the circuit traces in the pads reflected by the simulated circuit diagram are the same as those of the components and circuit layout in the circuit image.

[0131] In the embodiments of the present invention, the coordinate origin of the simulated circuit diagram can be randomly initialized, and then the corresponding components and circuit traces can be arranged according to the position coordinate data and the line coordinate data, where each component can be replaced by a standard component symbol.

[0132] In the embodiments of the present invention, the preset standard simulated circuit diagram is an accurate and standard graph of the layout of each component and circuit trace in the pads based on the original technical solution of the to-be-detected circuit board.

[0133] In the embodiments of the present invention, a standard simulated circuit matching the identity information of the to-be-detected circuit board can be queried in a mapping relationship table between a preset main board type and a standard simulated circuit diagram.

[0134] In the embodiments of the present invention, using the simulated circuit diagram to reflect the spatial layout of each component in the to-be-detected circuit board can effectively eliminate the interference information in the circuit image and realize the purity of the nuclear comparison and detection information.

[0135] S5. Perform a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, a pause instruction is sent to a preset production line control terminal through a 5G communication module and an abnormal warning instruction is sent to the automatic image acquisition terminal.

[0136] In an embodiment of the present invention, the image matching function of an image recognition model based on a neural network can be pre-trained, and the trained image recognition model is used to analyze the difference between the simulated circuit diagram and the standard simulated circuit diagram.

[0137] In another embodiment of the present invention, the consistency comparison between the analog circuit diagram and the preset standard analog circuit diagram is realized by using the corner features of the image. Usually, at the corners of an image, not only is the gradient value large, but also the rate of change of the gradient direction is large. That is to say, the corners show the positions where the gray level changes violently in the two-dimensional space of the image, and there are obvious differences between them and the surrounding adjacent points. Therefore, the information of each local part in the image can be obtained by calculating the corners in the image.

[0138] Specifically, referring to Figure 6 as shown, the consistency comparison between the analog circuit diagram and the preset standard analog circuit diagram to obtain a consistency comparison result includes:

[0139] S51. Respectively extract the corner features of the analog circuit diagram and the corner features of the preset standard analog circuit;

[0140] S52. Perform binary encoding on the corner features of the analog circuit diagram to obtain a comparison corner feature vector, and perform binary encoding on the corner features of the preset standard analog circuit to obtain a reference corner feature vector;

[0141] S53. Calculate the distance between the comparison corner feature vector and the reference corner feature vector;

[0142] S54. When the distance is less than or equal to a preset distance threshold, output result information indicating consistent comparison;

[0143] S55. When the distance is not less than the preset distance threshold, output result information indicating inconsistent comparison.

[0144] In the embodiment of the present invention, the Harris algorithm can be used to extract the corner features of the analog circuit diagram and the corner features of the preset standard analog circuit, and then the BRISK feature description method is used to convert the detected corner features into binary encoding, so as to obtain a corner feature vector described in a binary manner. Finally, the Hamming distance between the two is calculated. When the distance is not less than the preset distance threshold, it indicates that the difference between the two images is large, and vice versa, it indicates that the two images are similar. Among them, the distance threshold can be determined according to actual debugging data.

[0145] In the embodiment of the present invention, when the comparison result is inconsistent, warning industrial control information or pause industrial control information can be generated, etc. When the comparison result is consistent, production continuation industrial control information can be generated.

[0146] In the embodiments of the present invention, when outputting result information with inconsistent comparison, industrial control information of a pause instruction can be sent to a production line control terminal through a pre-installed 5G communication module, so as to control the production line control terminal to stop the production of the production line, and an abnormal warning instruction is sent to the corresponding image acquisition terminal to control the warning signal indicating device of the corresponding image acquisition terminal to issue an abnormal warning. For example, the abnormal warning can be a light warning of a preset color or a voice warning of a preset broadcast message.

