An industrial quality inspection method based on edge AI

Abstract

An embodiment of the present invention discloses an industrial quality inspection method based on edge AI, which relates to the field of edge computing technology. The method comprises: the cloud sends an AI quality inspection algorithm to a preset edge node device in a target factory; the edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result. The present invention sets the AI ​​quality inspection algorithm in the edge node device, and the edge node device automatically completes the quality inspection process. It has the advantages of high intelligent automation, strong scalability, high quality inspection efficiency, etc., and effectively saves labor costs.

Description

Technical Field

[0001] The present invention belongs to the field of edge computing technology, and in particular relates to an industrial quality inspection method based on edge AI. Background Art

[0002] In the fields of automobiles, 3C products, chips, photovoltaics, etc., industrial quality inspection is an indispensable link, which greatly affects the quality and speed of product output. Once a product quality problem occurs, it will bring great losses. At present, the industrial quality inspection method mainly relies on manual inspection. By deploying a large number of personnel on the production line, the product quality is inspected through experience and manpower. Although the quality inspection work can be completed and the product output quality is improved to a certain extent, there are also many shortcomings: First, the labor cost is high. Quality inspectors need to have certain professional inspection knowledge, so they need training. In addition, the assembly line quality inspection has high work intensity and high repetitiveness; second, the quality inspection is difficult. The complexity and refinement of industrial products are constantly improving, the size of parts is getting smaller, the styles and types are numerous, the shapes are irregular, and the replacement is frequent, which greatly increases the difficulty of quality inspection and control; third, the efficiency of manual quality inspection is low. With the development of AI technology, some industrial quality inspection methods based on AI technology have emerged, which have improved quality inspection efficiency and reduced labor costs to a certain extent. However, such quality inspection methods are currently deployed in a centralized manner, which is not convenient for large-scale promotion and deployment. In addition, all quality inspection work needs to be completed by remote servers, which leads to large quality inspection delays, further resulting in relatively low quality inspection efficiency. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides an industrial quality inspection method based on edge AI, which is used to solve the problems of high labor costs and low quality inspection efficiency in existing industrial quality inspection methods. The present invention can set the AI ​​quality inspection algorithm in the edge node device, and the edge node device automatically completes the quality inspection work, which has strong scalability, high quality inspection efficiency, and effectively saves labor costs.

[0004] An embodiment of the present invention provides an industrial quality inspection method based on edge AI, comprising:

[0005] The cloud sends the AI ​​quality inspection algorithm to the preset edge node devices in the target factory;

[0006] The edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result.

[0007] In an optional embodiment, the cloud sends the AI ​​quality inspection algorithm to a preset edge node device in the target factory, including:

[0008] The cloud uses different communication technologies to collaboratively transmit the AI ​​quality inspection algorithm to each preset edge node device in the target factory.

[0009] In an optional embodiment, the cloud collaboratively transmits the AI ​​quality inspection algorithm to each edge node device preset in the target factory through different communication technologies, including:

[0010] The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to different communication technologies;

[0011] The cloud uses different communication technologies to collaboratively transmit the encrypted data of the AI ​​quality inspection algorithms corresponding to different communication technologies to the preset edge node devices in the target factory.

[0012] In an optional embodiment, the different communication technologies include at least: 5G technology and 4G technology.

[0013] In an optional embodiment, the cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to different communication technologies, including:

[0014] The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm according to the first formula to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology;

[0015] Among them, the first formula is:

[0016]

[0017] In the first formula, J(5G) represents the encrypted data of the AI ​​quality inspection algorithm corresponding to 5G technology; J(4G) represents the encrypted data of the AI ​​quality inspection algorithm corresponding to 4G technology; D 16 Represents the hexadecimal form of the AI ​​quality inspection algorithm; Indicates that the data D 16 Each bit of data is XORed with 5 to get new hexadecimal data; Indicates that the data D 16 Each bit of data is XORed with 4 to get new hexadecimal data; Indicates circular left shift; Indicates circular right shift; len() indicates the number of bits of the data in the brackets; Indicates rounding up; Indicates rounding down; % indicates the remainder symbol;

[0018] The edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result, including:

[0019] The edge node device decrypts the received encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology according to the second formula to obtain the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology;

[0020] The edge node device performs AI intelligent detection on the local data to be detected according to the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology or the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology to obtain a quality inspection result;

[0021] Wherein, the second formula is:

[0022]

[0023] In the second formula, d(5G) represents the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology; d(4G) represents the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology; It means that each bit of data J(5G) is XORed with 5 to get new hexadecimal data; It means that each bit of data J(4G) is XORed with 4 to obtain new hexadecimal data.

