OCR (Optical Character Recognition) intelligent recognition and transmission system and method capable of recognizing metal laser lettering

Abstract

The invention relates to the technical field of OCR (Optical Character Recognition), in particular to an OCR intelligent recognition and transmission system and method capable of recognizing metal laser lettering. The system comprises N OCR (Optical Character Recognition) intelligent recognition devices and an upper system, when N OCR intelligent devices work at the same time, products detected by different devices are different, and a single device can recognize a plurality of characters at the same time. The N OCR intelligent devices interact with the MES system through the upper system; when the upper system interacts with the OCR intelligent recognition device, the upper system has an image acquisition function, an image processing function, a character recognition function, a result processing and feedback function, an exception handling and feedback function and a remote operation and data management function. When an upper system interacts with an MES system, interaction data is divided into three types, namely production data, alarm data and equipment performance data, and meanwhile, heartbeat detection is needed to maintain normal communication between equipment and the MES system.

Description

Technical Field

[0001] The present invention relates to the field of OCR recognition technology, and in particular to an OCR intelligent recognition and transmission system and method capable of recognizing metal laser engraving. Background Art

[0002] With the rapid development of information technology, OCR technology has been widely used in industrial production, especially in the production process, providing a data basis for the traceability of later production data, quality data, etc.

[0003] Existing OCR recognition methods and devices have deficiencies in software and hardware offline processing, remote data transmission, data storage, screen synchronization, optical system parameter adjustment, and character recognition, and cannot meet complex and changing production needs. Summary of the invention

[0004] In response to the above technical problems, an OCR intelligent recognition and transmission system and method that can recognize metal laser engraving is provided to solve the problem of rapid response and compatibility with the recognition of products of different varieties and batches, integrate OCR recognition into the production process, and realize human-machine collaboration.

[0005] Technical solution:

[0006] Provided is an OCR intelligent recognition and transmission system capable of recognizing metal laser engraving, comprising:

[0007] N OCR intelligent recognition devices and host systems;

[0008] When N OCR smart devices work at the same time, different devices detect different products, and a single device can recognize multiple characters at the same time; N OCR smart devices interact with the MES system through the upper system;

[0009] When the upper system interacts with the OCR intelligent recognition device, it has image acquisition function, image processing function, character recognition function, result processing and feedback function, exception processing and feedback function, remote operation and data management function;

[0010] When the upper system interacts with the MES system, the interactive data is divided into three categories: production data, alarm data, and equipment performance data. At the same time, in order to maintain normal communication between the equipment and the MES system, heartbeat detection is also required.

[0011] Furthermore, the production data is requested through the MES system, and the upper system returns the production index data after receiving the request. The communication method is http; the MES system is the client, and the upper system is the server, which is called by defining the IP address and port number; the production data is the recognized characters, processed characters and original pictures;

[0012] When the device generates an alarm, the upper system reports the abnormal alarm to the MES system in real time. The communication method is http. The upper system acts as the client and the MES system acts as the server.

[0013] According to the frequency of change of performance indicators, the MES system sends collection requests regularly, and the upper system responds and returns the corresponding equipment performance data. The communication method is http; the MES system acts as the client and the upper system acts as the server.

[0014] To ensure normal communication between the equipment and the MES system, the upper system needs to send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, the communication is considered to be interrupted.

[0015] Furthermore, the OCR intelligent recognition device mainly consists of a control box 1, a light source controller 2, a light source control panel 3, a start button 4, a product positioning tool 5, a light source 6, a protective cover 7, a camera lens 8, a laser indication module 9, a heat dissipation system 10, and a proximity switch 11;

[0016] The control box 1, the light source controller 2, and the heat dissipation system 9 are installed in the chassis at the bottom of the equipment to realize the recognition process control and the adjustment of the light source parameters; the product positioning tooling 5 is installed on the chassis at the bottom of the equipment, and the proximity switch 11 is fixed on the product positioning tooling 5, which can detect whether different products are loaded and unloaded; the light source 6 and the camera lens 8 are installed on the top of the equipment to realize the photo collection of characters; the light source control panel 3 and the start button 4 are installed at the front end of the chassis at the bottom of the equipment to display the light source parameters in real time; the laser indication module 9 is installed and fixed on the side of the camera lens, and the indication light is projected on the product positioning tooling 5, which is convenient for prompting manual loading.

