Automatic monitoring device of automatic measurement and control equipment

Abstract

The invention discloses an automatic monitoring device for automatic measurement and control equipment, and relates to the field of equipment monitoring, and the device comprises a management module which serves as a carrying end of each functional module and carries out the start-stop control of each piece of measurement and control equipment and each functional module; the image acquisition module is used for generally controlling a plurality of acquisition components and monitoring measurement and control equipment and products on a measurement and control production line in real time through the acquisition components; the area planning module is used for carrying out dynamic area division after carrying out key area intelligent identification on different acquired images in the rotating acquisition period of the acquisition part; the feature extraction module is used for extracting feature points, edges and texture feature data from the planning area through an image processing algorithm; by using the deep learning technology, the monitoring area and the feature extraction algorithm can be continuously optimized, the recognition accuracy and efficiency can be improved, areas with large changes are preferentially concerned on the basis of real-time sensitivity sorting, the monitoring area is dynamically adjusted, and it is ensured that the high-sensitivity area is continuously concerned.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and particularly to an automatic monitoring device for an automated measurement and control equipment. Background Art

[0002] Measurement and control equipment tests whether products or semi-finished products are qualified at a certain stage through certain detection means. Different types of measurement and control equipment constitute a measurement and control production line. Due to the variety of test equipment and the cumbersome actions of the test equipment in the measurement and control production line, the processing actions implemented on the products and the corresponding processing effects shown are different in different processing processes;

[0003] Traditional technologies often judge whether there are abnormalities in the equipment by combining monitoring screens and sensors. When the sensors also fail and cannot identify abnormalities, the management personnel can only analyze the abnormal content through the monitoring screen. However, even within a fixed monitoring screen, there are too many information elements to be identified. Therefore, the capture degree of the overall monitoring information is insufficient, and it is easy to miss the abnormal information of a certain equipment in the process, resulting in potential abnormal factors not being accurately and real-time identified.

[0004] Therefore, an automatic monitoring device for an automated measurement and control equipment is needed to solve the above problems. Summary of the Invention

[0005] (1) Technical Problems to be Solved

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides an automatic monitoring device for an automated measurement and control equipment, which can effectively solve the problems of the existing technology.

[0007] (2) Technical Solutions

[0008] To achieve the above object, the present invention is realized through the following technical solutions:

[0009] The present invention discloses an automatic monitoring device for an automated measurement and control equipment, including:

[0010] A management module, as the hosting end of each functional module, controls the start and stop of each measurement and control equipment and each functional module;

[0011] An image acquisition module, used to overall control a number of acquisition components, and to monitor the measurement and control equipment and products on the measurement and control production line in real time through the acquisition components;

[0012] A region planning module, used to perform intelligent identification of key regions on different acquisition images during the rotation acquisition cycle of the acquisition components, and then implement dynamic region division;

[0013] A feature extraction module, used to extract feature points, edges, and texture feature data from the planned region through image processing algorithms;

[0014] A judgment unit, configured to compare the acquired feature data with predefined standard data to determine whether there is an abnormality;

[0015] An anomaly analysis module, configured to analyze the submitted anomaly, identify the anomaly type and cause, and output relevant anomaly information and the involved abnormal devices;

[0016] A marking module, configured to mark a number of detected abnormal devices and a number of associated acquisition components;

[0017] An angle adjustment module, configured to analyze relevant anomaly information and the involved abnormal devices, generate and execute adjustment instructions for the monitoring angles of a number of marked associated acquisition components, obtain image states of the abnormal devices at different angles within a preset period, and resubmit them to the judgment unit for secondary analysis to finally determine whether there is an abnormality.

[0018] Furthermore, the area planning module uses historical data as samples, constructs a state model through a deep learning algorithm, obtains acquisition images in real time, performs preprocessing, analyzes the real-time data through the state model, divides a number of areas of the acquisition images, dynamically adjusts the areas to be monitored according to state changes, performs real-time sensitivity ranking on the identified areas, and preferentially focuses on high-sensitivity areas according to the historical adjustment parameters of the angle adjustment module.