[0024] In an optional embodiment, the cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies, including:

[0025] The cloud determines whether the remaining local CPU resources are below the preset threshold;

[0026] If the remaining local CPU resources are not less than the preset threshold, the cloud encrypts the AI ​​quality inspection algorithm according to the first formula using 5G technology and 4G technology respectively;

[0027] If the remaining local CPU resources are lower than the preset threshold, the cloud determines the priority order of using 5G technology and 4G technology to encrypt the AI ​​quality inspection algorithm based on the order in which the AI ​​quality inspection algorithm transmitted via 5G technology and 4G technology reaches the edge node device in the history that has been recorded;

[0028] The cloud encrypts the AI ​​quality inspection algorithm using the corresponding communication technology according to the first formula in accordance with the priority order of using the 5G technology and the 4G technology to encrypt the AI ​​quality inspection algorithm this time;

[0029] Before the edge node device decrypts the received encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology according to the second formula, the method further includes:

[0030] When the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology, the current time is recorded as the arrival time of the AI ​​quality inspection algorithm transmitted by the 5G technology this time, and when the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology, the current time is recorded as the arrival time of the AI ​​quality inspection algorithm transmitted by the 4G technology this time;

[0031] The edge node device determines the order in which the AI ​​quality inspection algorithms transmitted via 5G technology and 4G technology arrive at the edge node devices based on the arrival time of the AI ​​quality inspection algorithm transmitted via 5G technology and the arrival time of the AI ​​quality inspection algorithm transmitted via 4G technology, and sends the order in which the AI ​​quality inspection algorithms transmitted via 5G technology and 4G technology arrive at the edge node devices to the cloud for recording.

[0032] In an optional embodiment, the cloud determines the priority order of encrypting the AI ​​quality inspection algorithm using 5G technology and 4G technology this time according to the order in which the AI ​​quality inspection algorithm transmitted through 5G technology and 4G technology reaches the edge node device each time in the recorded history, including:

[0033] The cloud calculates the priority control value for encrypting the AI ​​quality inspection algorithm using 5G technology this time and the priority control value for encrypting the AI ​​quality inspection algorithm using 4G technology this time according to the third formula;

[0034] The cloud ranks the 5G technology and the 4G technology in descending order according to the priority control value for encrypting the AI ​​quality inspection algorithm, and obtains the priority order for encrypting the AI ​​quality inspection algorithm using the 5G technology and the 4G technology this time;

[0035] Wherein, the third formula is:

[0036]

[0037] In the third formula, Y 5G (n+1) represents the priority control value for encrypting the AI ​​quality inspection algorithm using 5G technology for the n+1th time; Y 4G(n+1) represents the priority control value for encrypting the AI ​​quality inspection algorithm using 4G technology for the n+1th time; n represents the total number of times that the cloud has transmitted the AI ​​quality inspection algorithm to the edge node device through 5G technology and 4G technology in history up to the current moment; a=1,2,…,n; B(a) represents the order value of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 5G technology and 4G technology to the edge node device in history. If the arrival time of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 5G technology in history is earlier than the arrival time of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 4G technology in history, then B(a)=1, otherwise B(a)=-1; Z[] represents a non-negative judgment function. If the value in the brackets is a non-negative number, the function value is 1, otherwise the function value is 0; F[] represents a negative judgment function. If the value in the brackets is a negative number, the function value is 1, otherwise the function value is 0.

[0038] The present invention provides an industrial quality inspection method based on edge AI, which first downloads the AI ​​quality inspection algorithm in the cloud to the target edge node device, and then the edge node device automatically performs industrial quality inspection according to the AI ​​quality inspection algorithm to obtain the quality inspection result. The present invention sinks the AI ​​quality inspection algorithm from the cloud to the edge node device at the edge position close to the user, realizes localized data processing and AI intelligent detection, and has the advantages of strong scalability, high quality inspection efficiency, and effective labor cost savings. 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 description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A flow chart of an industrial quality inspection method based on edge AI provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0042] It should be clear that the described embodiments are only some embodiments of the present invention, not all 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.