[0017] Furthermore, the visual real-time monitoring screen of the OCR intelligent recognition device is integrated into the MES system page, and the real-time monitoring image is displayed to the user through a unified interface.

[0018] Furthermore, the upper system provides a web service URL address that can be connected for the MES system to call and integrate the visual real-time monitoring screen.

[0019] Furthermore, the OCR intelligent recognition device is used to recognize reflective fonts on metal surfaces;

[0020] Light source 6 uses two strip lights and a ring light source for lighting. The brightness of the two groups of light sources is automatically adjusted according to the imaging effect through the program to ensure that the lighting position has obvious light and dark contrast at the metal engraving position and adapts to the light and dark contrast of different machine-processed product backgrounds.

[0021] In a second aspect, an OCR intelligent recognition and transmission method capable of recognizing metal laser engraving is provided, comprising:

[0022] Step 1. System initialization: startup and self-test process: turn on the power, activate the device power, and the system starts automatically;

[0023] Hardware self-check: The system performs a functional check on the camera, laser scanner, and other sensors to confirm that all hardware components are in normal condition and there are no faults.

[0024] Step 2: Software loading;

[0025] Receive MES system signals: The system detects whether the MES system sends a start-up instruction. If a task instruction is received, the system starts loading the program on the corresponding IP device;

[0026] Software loading process: The program loads all necessary software modules, including image processing software, machine learning algorithm library and user interface;

[0027] Network connection check: Verify the network interface between the device and the MES system to ensure that a stable network connection can be established with the remote service or control center; send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, it is considered that the communication is interrupted; if feedback is received from the MES system after sending the heartbeat, the program starts image acquisition;

[0028] Step 3: Image acquisition

[0029] Image acquisition settings: The program automatically adjusts the light intensity and focal length to optimize the lighting conditions on the metal surface and reduce the effects of reflections and shadows; the program adjusts the light source for comparison and analysis. If the recognition confidence is low, the light source brightness is automatically switched for recognition until the light source is adjusted to the optimal parameters and recognition and photography begins;

[0030] Camera start: Start the high-resolution camera, synchronously capture the metal surface, especially the engraving area, and process the recognized image;

[0031] Step 4: Image processing

[0032] Image reception: The system receives the original image data from the camera;

[0033] Denoising and enhancement: Apply digital filtering technology to remove image noise, adjust image contrast and brightness, and optimize subsequent feature recognition effects;

[0034] Grayscale conversion: convert to grayscale image to simplify data processing flow and improve processing speed;

[0035] Edge detection: Use edge detection technology to extract character boundaries in the image;

[0036] Shape recognition: Analyze the extracted edges to determine the basic shape and size of the characters;

[0037] Step 5: Character Recognition

[0038] Model application: The extracted features are input into the pre-trained deep learning model, and the model recognizes and classifies characters based on the learned data;

[0039] Result generation: Generate recognition results, including character types and related confidence scores; determine product category names for results that meet the confidence score;

[0040] Step 6: Result processing and feedback

[0041] Result collation: The recognition results are collated according to the set format, including the recognized characters and confidence levels. The collected sensor signals are used for judgment, bound to the product character results, and the final results are transmitted to the MES system; if the result is NG, the NG result and the position of the detected product are transmitted back to the MES system;

[0042] Display output: The processing results are displayed on the user interface embedded in the integrated computer for analysis and further operation by the operator and the MES system;

[0043] Step 7: Exception handling and feedback

[0044] Status monitoring: The upper system 11 continuously monitors the status and performance indicators of the entire identification process;

[0045] Abnormal alarm: Once any abnormal operation or performance deviation is found, the abnormal alarm system will be triggered immediately, and the device interface will prompt and send it to the MES system;

[0046] Step 8: Remote Operation and Data Management

[0047] Remote data transmission and control: It can automatically upload recognition results and operation logs to a remote server for storage, facilitating data backup and further analysis.