[0147] Preferably, if the analysis and comparison result is abnormal, an abnormal report can also be generated. The abnormal report includes, but is not limited to, information such as the abnormal mounting position or circuit position, and the possible consequences of the abnormality, and the abnormal report is sent to a pre-determined terminal through the 5G communication module.

[0148] In the embodiments of the present invention, the circuit image of the circuit board to be detected is collected. By identifying the components and circuit traces in the circuit image, the separation of the object to be detected from the pad background is realized, and the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image are calculated. Further, according to the position coordinate data and the line coordinate data, a simulated circuit diagram of the circuit board to be detected is generated, and the actual mounting result of the circuit board to be detected is mapped into the simulated circuit diagram. Finally, through the graphic consistency comparison between the simulated circuit diagram and the standard simulated circuit diagram, the detection of the mounting quality of the circuit board is realized, and this detection method improves the efficiency and reduces the cost compared with manual detection.

[0149] As Figure 7 shown, it is a functional module diagram of an intelligent industrial control device based on 5G technology provided by an embodiment of the present invention.

[0150] The intelligent industrial control device 100 based on 5G technology of the present invention can be installed in an electronic device. According to the functions realized, the intelligent industrial control device 100 based on 5G technology includes: an image and identity information receiving module 101, an image recognition module 102, a simulated circuit diagram generation module 103, a standard circuit diagram acquisition module 104, and an industrial control module 105. The modules of the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0151] In this embodiment, the functions of each module / unit are as follows:

[0152] The image and identity information receiving module 101 is used to receive the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by an automatic image acquisition terminal through a 5G communication module;

[0153] The image recognition module 102 is used to recognize the components and circuit traces in the circuit image by using a pre-trained circuit image recognition model;

[0154] The analog circuit diagram generation module 103 is used to calculate the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image, and generate an analog circuit diagram of the to-be-detected circuit board according to the position coordinate data and the line coordinate data;

[0155] The standard circuit diagram acquisition module 104 is used to acquire a preset standard analog circuit diagram corresponding to the identity information of the to-be-detected circuit board;

[0156] The industrial control module 105 is used to perform a consistency comparison between the analog circuit diagram and the preset standard analog circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, a pause instruction is sent to a preset production line control terminal through the 5G communication module and an abnormal warning instruction is sent to the automatic image acquisition terminal.

[0157] Specifically, the specific implementation manners of each module of the intelligent industrial control device 100 based on 5G technology are as follows:

[0158] Step 1: Receive the circuit image of the to-be-detected circuit board and the identity association data of the to-be-detected circuit board sent by the automatic image acquisition terminal through the 5G communication module;

[0159] In the embodiment of the present invention, a PCB (Printed Circuit Board) is taken as an example to illustrate an intelligent industrial control method based on 5G technology. It should be noted that the intelligent industrial control method based on 5G technology provided by the present invention application is also applicable to the mounting detection of circuit boards such as ceramic circuit boards, alumina ceramic circuit boards, circuit boards, and aluminum substrates.

[0160] In the embodiment of the present invention, in order to be able to collect the circuit board image and identity information of the to-be-detected circuit board, an automatic image acquisition terminal can be assembled between the pick-and-place machine and the corresponding next workstation. Among them, the selection of the camera, lens and the selection of the light source involved in the automatic image acquisition terminal can be set according to the actual situation, and the acquisition route and photographing frequency of the automatic image acquisition terminal can be set according to the movement path and speed of the actual to-be-detected PCB board.

[0161] Preferably, a warning signal indicating device and the preset 5G communication module can be set in the automatic image acquisition terminal, and the preset 5G communication module can be used to realize the high-speed and low-latency fast transmission of the collected circuit image of the to-be-detected circuit board.

[0162] In an embodiment of the present invention, the identity information of the circuit board to be detected can be realized by pre-spraying a two-dimensional code of the main board identity information at a preset position on the circuit board to be detected. The two-dimensional code information includes, but is not limited to, information such as the model number and production batch of the corresponding circuit board.