[0043] Figure 1 A flow chart of an industrial quality inspection method based on edge AI provided by an embodiment of the present invention. Figure 1The method comprises the following steps S101-S102:

[0044] S101: The cloud sends the AI ​​quality inspection algorithm to the preset edge node devices in the target factory.

[0045] S102: The edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result.

[0046] The beneficial effects of the above technical solution are as follows: an industrial quality inspection method based on edge AI provided by an embodiment of the present invention first downloads the AI ​​quality inspection algorithm in the cloud to the target edge node device, and then the edge node device automatically performs industrial quality inspection according to the AI ​​quality inspection algorithm to obtain the quality inspection result. The present invention sinks the AI ​​quality inspection algorithm from the cloud to the edge location close to the user, realizes localized data processing and AI intelligent detection, and has the advantages of strong scalability, high quality inspection efficiency, and effective savings in labor costs.

[0047] As an optional embodiment, the step S101 may include: the cloud collaboratively transmits the AI ​​quality inspection algorithm to each edge node device preset in the target factory through different communication technologies.

[0048] Preferably, the different communication technologies include at least: 5G technology and 4G technology.

[0049] The beneficial effects of the above technical solution are: using different communication technologies to transmit AI quality inspection algorithm data, such as 5G network, which has a higher bandwidth than 4G network, can effectively improve the efficiency of data transmission; although the 4G network is not as fast as the 5G network transmission speed, the 4G network is currently widely deployed and can be used as a supplement to the 5G network transmission. When the network fails, the present invention can still work normally, so that the AI ​​quality inspection algorithm data transmission has higher reliability.

[0050] As an optional embodiment, the cloud collaboratively transmits the AI ​​quality inspection algorithm to each edge node device preset in the target factory through different communication technologies, which may include the following steps S201-S202:

[0051] S201: The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to the different communication technologies;

[0052] S202: The cloud uses different communication technologies to collaboratively transmit the encrypted data of the AI ​​quality inspection algorithms corresponding to different communication technologies to the preset edge node devices in the target factory.

[0053] The beneficial effect of the above technical solution is: according to different communication technologies, the encryption algorithm data of the corresponding technology is obtained, so that different encryption algorithms can more effectively protect the AI ​​quality inspection algorithm data and ensure the security of the data.

[0054] As an optional embodiment, the step S201 may include: the cloud performs encryption processing on the AI ​​quality inspection algorithm in 5G and 4G forms according to the first formula to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology;

[0055] Among them, the first formula is:

[0056]

[0057] In the first formula, J(5G) represents the encrypted data of the AI ​​quality inspection algorithm corresponding to 5G technology; J(4G) represents the encrypted data of the AI ​​quality inspection algorithm corresponding to 4G technology; D 16 Represents the hexadecimal form of the AI ​​quality inspection algorithm; Indicates that the data D 16 Each bit of data is XORed with 5 to get new hexadecimal data; Indicates that the data D 16 Each bit of data is XORed with 4 to get new hexadecimal data; Indicates circular left shift; Indicates circular right shift; len() indicates the number of bits of the data in the brackets; Indicates rounding up; Indicates rounding down; % indicates the remainder symbol.

[0058] Preferably, the step S102 may include the following steps S301-S302:

[0059] S301: The edge node device decrypts the received encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology according to the second formula to obtain the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology;

[0060] S302: The edge node device performs AI intelligent detection on the local data to be detected according to the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology or the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology to obtain a quality inspection result;

[0061] Wherein, the second formula is:

[0062]

[0063] In the second formula, d(5G) represents the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology; d(4G) represents the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology; It means that each bit of data J(5G) is XORed with 5 to get new hexadecimal data; It means that each bit of data J(4G) is XORed with 4 to obtain new hexadecimal data.

[0064] The beneficial effects of the above technical solution are: using the first formula (1) to perform different forms of encryption processing on the AI ​​algorithm according to 5G technology and 4G technology, to obtain encryption algorithm data of the corresponding technology, so that different encryption algorithms can more effectively protect data and ensure data security; then using the second formula (2) according to the transmission technology of the encryption algorithm data to decrypt the encryption algorithm data, ensure that the data can be automatically and reliably decrypted, and ensure the reliability of the data and the accuracy after decryption.

