Granted Patent B2

US 12,412,366 · App. 18/498,112 · Granted Sep 9, 2025

Inventors: Zehua Shao (Chengdu, CN); Bin Liu (Chengdu, CN); Yuefei Wu (Chengdu, CN); Yaqiang Quan (Chengdu, CN); Junyan Zhou (Chengdu, CN)

Assignee: CHENGDU QINCHUAN IOT TECHNOLOGY CO., LTD.

G06V10/761G06F21/602G06V10/40G06V10/764G16Y10/25G16Y30/10H04L67/12G06V2201/06

Abstract

Claims (89)

1. An Industrial Internet of Things (IIoT) system for industrial production data management, comprising:

a non-transitory computer-readable storage medium storing executable instructions; and

at least one processor in communication with the non-transitory computer-readable storage medium, when executing the executable instructions, the at least one processor is directed to cause the IIoT system to:

identify first image data and second image data based on image data of a production line;

encrypt the second image data to form third image data;

decrypt the third image data to form decrypted second image data;

obtain reference image data;

determine image similarities through image feature processing based on the decrypted second image data and the reference image data, wherein the image similarities include first image similarities and second image similarities, and the second image similarities are determined by a judgment model based on the decrypted second image data and the reference image data, wherein the judgment model is a machine learning model; and

obtain target images based on the image similarities and send the target images to a user platform.

2. The IIoT system for industrial production data management of claim 1 , wherein the at least one processor is further directed to cause the IIoT system to:

extract first features corresponding to the decrypted second image data based on the decrypted second image data;

extract second features corresponding to the reference image data based on the reference image data; and

determine the first image similarities based on the first features and the second features.

3. The IIoT system for industrial production data management of claim 1 , wherein the judgment model includes a feature extraction layer and a similarity judgment layer;

the feature extraction layer is configured to process the decrypted second image data and the reference image data to determine a third feature vector corresponding to the decrypted second image data and a fourth feature vector corresponding to the reference image data; and

the similarity judgment layer is configured to process the third feature vector and the fourth feature vector to determine the second image similarities.

4. The IIoT system for industrial production data management of claim 1 , wherein the judgment model is obtained through training based on a plurality of groups of training data and labels, wherein each group of the training data in the plurality of groups of training data includes two sample image data, and label of the group of the training data is a similarity of the two sample image data, wherein the at least one processor is further directed to cause the IIoT system to:

determine the plurality of groups of training data based on a process map, wherein nodes of the process map include image nodes, feature nodes, and feature type nodes, and edges of the process map are used to connect two associated nodes, wherein the image nodes correspond to historical process images, node attributes of the feature nodes include process parameters or product parameters, and node attributes of the feature type nodes include types of process parameters or product parameters.

5. The IIoT system for industrial production data management of claim 4 , wherein the at least one processor is further directed to cause the IIoT system to:

obtain common feature nodes and difference feature nodes of two historical process images based on the process map;

determine a first number of the common feature nodes based on the common feature nodes, and determine a second number of the difference feature nodes belonging to a same feature type in the difference feature nodes based on the difference feature nodes;

determine a first judgment result based on the first number and a first preset threshold;

determine a second judgment result based on the second number and a second preset threshold; and

determine the plurality of groups of training data based on the first judgment result and the second judgment result.

6. The IIoT system for industrial production data management of claim 1 , wherein the at least one processor is further directed to cause the IIoT system to:

obtain corresponding decrypted second image data whose a similarity with the reference image data meets a preset condition, wherein the preset condition includes the similarity being maximum or the similarity being higher a preset threshold; and

take images of the corresponding decrypted second image data as the target images.

7. The IIoT system for industrial production data management of claim 1 , further comprising: a service platform, a management platform, and a sensor network platform that are interacted sequentially from top to bottom, wherein the sensor network platform includes a general platform of the sensor network platform and a plurality of sub platforms of the sensor network platform, and the service platform includes a general platform of the service platform and a plurality of sub platforms of the service platform, wherein the at least one processor is further directed to cause the IIoT system to:

receive the image data of the production line detected by sensors of the production line, and identify the image data of the production line as the first image data and the second image data, wherein the first image data is image data that requires the management platform to perform identification processing, and the second image data is image data that requires to be encrypted and transmitted to the service platform;

encrypt and store the second image data into the first image data to form the third image data, and send the third image data to corresponding sub platforms of the sensor network platform according to types of the third image data;

forward the received third image data to the management platform;

identify the third image data to generate control parameters, and send the control parameters to the production line through the sensor network platform; and

send the third image data to the service platform;

receive the third image data and decrypt the corresponding second image data from the third image data; and

summarize all the second image data and present the second image data to a user platform.

8. The IIoT system for industrial production data management of claim 7 , wherein

the second image data and the first image data are both plural, the general platform of the sensor network platform is configured with a plurality of serial numbers of the first image data, and the general platform of the sensor network platform is configured with encryption keys corresponding to different second image data;

when encrypting and storing the second image data into the first image data to form the third image data, the at least one processor is further directed to cause the IIoT system to encrypt and store the second image data into the first image data according to the encryption keys corresponding to the second image data, and assign the serial numbers corresponding to the first image data to the corresponding third image data; and

classify the third image data according to the serial numbers of the third image data and send the third image data to the corresponding sub platforms of the sensor network platform according to the types of the third image data.

9. The IIoT system for industrial production data management of claim 8 , wherein the sub platforms of the service platform are configured with decryption keys corresponding to the different second image data; and

when receiving the third image data, the at least one processor is further directed to cause the IIoT system to identify the third image data according to the serial numbers of the third image data, and extract the second image data from the identified third image data according to the decryption keys.