[0163] Step 2: Use the pre-trained circuit image recognition model to recognize the components and circuit traces in the circuit image;

[0164] It can be understood that in the circuit image, each patch area, that is, the area where a single component or circuit trace is located, is a very small area relative to the entire circuit image area. In order to accurately obtain the patch area and exclude the interference area, before using the pre-trained circuit image recognition model to recognize the components and circuit traces in the circuit image, preprocessing operations need to be performed on the circuit image.

[0165] Exemplarily, the preprocessing operations on the circuit image include, but are not limited to: using a flat-field correction algorithm to perform pixel correction on the circuit image, performing binarization processing on the image after pixel correction to obtain a binarized image, using a closing operation algorithm to perform hole area filling processing on the binarized image, and using an opening operation algorithm to perform edge smoothing processing on the binarized image.

[0166] In an embodiment of the present invention, performing flat-field calibration on the circuit image can eliminate the problem of non-uniform pixel response in the circuit image and ensure that the gray values of all pixel points in the circuit image are relatively uniform. Performing binarization processing on the circuit image is because the pixel values of the gray-scale image are 256 levels, and performing data operations on the gray-scale image is very time-consuming. If the gray-scale image is converted into a binary image with pixel values of 2 levels, a large amount of data operations will be reduced. At the same time, the binary image can reflect the geometric features of the image, which is beneficial to the separation of the patch area and the background area. Filling the hole area and smoothing the edge of the circuit image can make the shape features of the pads in the circuit image unified.

[0167] In an embodiment of the present invention, the pre-trained circuit image recognition model can be an image recognition model based on a neural network. Using a large amount of circuit board image samples to train the circuit image recognition model. During this training process, the circuit image recognition model continuously learns the image features in the circuit board image samples, and then maps the features possessed by the image to the neural network for image recognition and classification.

[0168] Specifically, the use of the pre-trained circuit image recognition model to recognize the components and circuit traces in the circuit image includes:

[0169] Perform a vector conversion operation on the circuit image to obtain a vector matrix corresponding to the circuit image;

[0170] Use the pre-trained circuit image recognition model to extract the component features and circuit trace features of the vector matrix;

[0171] Use a pre-trained activation function to match the preset component labels corresponding to each of the component features, and label the corresponding components according to the matched component labels;

[0172] Mark the circuit trace pixel points in the circuit image according to the circuit trace features, and connect the circuit trace pixel points to obtain the circuit traces.

[0173] In an embodiment of the present invention, the pre-trained circuit image recognition model includes an input layer, a convolutional layer, and an output layer. Among them, the circuit image is vector-encoded through the input layer, and the circuit image information is converted into digital information recognizable by a computer. In the convolutional layer, the component features and circuit trace features of the circuit image are extracted by performing convolutional calculations on the vector matrix. In the output layer, by connecting with a pre-trained activation function, the relative probability value between the component feature and the preset component label is calculated, and the component label with the highest probability value is selected as the classification of the corresponding component.

[0174] In an embodiment of the present invention, the pre-trained activation function includes but is not limited to the softmax activation function, sigmoid activation function, relu activation function, and the preset component labels include but are not limited to resistors, capacitors, diodes, voltage regulators, and triodes, etc.

[0175] In one embodiment of the present invention, the relative probability value can be calculated using the following activation function:

[0176]

[0177] where p(a|x) is the relative probability value between the component feature x and the component label a, w a is the weight vector of the component label a, T is the transpose operation symbol, exp is the expectation operation symbol, and A is the number of preset component labels.

[0178] In an embodiment of the present invention, by means of a pre-trained circuit image recognition model, the components and circuit traces in the circuit image are recognized, realizing the separation of the patch area and the pad background area.

[0179] Step three: Calculate the position coordinate data of each component in the circuit image and calculate the line coordinate data of the circuit traces in the circuit image;

[0180] It can be understood that in the usual circuit board mounting process, a plurality of optical positioning points, also known as fiducial points or Mark points, are set at preset positions on the circuit board. The fiducial points are used as common measurable points in all steps of the mounting process. Therefore, in the embodiments of the present invention, the positioning of each of the components and the circuit traces is achieved by means of the preset fiducial points in the circuit board to be detected.