[0065] As an optional embodiment, the step S201 may include the following steps S401-S404:

[0066] S401: The cloud determines whether the remaining local CPU resources are lower than a preset threshold; if yes, execute S402; otherwise, execute S403.

[0067] S402: The cloud encrypts the AI ​​quality inspection algorithm according to the first formula using 5G technology and 4G technology to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology;

[0068] In this embodiment, when the remaining local CPU resources in the cloud are sufficient, the AI ​​quality inspection algorithm is encrypted using 5G technology and 4G technology respectively.

[0069] S403: The cloud determines the priority order of encrypting the AI ​​quality inspection algorithm using the 5G technology and the 4G technology this time according to the order in which the AI ​​quality inspection algorithm transmitted through the 5G technology and the 4G technology reaches the edge node device in the recorded history;

[0070] For example, if in most cases in history, the AI ​​quality inspection algorithm transmitted via 5G technology reaches the edge node device before the AI ​​quality inspection algorithm transmitted via 4G technology, then the priority order for using 5G technology and 4G technology to encrypt the AI ​​quality inspection algorithm this time is determined to be that 5G technology has a higher priority than 4G technology, that is: when the remaining local CPU resources in the cloud are tight, 5G technology is used first to encrypt and transmit the AI ​​quality inspection algorithm, and then 4G technology is used to encrypt and transmit the AI ​​quality inspection algorithm.

[0071] S404: The cloud encrypts the AI ​​quality inspection algorithm according to the priority order of using 5G technology and 4G technology to encrypt the AI ​​quality inspection algorithm, and uses the corresponding communication technology according to the first formula to obtain the encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology.

[0072] Preferably, before step S301, steps S501-S502 may also be included:

[0073] S501: When the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology, the edge node device records the current time as the arrival time of the AI ​​quality inspection algorithm transmitted through the 5G technology this time, and when the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology, the edge node device records the current time as the arrival time of the AI ​​quality inspection algorithm transmitted through the 4G technology this time;

[0074] S502: The edge node device determines the order in which the AI ​​quality inspection algorithms transmitted this time via 5G technology and 4G technology arrive at the edge node devices according to the arrival time of the AI ​​quality inspection algorithms transmitted this time via 5G technology and the arrival time of the AI ​​quality inspection algorithms transmitted this time via 4G technology, and sends the order in which the AI ​​quality inspection algorithms transmitted this time via 5G technology and 4G technology arrive at the edge node devices to the cloud for recording.

[0075] For example: If the time when the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to 5G technology is 13:00, then the arrival time of the AI ​​quality inspection algorithm transmitted through 5G technology is recorded as 13:00. Similarly, if the arrival time of the AI ​​quality inspection algorithm transmitted through 4G technology is recorded as 13:05, the order in which the AI ​​quality inspection algorithms transmitted through 5G technology and 4G technology arrive at the edge node device is determined as follows: 5G technology precedes 4G technology.

[0076] The beneficial effect of the above technical solution is: when the cloud CPU is too busy to encrypt data from 5G technology and 4G technology at the same time, the priority order of using 5G technology and 4G technology to encrypt the AI ​​quality inspection algorithm is determined based on the historical time when the AI ​​algorithm data corresponding to 5G technology and 4G technology is downloaded from the cloud to the edge location (and edge node device) close to the user, so as to ensure the data transmission efficiency when the cloud CPU is relatively busy.

[0077] As an optional embodiment, the step S403 may include the following steps S601-S602:

[0078] S601: The cloud calculates the priority control value for encrypting the AI ​​quality inspection algorithm using the 5G technology this time and the priority control value for encrypting the AI ​​quality inspection algorithm using the 4G technology this time according to the third formula;

[0079] S602: The cloud ranks the 5G technology and the 4G technology in descending order according to the priority control value for encrypting the AI ​​quality inspection algorithm, and obtains the priority order for using the 5G technology and the 4G technology to encrypt the AI ​​quality inspection algorithm this time;

[0080] Wherein, the third formula is:

[0081]