[0019] Furthermore, the judgment unit includes a template database, an action definition module, and a scan definition module. The template database is configured to compare the device actions involved in the acquired image content with a standard template to determine whether the device operating state meets the expectation. The action definition module is configured to compare the acquired product state with a preset standard module to determine whether there is an abnormality in the product. The scan definition module establishes a refined action standard template library and a standard module library, covering different actions of different devices and different angular postures of different products.

[0020] Furthermore, the judgment unit analyzes the matching degree between the acquired data and the standard data. If the matching degree is lower than a preset threshold, it is determined that the current acquired data and the standard data are normal; otherwise, it is determined that there is an abnormality and it is immediately submitted to the angle adjustment module.

[0021] Furthermore, the feature extraction module synchronously upgrades the feature extraction algorithm according to the update periods of the measurement and control devices and the acquisition components by using a deep learning algorithm.

[0022] Furthermore, the anomaly analysis module sets reasonable thresholds based on historical data, quickly screens the real-time data, classifies the identified anomalies, and identifies the credibility of the anomaly results through a combination of data simulation and historical data verification, generating a detailed anomaly report including anomaly type, cause, occurrence time, and affected device information.

[0023] Furthermore, when the angle adjustment module finally determines an anomaly, it submits an alarm message to the management module, focuses on the associated acquisition components, continuously acquires images, and continuously submits them to the management end.

[0024] Furthermore, the angle adjustment module is connected to a memory module through wireless network interaction. The memory module is used to record the historical adjustment data of the angle adjustment module and classify it into several adjustment modes. When encountering similar anomalies in the future, it directly calls the historical adjustment mode with a matching degree to the angle adjustment module for implementation.

[0025] Furthermore, the attributes of the adjustment mode include: the number of target acquisition components, rotation angle, rotation order, and duration.

[0026] Furthermore, the management module is connected to the image acquisition module through wireless network interaction. The image acquisition module is connected to the area planning module and the angle adjustment module through wireless network interaction. The area planning module is connected to the feature extraction module through wireless network interaction. The feature extraction module is connected to the judgment unit through wireless network interaction. The judgment unit is connected to the anomaly analysis module through wireless network interaction. The anomaly analysis module is connected to the marking module through wireless network interaction. The marking module is connected to the angle adjustment module through wireless network interaction.

[0027] (III) Beneficial Effects

[0028] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects.

[0029] 1. Through the master control acquisition component and the intelligent recognition algorithm, dynamically obtain and analyze images, reduce the need for manual intervention, construct a state model based on historical data, design a dynamic area division and priority attention mechanism, so as to utilize deep learning technology, continuously optimize the monitoring area and the feature extraction algorithm, improve the accuracy and efficiency of recognition, and based on the real-time sensitivity ranking, give priority to areas with large changes, dynamically adjust the monitoring area, and ensure that high-sensitivity areas are continuously concerned, thereby improving the sensitivity of anomaly detection.

[0030] 2. By comprehensively analyzing abnormal data, the credibility of abnormal results is improved, misjudgment is reduced, and a refined action standard template library is established. Screening and classification are carried out on the reasonable thresholds obtained from historical data and simulation data, and a detailed abnormal report is generated, making abnormal judgment more reliable. By recording and classifying historical data adjustments, when encountering similar situations, the matching historical adjustment mode can be quickly called to quickly respond to similar abnormalities and improve the response speed. The invocation of the historical adjustment mode effectively reduces the uncertainty brought by human decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0032] Figure 1 is a schematic framework diagram of the present invention;

[0033] Figure 2 is a test schematic diagram of video transmission acquisition in the present invention.

[0034] The reference numerals in the figure respectively represent: 1, management module; 2, image acquisition module; 3, area planning module; 4, feature extraction module; 5, judgment unit; 51, template database; 52, action definition module; 53, scanning definition module; 6, abnormal analysis module; 7, marking module; 8, angle adjustment module; 9, memory module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0036] The following further describes the present invention with reference to the embodiments.