[0181] Specifically, calculating the position coordinate data of each of the components in the circuit image includes:

[0182] Identifying the fiducial point region corresponding to the preset fiducial point in the circuit board to be detected in the circuit image;

[0183] Obtaining the center of the circle of each of the fiducial point regions, and generating the coordinate origin of the circuit image according to the spatial correspondence relationship between the centers of the circles of each of the fiducial point regions;

[0184] Performing edge detection processing on the labeled components to obtain the graphic border corresponding to each of the components;

[0185] Randomly selecting a preset number of pixel points from each of the graphic borders as measurement points in sequence, calculating the spatial distance between each of the measurement points and the coordinate origin, and obtaining the coordinate value of each of the measurement points;

[0186] Collecting the coordinate values of each of the measurement points to obtain the position coordinate data of the corresponding component.

[0187] In the embodiments of the present invention, the pixel features of the preset fiducial points can be extracted by using the pre-trained circuit image recognition model, and mapped to the corresponding fiducial point regions according to the pixel features of the preset fiducial points.

[0188] It can be understood that usually there can be multiple fiducial points on a circuit board, and each fiducial point is mostly circular. The accurate position of the corresponding fiducial point can be determined by obtaining the center of the circle of each of the fiducial point regions. In practical applications, since the fiducial point region sometimes deforms, which affects the accurate positioning of the fiducial point, in the embodiments of the present invention, the following method is used to perform elliptical fitting on the fiducial point region to ensure the accuracy of fiducial point positioning:

[0189] Separating the fiducial point region from the background of the circuit image;

[0190] Calculating the edge points of the separated fiducial point region to obtain the image contour corresponding to the fiducial point region;

[0191] Performing elliptical fitting on the image contour, and generating the circumscribed rectangle corresponding to the image contour after elliptical fitting;

[0192] Calculate the ratio of the difference between the length and the width of the circumscribed rectangle to the width of the circumscribed rectangle;

[0193] When the ratio is less than a preset ratio threshold, the center point of the fitted image contour is taken as the center of the corresponding reference area.

[0194] In the embodiment of the present invention, the preset ratio threshold can be determined according to actual debugging data. For example, the preset ratio threshold can be 0.1.

[0195] In the embodiment of the present invention, a point with the smallest average distance from the center of each reference point area can be selected through calculation as the coordinate origin of the circuit image.

[0196] It can be understood that the circuit traces in a circuit board usually include shapes such as straight lines and curves, and corresponding coordinate data can be collected according to the different shapes of the circuit traces.

[0197] Specifically, calculating the line coordinate data of the circuit traces in the circuit image includes:

[0198] Segment the circuit trace according to the shape of the circuit trace;

[0199] Calculate the starting coordinate value of the starting point of each segment relative to the coordinate origin and the ending coordinate value of the ending point of each segment relative to the coordinate origin in sequence;

[0200] Calculate the azimuth angle of the corresponding segment according to the starting coordinate value and the ending coordinate value;

[0201] When the segment is in the shape of a circular curve, calculate the radius of the circular curve of the corresponding segment;

[0202] Collect the starting coordinate value, ending coordinate value, azimuth angle and circular curve radius of each segment to obtain the line coordinate data.

[0203] In the embodiment of the present invention, the coordinate data of each segment can be calculated by using a line coordinate calculation formula.

[0204] Exemplarily, when the segment is a straight line, the following line coordinate formula can be used to calculate the coordinate value of the segment.

[0205] X = X0 + L * cosα1

[0206] Y = Y0 + L * sinα1

[0207] Wherein, X and Y are respectively the abscissa and ordinate of the segment relative to the coordinate origin, X0 is the starting coordinate value of the segment, Y0 is the ending coordinate value of the segment, α1 is the azimuth angle of the segment, and L is the distance from the coordinate origin to the starting point of the segment.