[0082] In the third formula, Y 5G (n+1) represents the priority control value for encrypting the AI ​​quality inspection algorithm using 5G technology for the n+1th time; Y 4G (n+1) represents the priority control value for encrypting the AI ​​quality inspection algorithm using 4G technology for the n+1th time. If Y 5G (n+1) is greater than Y 4G (n+1) indicates that the cloud will use 5G technology to encrypt the AI ​​quality inspection algorithm first. 5G (n+1) is less than Y 4G (n+1) indicates that the cloud will use 4G technology to encrypt the AI ​​quality inspection algorithm first. 5G (n+1) equals Y 4GWhen (n+1), it indicates that the cloud should select either 5G technology or 4G technology to encrypt the AI ​​quality inspection algorithm; n indicates the total number of times that the cloud has transmitted the AI ​​quality inspection algorithm to the edge node device through 5G technology and 4G technology in history up to the current moment; a=1,2,…,n; B(a) indicates the order value of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 5G technology and 4G technology to the edge node device in history. If the arrival time of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 5G technology in history is earlier than the arrival time of the AI ​​quality inspection algorithm transmitted by the cloud for the ath time through 4G technology in history, then B(a)=1, otherwise B(a)=-1; Z[] indicates a non-negative judgment function. If the value in the brackets is a non-negative number, the function value is 1, otherwise the function value is 0; F[] indicates a negative judgment function. If the value in the brackets is a negative number, the function value is 1, otherwise the function value is 0.

[0083] The beneficial effect of the above technical solution is: using the third formula (3) according to the arrival order of the 5G technology and the 4G technology when the historical AI algorithm is sunk from the cloud to the edge position close to the user, the encryption algorithm priority of the 5G technology and the 4G technology when the AI ​​algorithm is sunk from the cloud to the edge position close to the user is controlled next time, thereby maintaining the transmission advantage of the 5G technology and the 4G technology and ensuring efficient data transmission.

[0084] It can be seen from the contents of the above embodiments that the edge AI-based industrial quality inspection method proposed in the embodiments of the present invention sinks AI from the cloud to the edge location close to the user, realizes localized data processing and AI intelligent detection, and combines 5G technology to reduce detection latency, has high real-time performance, and has extremely strong security. When the network fails, it can still work normally with high reliability, and realizes automatic learning and uninterrupted parallel detection to reduce labor costs. The present invention also uses 5G technology and 4G technology to collaboratively transmit data to sink the AI ​​algorithm from the cloud to the edge location close to the user, and 5G technology and 4G technology adopt different data transmission forms for transmission to ensure data security.

[0085] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1An apparatus for the method specified in one or more blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The method specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of a method specified in one or more boxes.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these changes and variations. The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An industrial quality inspection method based on edge AI, characterized in that: include: The cloud sends the AI ​​quality inspection algorithm to the preset edge node devices in the target factory; The edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result; The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to different communication technologies, including: The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm according to the first formula to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology; Among them, the first formula is: In the first formula, Indicates the encrypted data of the AI ​​quality inspection algorithm corresponding to 5G technology; Indicates the encrypted data of the AI ​​quality inspection algorithm corresponding to 4G technology; Represents the hexadecimal form of the AI ​​quality inspection algorithm; Indicates that the data Each bit of data is XORed with 5 to get new hexadecimal data; Indicates that the data Each bit of data is XORed with 4 to get new hexadecimal data; Indicates circular left shift; Indicates circular right shift; Indicates the number of digits to be obtained for the data in brackets; Indicates rounding up; Indicates rounding down; Indicates the remainder symbol; The different communication technologies include at least: 5G technology and 4G technology; The edge node device performs AI intelligent detection on the local data to be detected according to the received AI quality inspection algorithm to obtain a quality inspection result, including: The edge node device decrypts the received encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology according to the second formula to obtain the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology; The edge node device performs AI intelligent detection on the local data to be detected according to the decrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology or the decrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology to obtain a quality inspection result; Wherein, the second formula is: In the second formula, Indicates the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to 5G technology; Indicates the hexadecimal form of the decrypted data of the AI ​​quality inspection algorithm corresponding to 4G technology; It means that each bit of the data is XORed with 5 to get a new hexadecimal data. It means that each bit of the data is XORed with 4 to get new hexadecimal data.