[0037] Embodiment 1

[0038] An automatic monitoring device for an automatic measurement and control device in this embodiment, as Figure 1 shown, includes:

[0039] The management module 1, as the carrier of each functional module, controls the start and stop of each measurement and control device and each functional module;

[0040] The image acquisition module 2 is used to overall control a number of acquisition components, and through the acquisition components, it monitors the measurement and control equipment and products on the measurement and control production line in real time;

[0041] The area planning module 3 is used to intelligently identify key areas of different acquired images during the rotation acquisition cycle of the acquisition components, and then implement dynamic area division, focusing on important areas, thereby improving the detection efficiency and reducing the time and resources of ineffective monitoring;

[0042] The feature extraction module 4 is used to extract feature points, edges and texture feature data from the planned area through image processing algorithms. The feature extraction module 4 adopts a deep learning algorithm to synchronously upgrade the feature extraction algorithm according to the update cycle of the measurement and control equipment and the acquisition components; it provides more dimensional information for comparison data, greatly improving the accuracy of anomaly judgment;

[0043] The judgment unit 5 is used to compare the acquired feature data with predefined standard data to judge whether there is an anomaly; by comparing the acquired feature data with the predefined standard for anomaly judgment, the high reliability of the system is ensured;

[0044] The anomaly analysis module 6 is used to analyze the submitted anomalies, identify the anomaly types and causes, output relevant anomaly information and the involved anomaly equipment. The anomaly analysis module 6 sets reasonable thresholds based on historical data, quickly screens the real-time data and classifies the identified anomalies, and identifies the credibility of the anomaly results through a combination of data simulation and historical data verification, generating a detailed anomaly report, including anomaly types, causes, occurrence times and affected equipment information; it provides a basis for subsequent fault troubleshooting and helps the operation and maintenance personnel to make a quick response;

[0045] The marking module 7 is used to mark a number of detected anomaly equipment and a number of associated acquisition components;

[0046] The angle adjustment module 8 is used to analyze relevant anomaly information and the involved anomaly equipment, generate and run adjustment instructions for the monitoring angles of a number of associated acquisition components to be marked, obtain the image states of the anomaly equipment at different angles within a preset cycle, and resubmit them to the judgment unit 5 for secondary analysis. Finally, it judges whether there is an anomaly. When the angle adjustment module 8 finally judges an anomaly, it submits an alarm message to the management module 1, focuses the attention of the associated acquisition components, continuously acquires images, and continuously submits them to the management end. Through a combination of data simulation and historical data verification, the credibility of the judgment result is enhanced, the chance of misjudgment is reduced, reasonable thresholds are set according to historical data, the real-time data is quickly screened and classified, and the efficiency of anomaly detection is greatly improved.

[0047] The angle adjustment module 8 is connected by wireless network interaction with a memory module 9. The memory module 9 is used to record the historical adjustment data of the angle adjustment module 8 and classify it into several adjustment modes. The attributes of the adjustment modes include: the number of target acquisition components, the rotation angle, the rotation sequence, and the duration. When encountering similar anomalies in the future, the historical adjustment mode with a matching degree will be directly called to the angle adjustment module 8 for implementation. Recording the historical adjustment data facilitates the quick call of the previous successful adjustment modes under similar anomalies in the future, reduces the repeated learning process, and helps the system to adapt quickly.

[0048] As a preferred implementation manner in this embodiment, as Figure 1 shown, the management module 1 is connected by wireless network interaction with the image acquisition module 2. The image acquisition module 2 is connected by wireless network interaction with the area planning module 3 and the angle adjustment module 8. The area planning module 3 is connected by wireless network interaction with the feature extraction module 4. The feature extraction module 4 is connected by wireless network interaction with the judgment unit 5. The judgment unit 5 is connected by wireless network interaction with the anomaly analysis module 6. The anomaly analysis module 6 is connected by wireless network interaction with the marking module 7. The marking module 7 is connected by wireless network interaction with the angle adjustment module 8.

[0049] Compared with the prior art, by introducing multi-layer intelligent processing, real-time data monitoring, intelligent anomaly recognition, and flexible adjustment strategies, the accuracy and response speed of monitoring have been greatly improved. The design of this system is based on in-depth learning algorithms and fine analysis of historical data, enabling the device to intervene more accurately and timely in the face of potential anomalies. Compared with the prior art, this intelligent and dynamically adjustable feature makes the production process smoother, improving the overall efficiency and production safety.