[0208] In the embodiments of the present invention, by calculating the position coordinate data of each component in the circuit image and calculating the line coordinate data of the circuit traces in the circuit image, it is beneficial to form the spatial layout relationship of each component and circuit traces in the to-be-detected circuit board under the background of the circuit board pads with the help of the position coordinate data and line coordinate data, facilitating the subsequent investigation of the problem of component offset.

[0209] Step Four: Generate a simulated circuit diagram of the to-be-detected circuit board according to the position coordinate data and the line coordinate data, and obtain a preset standard simulated circuit diagram corresponding to the identity information of the to-be-detected circuit board;

[0210] In the embodiments of the present invention, the spatial position relationship of each component in the pads, the spatial relative relationship between each component, and the spatial position relationship of the circuit traces in the pads reflected by the simulated circuit diagram are the same as those of the components and circuit layout in the circuit image.

[0211] In the embodiments of the present invention, the coordinate origin of the simulated circuit diagram can be randomly initialized, and then the corresponding components and circuit traces can be arranged according to the position coordinate data and the line coordinate data, where each component can be replaced by a standard component symbol.

[0212] In the embodiments of the present invention, the preset standard simulated circuit diagram is an accurate and standard graphic of the layout of each component and circuit traces in the pads based on the original technical solution of the to-be-detected circuit board.

[0213] In the embodiments of the present invention, a standard simulated circuit matching the identity information of the to-be-detected circuit board can be queried in the mapping relationship table between the preset main board types and the standard simulated circuit diagrams.

[0214] In the embodiments of the present invention, using the simulated circuit diagram to reflect the spatial layout of each component in the to-be-detected circuit board can effectively eliminate the interference information in the circuit image and achieve the purity of the nuclear comparison and detection information.

[0215] Step Five: Perform a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, a pause instruction is sent to a preset production line control terminal through a 5G communication module and an abnormal warning instruction is sent to the automatic image acquisition terminal.

[0216] In one embodiment of the present invention, the image matching function of the neural network-based image recognition model can be pre-trained, and the trained image recognition model is used to analyze the differences between the simulated circuit diagram and the standard simulated circuit diagram.

[0217] In another embodiment of the present invention, the consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram is realized by using the corner features of the image. Usually, at the corners of an image, not only the gradient value is large, but also the change rate of the gradient direction is large. That is to say, the corners show the positions where the gray level changes violently in the two-dimensional space of the image, and there are obvious differences between them and the surrounding neighboring points. Therefore, the information of each local part of the image can be obtained by calculating the corners in the image.

[0218] Specifically, the consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result includes:

[0219] Extract the corner features of the simulated circuit diagram and the corner features of the preset standard simulated circuit respectively;

[0220] Perform binary encoding on the corner features of the simulated circuit diagram to obtain a comparison corner feature vector, and perform binary encoding on the corner features of the preset standard simulated circuit to obtain a reference corner feature vector;

[0221] Calculate the distance between the comparison corner feature vector and the reference corner feature vector;

[0222] When the distance is less than or equal to a preset distance threshold, output the result information indicating consistent comparison;

[0223] When the distance is not less than the preset distance threshold, output the result information indicating inconsistent comparison.

[0224] In the embodiment of the present invention, the Harris algorithm can be used to extract the corner features of the simulated circuit diagram and the corner features of the preset standard simulated circuit, and then the BRISK feature description method is used to convert the detected corner features into binary encoding, so as to obtain the corner feature vector described in binary form. Finally, the Hamming distance between the two is calculated. When the distance is not less than the preset distance threshold, it indicates that the differences between the two images are large, and vice versa, it indicates that the two images are similar. Among them, the distance threshold can be determined according to the actual debugging data.

[0225] In the embodiment of the present invention, when the comparison result is inconsistent, warning industrial control information or pause industrial control information can be generated, etc. When the comparison result is consistent, production continuation industrial control information can be generated.