2. The industrial quality inspection method based on edge AI according to claim 1, characterized in that: The cloud sends the AI ​​quality inspection algorithm to the preset edge node devices in the target factory, including: The cloud uses different communication technologies to collaboratively transmit the AI ​​quality inspection algorithm to each preset edge node device in the target factory.

3. The edge AI-based industrial quality inspection method according to claim 2, characterized in that: The cloud collaboratively transmits the AI ​​quality inspection algorithm to each edge node device preset in the target factory through different communication technologies, including: The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies to obtain encrypted data of the AI ​​quality inspection algorithm corresponding to different communication technologies; The cloud uses different communication technologies to collaboratively transmit the encrypted data of the AI ​​quality inspection algorithms corresponding to different communication technologies to the preset edge node devices in the target factory.

4. The industrial quality inspection method based on edge AI according to claim 1, characterized in that: The cloud performs different forms of encryption processing on the AI ​​quality inspection algorithm through different communication technologies, including: The cloud determines whether the remaining local CPU resources are below the preset threshold; If the remaining local CPU resources are not less than the preset threshold, the cloud encrypts the AI ​​quality inspection algorithm according to the first formula using 5G technology and 4G technology respectively; If the remaining local CPU resources are lower than the preset threshold, the cloud determines the priority order of using 5G technology and 4G technology to encrypt the AI ​​quality inspection algorithm based on the order in which the AI ​​quality inspection algorithm transmitted via 5G technology and 4G technology reaches the edge node device in the history that has been recorded; The cloud encrypts the AI ​​quality inspection algorithm using the corresponding communication technology according to the first formula in accordance with the priority order of using the 5G technology and the 4G technology to encrypt the AI ​​quality inspection algorithm this time; Before the edge node device decrypts the received encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology and the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology according to the second formula, the method further includes: When the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 5G technology, the current time is recorded as the arrival time of the AI ​​quality inspection algorithm transmitted by the 5G technology this time, and when the edge node device receives the encrypted data of the AI ​​quality inspection algorithm corresponding to the 4G technology, the current time is recorded as the arrival time of the AI ​​quality inspection algorithm transmitted by the 4G technology this time; The edge node device determines the order in which the AI ​​quality inspection algorithms transmitted via 5G technology and 4G technology arrive at the edge node devices based on the arrival time of the AI ​​quality inspection algorithm transmitted via 5G technology and the arrival time of the AI ​​quality inspection algorithm transmitted via 4G technology, and sends the order in which the AI ​​quality inspection algorithms transmitted via 5G technology and 4G technology arrive at the edge node devices to the cloud for recording.

5. The edge AI-based industrial quality inspection method according to claim 1, characterized in that: The cloud determines the priority order of encrypting the AI ​​quality inspection algorithm using 5G technology and 4G technology this time according to the order in which the AI ​​quality inspection algorithm transmitted through 5G technology and 4G technology reaches the edge node device each time in the recorded history, including: The cloud calculates the priority control value for encrypting the AI ​​quality inspection algorithm using 5G technology this time and the priority control value for encrypting the AI ​​quality inspection algorithm using 4G technology this time according to the third formula; The cloud ranks the 5G technology and the 4G technology in descending order according to the priority control value for encrypting the AI ​​quality inspection algorithm, and obtains the priority order for encrypting the AI ​​quality inspection algorithm using the 5G technology and the 4G technology this time; Wherein, the third formula is: In the third formula, Indicates The priority control value of the AI ​​quality inspection algorithm encrypted using 5G technology; Indicates The priority control value of the AI ​​quality inspection algorithm encrypted using 4G technology; Indicates the total number of times that the cloud has sent the AI ​​quality inspection algorithm to edge node devices through 5G and 4G technologies up to the current moment; ; Indicates the history of cloud The order of the AI ​​quality inspection algorithm transmitted by 5G technology and 4G technology to the edge node device is the order of the first The first AI quality inspection algorithm transmitted via 5G technology arrived earlier than the first in history. The arrival time of the AI ​​quality inspection algorithm transmitted via 4G technology is , otherwise ; Represents a non-negative judgment function. If the value in the brackets is non-negative, the function value is 1, otherwise the function value is 0; Represents a negative number judgment function. If the value in the brackets is a negative number, the function value is 1, otherwise the function value is 0.

Images