10. A control method of an Industrial Internet of Things (IIoT) system for industrial production data management, wherein the method is executed by at least one processor of the IIoT system, comprising:

identifying first image data and second image data based on image data of a production line;

encrypting the second image data to form third image data;

decrypting the third image data to form decrypted second image data;

searching one or more target images in the decrypted second image data;

obtaining reference image data;

determining image similarities through image feature processing based on the decrypted second image data and the reference image data, wherein the image similarities include first image similarities and second image similarities, and the second image similarities are determined by a judgment model based on the decrypted second image data and the reference image data, wherein the judgment model is a machine learning model; and

obtaining target images based on the image similarities and send the target images to a user platform.

11. The control method of the IIoT system for industrial production data management of claim 10 , wherein the first image similarities are determined by a process including:

extracting first features corresponding to the decrypted second image data based on the decrypted second image data;

extracting second features corresponding to the reference image data based on the reference image data; and

determining the first image similarities based on the first features and the second features.

12. The control method of the IIoT system for industrial production data management of claim 10 , wherein the judgment model includes a feature extraction layer and a similarity judgment layer;

the feature extraction layer is configured to process the decrypted second image data and the reference image data to determine a third feature vector corresponding to the decrypted second image data and a fourth feature vector corresponding to the reference image data; and

the similarity judgment layer is configured to process the third feature vector and the fourth feature vector to determine the second image similarities.

13. The control method of the IIoT system for industrial production data management of claim 10 , wherein the judgment model is obtained through training based on a plurality of groups of training data and labels, wherein each group of the training data in the plurality of groups of training data includes two sample image data, and label of the group of the training data is a similarity of the two sample image data, further comprising:

determining the plurality of groups of training data based on a process map, wherein nodes of the process map include image nodes, feature nodes, and feature type nodes, and edges of the process map are used to connect two associated nodes, wherein the image nodes correspond to historical process images, node attributes of the feature nodes include process parameters or product parameters, and node attributes of the feature type nodes include types of process parameters or product parameters.

14. The control method of the IIoT system for industrial production data management of claim 13 , wherein the determining the plurality of groups of training data based on a process map includes:

obtaining common feature nodes and difference feature nodes of two historical process images based on the process map;

determining a first number of the common feature nodes based on the common feature nodes, and determining a second number of the difference feature nodes belonging to a same feature type in the difference feature nodes based on the difference feature nodes;

determining a first judgment result based on the first number and a first preset threshold;

determining a second judgment result based on the second number and a second preset threshold; and

determining the plurality of groups of training data based on the first judgment result and the second judgment result.

15. The control method of the IIoT system for industrial production data management of claim 10 , further comprising:

obtaining corresponding decrypted second image data whose a similarity with the reference image data meets a preset condition, wherein the preset condition includes the similarity being maximum or the similarity being higher a preset threshold; and

taking images of the corresponding decrypted second image data as the target images.

16. The control method of the IIoT system for industrial production data management of claim 10 , wherein the method is applied to a service platform, a management platform, and a sensor network platform that are interacted sequentially from top to bottom, the sensor network platform includes a general platform of the sensor network platform and a plurality of sub platforms of the sensor network platform; and the service platform includes a general platform of the service platform and a plurality of sub platforms of the service platform; and

the method comprises:

receiving, by the general platform of the sensor network platform, the image data of the production line detected by sensors of the production line, and identifying the image data of the production line as the first image data and the second image data, wherein the first image data is image data that requires the management platform to perform identification processing, and the second image data is image data that requires to be encrypted and transmitted to the service platform;

encrypting and storing, by the general platform of the sensor network platform, the second image data into the first image data to form the third image data, and sending the third image data to corresponding sub platforms of the sensor network according to types of the third image data;

forwarding, by the sub platforms of the sensor network platform, the received third image data to the management platform;

identifying, by the management platform, the third image data to generate control parameters, and sending the control parameters to the production line through the sensor network platform;

sending, by the management platform, the third image data to the service platform;

receiving, by the sub platforms of the service platform, the third image data and decrypting the corresponding second image data from the third image data; and

summarizing, by the general platform of the sensor network platform, all the second image data and presenting the second image data to a user platform.

17. The control method of the IIoT system for industrial production data management of claim 16 , wherein

the second image data and the first image data are both plural, the general platform of the sensor network platform is configured with a plurality of serial numbers of the first image data, and the general platform of the sensor network platform is configured with encryption keys corresponding to different second image data;

when encrypting and storing the second image data into the first image data to form the third image data, encrypting and storing, by the general platform of the sensor network platform, the second image data into the first image data according to the encryption keys corresponding to the second image data, and assigning the serial numbers corresponding to the first image data to the corresponding third image data; and

classifying, by the general platform of the sensor network platform, the third image data according to the serial numbers of the third image data and sending the third image data to the corresponding sub platforms of the sensor network platform according to the types of the third image data.

18. The control method of the IIoT system for industrial production data management of claim 17 , wherein the sub platforms of the service platform are configured with decryption keys corresponding to the different second image data; and

when receiving the third image data, identifying, by the sub platforms of the service platform, the third image data according to the serial numbers of the third image data, and extracting the second image data from the identified third image data according to the decryption keys.

19. A non-transitory computer-readable storage medium, wherein the storage medium stores computer instructions, and when the computer instructions are executed by a processor, the processor is directed to cause an Industrial Internet of Things (IIoT) system to:

identify first image data and second image data based on image data of a production line;

encrypt the second image data to form third image data;

decrypt the third image data to form decrypted second image data;

obtain reference image data;

determine image similarities through image feature processing based on the decrypted second image data and the reference image data, wherein the image similarities include first image similarities and second image similarities, and the second image similarities are determined by a judgment model based on the decrypted second image data and the reference image data, wherein the judgment model is a machine learning model; and

obtain target images based on the image similarities and send the target images to a user platform.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (20)