[0226] In an embodiment of the present invention, when outputting result information with inconsistent comparison, industrial control information of a pause instruction can be sent to a production line control terminal through a pre-installed 5G communication module to control the production line control terminal to stop the production of the production line, and an abnormal warning instruction can be sent to a corresponding image acquisition terminal to control a warning signal indicating device of the corresponding image acquisition terminal to issue an abnormal warning. For example, the abnormal warning can be a light warning of a preset color or a voice warning of a preset broadcast message.

[0227] Preferably, if the analysis and comparison result is abnormal, an abnormal report can also be generated. The abnormal report includes, but is not limited to, information such as the abnormal mounting position or circuit position, and the possible consequences of the abnormality, and the abnormal report is sent to a pre-determined terminal through the 5G communication module.

[0228] In an embodiment of the present invention, a circuit image of a circuit board to be detected is collected. By identifying components and circuit traces in the circuit image, separation of the target to be detected from the pad background is achieved, and position coordinate data of each component in the circuit image and line coordinate data of the circuit traces in the circuit image are calculated. Further, according to the position coordinate data and the line coordinate data, a simulated circuit diagram of the circuit board to be detected is generated, and the actual mounting result of the circuit board to be detected is mapped into the simulated circuit diagram. Finally, through graphic consistency comparison between the simulated circuit diagram and a standard simulated circuit diagram, detection of the mounting quality of the circuit board is realized. Therefore, the intelligent industrial control device based on 5G technology proposed by the present invention improves the detection efficiency of circuit board mounting and reduces costs compared with manual detection.

[0229] As Figure 8 shown, it is a schematic structural diagram of an electronic device for implementing an intelligent industrial control method based on 5G technology provided by an embodiment of the present invention.

[0230] The electronic device 1 may include a processor 10, a memory 11, and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an intelligent industrial control program based on 5G technology.

[0231] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 11 can be an internal storage unit of the electronic device 1 in some embodiments, such as the mobile hard disk of the electronic device 1. The memory 11 can also be an external storage device of the electronic device 1 in some other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 1. Further, the memory 11 can also include both the internal storage unit and the external storage device of the electronic device 1. The memory 11 can be used not only to store application software installed on the electronic device 1 and various types of data, such as the code of an intelligent industrial control program based on 5G technology, etc., but also to temporarily store data that has been output or will be output.

[0232] The processor 10 can be composed of integrated circuits in some embodiments. For example, it can be composed of a single packaged integrated circuit, or can also be composed of multiple integrated circuits with the same or different functions packaged together, including a combination of one or more Central Processing Units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and lines, and by running or executing programs or modules stored in the memory 11 (such as an intelligent industrial control program based on 5G technology, etc.), and calling data stored in the memory 11, to execute various functions of the electronic device 1 and process data.

[0233] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is set to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0234] Figure 8 Only the electronic device with components is shown. Those skilled in the art can understand that, Figure 8The structure shown does not constitute a limitation on the electronic device 1, and it may include fewer or more components than those shown, or combine certain components, or have a different component arrangement.

[0235] For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component. Preferably, the power source can be logically connected to the at least one processor 10 through a power management device, so as to implement functions such as charging management, discharging management, and power consumption management through the power management device. The power source may also include any components such as one or more DC or AC power sources, a recharge device, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may also include various sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0236] Furthermore, the electronic device 1 may further include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0237] Optionally, the electronic device 1 may further include a user interface. The user interface may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visual user interface.

[0238] It should be understood that the embodiments are only for illustration purposes and are not limited by this structure in the scope of the patent application.

[0239] The intelligent industrial control program based on 5G technology stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can implement:

[0240] Receiving the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module;

[0241] Using a pre-trained circuit image recognition model to identify the components and circuit traces in the circuit image;

[0242] Calculate the position coordinate data of each of the components in the circuit image and calculate the line coordinate data of the circuit traces in the circuit image;

[0243] Generate a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data;

[0244] Obtain a preset standard simulated circuit diagram corresponding to the identity association data of the circuit board to be detected;

[0245] Perform a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, send a pause instruction to a preset production line control terminal through a 5G communication module and send an abnormal warning instruction to the automatic image acquisition terminal.