[0050] Embodiment 2

[0051] On other levels, the judgment unit 5 provided in this embodiment, as Figure 1 shown, includes a template database 51, an action definition module 52, and a scan definition module 53. The template database 51 is used to compare the device actions involved in the collected image content with the standard templates to determine whether the device operation status meets the expectations. The action definition module 52 is used to compare the collected product status with the preset standard modules to determine whether there are anomalies in the products. The scan definition module 53 establishes a refined action standard template library and a standard module library, covering different actions of different devices and different angular postures of different products;

[0052] The judgment unit 5 analyzes the matching degree between the collected data and the standard data. If the matching degree is lower than the preset threshold, it is judged that the current collected data and the standard data are normal. Otherwise, it is judged as abnormal and immediately submitted to the angle adjustment module 8.

[0053] Compared with the prior art, the template database 51 establishes a refined action standard template library, covering different actions and product postures of different devices. The action definition module 52 can perform serialization analysis using a deep learning model to better capture dynamic actions. The scanning definition module 53 can introduce multispectral imaging technology to improve detection accuracy, and establish a sound standard module library based on machine learning and big data analysis, and gradually optimize the threshold parameters.

[0054] Embodiment 3

[0055] In this embodiment, a design process for monitoring data collection and transmission of an automatic monitoring device for an automated measurement and control device is provided. For example, Figure 2 , the image acquisition module 2 uses the MJPG-streamer open-source software to implement the transmission of the video stream, and can transmit JPEG files to the browser. The design uses the input_uvc.so and output_http.so channels, and uses the JPEG library as the support for the video stream during transmission. Set the video image information, set the default video device to be opened as / dev / video0, the video output resolution to be 640×480, the number of frames to be 15, and the video format to be V4L2_PIX_FMT_MJPEG;

[0056] Then compile the server source code to generate the dynamic link library.so file and the executable file mjpg_streamr, and copy the.so file, the mjpg_streamr executable file, the start.h and the www folder to the gateway Linux system;

[0057] Finally, test the video transmission. Run the start.sh file in the gateway and enter the gateway IP address information in the Web browser;

[0058] The test results are as Figure 2 shown, indicating that the video stream transmission is successful. Set the picture source address in the monitoring web page to the link in the figure to achieve video transmission.

[0059] In summary, when this device is carried, the user needs to deploy the acquisition component in the measurement and control area and connect each functional module to the Internet and the power supply;

[0060] In the measurement and control scenario, products or semi-finished products are transported to the measurement and control equipment via a transport device (such as a conveyor belt). The measurement and control equipment makes a measurement and control reaction and generates an action. This device edits and sends control commands through the management module 1, centrally controls all deployed acquisition components through the image acquisition module 2 for video transmission, real-time identifies and divides the monitoring area in the acquired image through the area planning module 3, extracts the features of the content in the divided area through the feature extraction module 4, and determines whether there is a problem through the judgment unit 5. Specifically, it analyzes whether there is a problem with the movement trajectory of the measurement and control equipment through the template database 51, analyzes whether there is a problem with the products passing through the measurement and control through the action definition module 52, provides a standard template through the scan definition module 53, analyzes the data with a judgment result of "yes" through the exception analysis module 6 for specific problem analysis, marks the measurement and control equipment and acquisition components involved through the marking module 7, and makes the acquisition attention of the acquisition components related to the exception focus on the measurement and control components through the angle adjustment module 8 to obtain a complete verification image within a certain period. After secondary analysis by the judgment unit 5 and the exception analysis module 6, the authenticity of the exception is determined. The memory module 9 memorizes the successful adjustment module, and when a similar event occurs, it directly applies the mode;

[0061] Through modular design, it realizes real-time, intelligent, and efficient device monitoring and exception analysis capabilities, thereby greatly improving the monitoring accuracy and response speed of the production line, and effectively reducing the risks of potential equipment failures and production stagnation.