[0048] Remote control: allows users to remotely log in to the device to adjust settings or update operating instructions to meet different recognition needs;

[0049] Step 9: The equipment continuously detects different products

[0050] Product compatibility: After the system loads products in different tooling positions, it automatically triggers the loading system through the proximity switch to achieve automatic detection, and at the same time distinguishes different product categories and returns the category information and characters.

[0051] Furthermore, character recognition includes: character products recognized by multiple devices have intersections, and recognition is performed through database comparison, deep learning algorithm, and character feature comparison, and the recognition character data streams of multiple devices are interoperable.

[0052] Beneficial effects:

[0053] 1. Automatic reconnection when software and hardware are disconnected, and automatic alarm when abnormalities occur: The device has built-in disconnection detection and automatic reconnection mechanisms. When software and hardware connections are abnormal, it can automatically detect disconnections and reconnect. At the same time, the device also has abnormality detection and automatic alarm functions. Once an abnormal situation is detected, such as too high recognition error rate, too high device temperature, etc., the alarm mechanism will be automatically triggered to ensure the stability and safety of the device operation.

[0054] 2. The software supports TCP / IP protocol and Http protocol remote data transmission: The software processing unit supports TCP / IP protocol and Http protocol, and can realize remote (LAN or company intranet) data transmission function. Users can access the device remotely through the network (LAN or company intranet), obtain recognition results, device status and other information, and can remotely control the device to perform related operations.

[0055] 2. The software supports remote data storage function: The device has a built-in remote data storage module, which can store recognition results, device status and other information in real time to a remote server. Users can access the server at any time to obtain the required data information and conduct further processing and analysis.

[0056] 3. The software supports remote screen synchronization function and realizes page integration with the production line management and control system: The device supports remote screen synchronization function, which can synchronously transmit the real-time screen of the device to the remote terminal. At the same time, the device also supports page integration with the production line management and control system. Users can directly view the operation status, identification results and other information of the equipment on the interface of the production line management and control system to realize comprehensive production line monitoring and management.

[0057] 4. The software supports automatic switching of optical system parameters according to various products, and the device realizes multi-position character recognition of various products: the device has built-in multiple optical system parameter templates, which can automatically switch the corresponding parameter templates according to different products. At the same time, the device also supports multi-position character recognition function, which can simultaneously recognize multiple character positions on the product, improving recognition efficiency and accuracy.

[0058] 5. The software supports deep character learning and intelligent logic detection: The device has built-in deep learning algorithms and intelligent logic detection modules, which can perform deep learning and intelligent logic detection on the recognition results. Through continuous learning and optimization, the recognition accuracy and efficiency are improved, and some machine-engraved characters can be intelligently judged and processed. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. It is obvious that the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 This is a schematic diagram of the structure of an OCR intelligent recognition device in an embodiment of the present invention;

[0061] Figure 2 It is a structural schematic diagram of an OCR intelligent recognition and transmission system capable of recognizing metal laser engraving in an embodiment of the present invention;

[0062] Figure 3 It is a flow chart of an OCR intelligent recognition and transmission method capable of recognizing metal laser engraving in an embodiment of the present invention;