[0246] Further, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory).

[0247] The present invention also provides a computer-readable storage medium. The readable storage medium stores a computer program. When the computer program is executed by a processor of an electronic device, it can implement:

[0248] Receive the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module;

[0249] Use a pre-trained circuit image recognition model to identify the components and circuit traces in the circuit image;

[0250] Calculate the position coordinate data of each of the components in the circuit image and calculate the line coordinate data of the circuit traces in the circuit image;

[0251] Generate a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data;

[0252] Obtain a preset standard simulated circuit diagram corresponding to the identity association data of the circuit board to be detected;

[0253] Perform a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, send a pause instruction to the preset production line control terminal through the 5G communication module and send an abnormal warning instruction to the automatic image acquisition terminal.

[0254] In addition, in each embodiment of the present invention, each functional module can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a hardware plus software functional module.

[0255] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0256] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any associated drawing marks in the claims should not be regarded as limiting the claims involved.

[0257] The blockchain referred to in the present invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithms. Blockchain, essentially a decentralized database, is a string of data blocks generated by using cryptographic methods. Each data block contains information about a batch of network transactions, which is used to verify the validity (anti-counterfeiting) of the information and generate the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, an application service layer, etc.

[0258] The embodiments of the present application can acquire and process relevant data based on holographic projection technology. Among them, Artificial Intelligence (AI) is the theory, method, technology, and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0259] In addition, obviously, the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices stated in the system claims can also be implemented by one unit or device through software or hardware. Words such as second are used to denote names and do not denote any specific order.

[0260] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent industrial control method based on 5G technology, characterized in that, The method includes: Receiving the circuit image of the circuit board to be detected and the identity - associated data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module; Performing vector encoding on the circuit image by using the input layer of the pre - trained circuit image recognition model to obtain a vector matrix corresponding to the circuit image, performing convolution calculation on the vector matrix by using the convolution layer of the circuit image recognition model to extract the component features and circuit trace features of the circuit image, using the pre - trained activation function to match the component label corresponding to each component feature, marking the corresponding components in the circuit image according to the component labels to obtain the components in the circuit image, marking the circuit trace pixel points in the circuit image according to the circuit trace features, and connecting the circuit trace pixel points to obtain the circuit traces in the circuit image; Calculating the position coordinate data of each component in the circuit image and calculating the line coordinate data of the circuit traces in the circuit image to obtain the spatial layout relationship of each component and circuit trace in the circuit board to be detected; Generating a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data; Obtaining a preset standard simulated circuit diagram corresponding to the identity - associated data of the circuit board to be detected; Performing a consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, a pause instruction is sent to a preset production line control terminal through the 5G communication module and an abnormal warning instruction is sent to the automatic image acquisition terminal.

2. The intelligent industrial control method based on 5G technology according to claim 1, characterized in that, The calculating the position coordinate data of each component in the circuit image includes: Identifying the reference point area corresponding to the preset reference point in the circuit board to be detected in the circuit image; Obtaining the center of each reference point area, and generating the coordinate origin of the circuit image according to the spatial correspondence relationship between the centers of each reference point area; Performing edge detection processing on the marked components to obtain a graphic border corresponding to each component; Randomly selecting a preset number of pixel points from each graphic border as measurement points in sequence, calculating the spatial distance between each measurement point and the coordinate origin to obtain the coordinate value of each measurement point; Collecting the coordinate values of each measurement point to obtain the position coordinate data of the corresponding component.