[0062] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic monitoring device for automated measurement and control equipment, characterized in that: include: The management module (1), as the carrier of each functional module, performs start and stop control of each measurement and control device and each functional module; An image acquisition module (2) is used to control a plurality of acquisition components and to monitor the measurement and control equipment and products on the measurement and control production line in real time through the acquisition components; The area planning module (3) is used to perform dynamic area division after intelligently identifying key areas of different collected images during the rotation and collection cycle of the collection component; A feature extraction module (4), used for extracting feature points, edges and texture feature data from the planning area through an image processing algorithm; A judgment unit (5) is used to compare the acquired characteristic data with predefined standard data to determine whether there is an abnormality; An exception analysis module (6) is used to analyze the submitted exception, identify the type and cause of the exception, and output relevant exception information and the abnormal device involved; A marking module (7), used for marking a number of detected abnormal devices and a number of associated collection components; The angle adjustment module (8) is used to analyze relevant abnormal information and the abnormal equipment involved, generate and run adjustment instructions for monitoring angles of several associated acquisition components that are marked, obtain image states of different angles of the abnormal equipment within a preset period, resubmit them to the judgment unit (5) for secondary analysis, and finally judge whether there is an abnormality.

2. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The regional planning module (3) uses historical data as samples, constructs a state model through a deep learning algorithm, obtains collected images in real time, performs preprocessing, analyzes real-time data through the state model, divides the collected images into several areas, dynamically adjusts the areas to be monitored according to state changes, sorts the identified areas in real time by sensitivity, and gives priority to high-sensitivity areas based on the historical adjustment parameters of the angle adjustment module (8).

3. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The judgment unit (5) comprises a template database (51), an action definition module (52) and a scan definition module (53); the template database (51) is used to compare the device actions involved in the collected image content with the standard template to determine whether the device operation status meets expectations; the action definition module (52) is used to compare the collected product status with the preset standard module to determine whether the product has any abnormality; the scan definition module (53) establishes a refined action standard template library and a standard module library, covering different actions of different devices and different angles and postures of different products.

4. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The judgment unit (5) analyzes the comparison and matching degree between the collected data and the standard data. If the comparison and matching degree is lower than a preset threshold, it is judged that the current collected data and the standard data are normal. Otherwise, it is judged to be abnormal and immediately submitted to the angle adjustment module (8).

5. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The feature extraction module (4) uses a deep learning algorithm to synchronously upgrade the feature extraction algorithm according to the update cycle of the measurement and control equipment and the acquisition component.

6. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The anomaly analysis module (6) sets a reasonable threshold value derived from historical data, quickly screens real-time data and classifies the identified anomalies, identifies the credibility of the anomaly results by combining data simulation with historical data verification, and generates a detailed anomaly report including the anomaly type, cause, occurrence time, and affected device information.

7. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: When the angle adjustment module (8) finally determines that it is abnormal, it submits alarm information to the management module (1), focuses the attention of the associated acquisition components, continuously acquires images, and continuously submits them to the management end.

8. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The angle adjustment module (8) is interactively connected to a memory module (9) via a wireless network. The memory module (9) is used to record historical adjustment data of the angle adjustment module (8) and classify the data into a number of adjustment modes. When a similar anomaly is encountered in the future, the historical adjustment mode that meets the matching degree is directly called to the angle adjustment module (8) for implementation.

9. The automatic monitoring device for automatic measurement and control equipment according to claim 8, characterized in that: The attributes of the adjustment mode include: the number of target acquisition components, the rotation angle, the rotation sequence and the duration.

10. The automatic monitoring device for automatic measurement and control equipment according to claim 1, characterized in that: The management module (1) is interactively connected to the image acquisition module (2) via a wireless network, the image acquisition module (2) is interactively connected to the area planning module (3) and the angle adjustment module (8) via a wireless network, the area planning module (3) is interactively connected to the feature extraction module (4) via a wireless network, the feature extraction module (4) is interactively connected to the judgment unit (5) via a wireless network, the judgment unit (5) is interactively connected to the abnormality analysis module (6) via a wireless network, the abnormality analysis module (6) is interactively connected to the marking module (7) via a wireless network, and the marking module (7) is interactively connected to the angle adjustment module (8) via a wireless network.

Images