3. The intelligent industrial control method based on 5G technology according to claim 2, characterized in that, The calculating the line coordinate data of the circuit traces in the circuit image includes: Segmenting the circuit traces according to the shape of the circuit traces; Calculating the starting coordinate value of each segment relative to the coordinate origin and the ending coordinate value of each segment relative to the coordinate origin in sequence; Calculating the azimuth angle of the corresponding segment according to the starting coordinate value and the ending coordinate value; When the segment is in a circular curve shape, calculating the radius of the circular curve of the corresponding segment; Collecting the starting coordinate value, ending coordinate value, azimuth angle and circular curve radius of each segment to obtain the line coordinate data.

4. The intelligent industrial control method based on 5G technology according to claim 2, characterized in that, The obtaining the center of each reference point area includes: Separate the reference point area from the background of the circuit image; Calculate the edge points of the separated reference point area to obtain the image contour corresponding to the reference point area; Perform elliptical fitting on the image contour and generate a circumscribed rectangle corresponding to the image contour after elliptical fitting; Calculate the ratio of the difference between the length and width of the circumscribed rectangle to the width of the circumscribed rectangle; When the ratio is less than a preset ratio threshold, use the center point of the image contour after fitting as the center of the circle of the corresponding reference area.

5. The intelligent industrial control method based on 5G technology according to claim 1, characterized in that, The consistency comparison between the simulated circuit diagram and the preset standard simulated circuit diagram includes: Extract the corner point features of the simulated circuit diagram and the corner point features of the preset standard simulated circuit respectively; Perform binary encoding on the corner point features of the simulated circuit diagram to obtain a comparison corner point feature vector, and perform binary encoding on the corner point features of the preset standard simulated circuit to obtain a reference corner point feature vector; Calculate the distance between the comparison corner point feature vector and the reference corner point feature vector; When the distance is less than or equal to a preset distance threshold, output the result information of consistent comparison; When the distance is not less than the preset distance threshold, output the result information of inconsistent comparison.

6. The intelligent industrial control method based on 5G technology according to any one of claims 1 to 5, characterized in that,Before using the pre-trained circuit image recognition model to recognize the components and circuit traces in the circuit image, the method further includes: Perform pixel correction on the circuit image using a flat field correction algorithm; Perform binarization processing on the image after pixel correction to obtain a binarized image; Perform hole area filling processing on the binarized image using a closing operation algorithm; and Perform edge smoothing processing on the binarized image using an opening operation algorithm.

7. An intelligent industrial control device based on 5G technology, characterized in that, The device includes: An image and identity information receiving module, configured to receive the circuit image of the circuit board to be detected and the identity association data of the circuit board to be detected sent by the automatic image acquisition terminal through the 5G communication module; An image recognition module, configured to perform vector encoding on the circuit image using the input layer of the pre-trained circuit image recognition model to obtain a vector matrix corresponding to the circuit image, perform convolution calculation on the vector matrix using the convolution layer of the circuit image recognition model, extract the component features and circuit trace features of the circuit image, match each component feature with a corresponding component label using a pre-trained activation function, label the corresponding components in the circuit image according to the component labels to obtain the components in the circuit image, label the circuit trace pixel points in the circuit image according to the circuit trace features, and connect the circuit trace pixel points to obtain the circuit traces in the circuit image; A simulated circuit diagram generation module, configured to calculate the position coordinate data of each component in the circuit image and the line coordinate data of the circuit traces in the circuit image, obtain the spatial layout relationship of each component and circuit trace in the circuit board to be detected, and generate a simulated circuit diagram of the circuit board to be detected according to the position coordinate data and the line coordinate data; A standard circuit diagram acquisition module, configured to acquire a preset standard analog circuit diagram corresponding to the identity association data of the circuit board to be detected; An industrial control module, configured to perform a consistency comparison between the analog circuit diagram and the preset standard analog circuit diagram to obtain a consistency comparison result. When the consistency comparison result is abnormal, a pause instruction is sent to a preset production line control terminal through a 5G communication module and an abnormal warning instruction is sent to the automatic image acquisition terminal.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can execute the intelligent industrial control method based on 5G technology according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent industrial control method based on 5G technology according to any one of claims 1 to 6.

Images