[0063] Figure 4 This is a flowchart of OCR deep learning in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The embodiment of the present invention provides an OCR intelligent recognition device capable of recognizing metal laser engraving. In order to enable those skilled in the art to better understand the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0065] like Figure 2 As shown, the OCR intelligent recognition and transmission system includes N OCR intelligent recognition devices and a host system 12. The OCR intelligent recognition device is mainly composed of a control box 1, a light source controller 2, a light source control panel 3, a start button 4, a product positioning tool 5, a light source 6, a protective cover 7, a camera lens 8, a laser indication module 9, a heat dissipation system 10, and a proximity switch 11. Figure 1 As shown. Among them, the control box 1, the light source controller 2, and the heat dissipation system 9 are installed in the chassis at the bottom of the equipment to realize the recognition process control and the adjustment of the light source parameters. The proximity switch 11 is fixed on the product positioning tooling 5, which can detect whether the tooling at different positions is loading and unloading. The light source 6 and the camera lens 8 are installed on the top of the equipment to realize the photo collection of characters. The light source 6 uses two strip lights and one ring light for lighting, and can automatically adjust the brightness of the light source according to the recognition results. The light source control panel 3 is installed at the front end of the equipment to display the light source parameters in real time. The laser indication module 9 is installed and fixed on the side of the lens, and the indication light is projected on the product positioning tooling 5, which is convenient for prompting manual loading. The upper system 12 of the equipment is deployed on an integrated industrial computer, and the collection and processing results and images can be saved locally in real time, and can also be connected and transmitted with the MES system. The device interface is nested in the MES system in the form of a WEB page to realize the real-time transmission of data and pages.

[0066] When multiple OCR smart devices work at the same time, different devices detect different products. A single device can recognize multiple characters at the same time, such as Figure 2 The figure shows a logical flow diagram of multiple devices interacting with the MES system when working simultaneously. Taking the application of the device in a production line as an example, a total of 5 devices are integrated in the production line. Each device is connected to an integrated computer via a network cable, and information is exchanged between the upper system 12 and the MES system. The interface method is unified as a Restful interface, and the request type is HTTP. The request content is in json format and is placed in the body for transmission. The data content returned by the interface is also in json format, and the specific URL request parameters can be customized. The visual real-time monitoring screen is integrated into the MES system page, and the real-time monitoring image is displayed to the user through a unified interface. The upper system 12 provides a dockable URL address for the web service for the MES system to call and integrate.

[0067] Among them, the data interacting with MES is divided into three categories, namely: production data, alarm data, and equipment performance data. At the same time, in order to maintain normal communication between the equipment and the MES system, heartbeat detection is also required. The production data is sent through the MES system, and the upper system 11 returns the production index data after receiving the request. The communication method is http, the MES system is the client, and the upper system 11 is the server, and the call is made by defining the IP address and port number. The production data is the recognized characters, the processed characters, and the original pictures, and the response data format is json format. When the equipment generates an alarm, the upper system 11 reports the abnormal alarm to the MES system in real time. The communication method is http, the upper system 11 is the client, and the MES system is the server. According to the frequency of change of the performance index, the MES system sends a collection request regularly, and the upper system 11 responds and returns the corresponding data. The communication method is http, the MES system is the client, the upper system 11 is the server, and the data format is json. To ensure normal communication between the device and the MES system, the upper system 11 needs to send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, it is considered that the communication is interrupted.

[0068] When the device is performing identification and analysis, multiple products will be detected at the same time. The upper system 11 of the device controls the execution logic as follows Figure 3 shown.

[0069] Step 1: System Initialization

[0070] Startup and self-test process: Turn on the power, activate the device power, and the system starts automatically.

[0071] Hardware self-check: The system performs functional checks on cameras, laser scanners, and other sensors to confirm that all hardware components are in normal condition and no faults exist.

[0072] Step 2: Software Loading

[0073] Receive MES system signals: The system detects whether the MES system sends a start-up instruction. If a task instruction is received, the system starts loading the program on the corresponding IP device.

[0074] Software loading process: The program loads all necessary software modules, including image processing software, machine learning algorithm library and user interface.

[0075] Network connection check: Verify the network interface between the device and the MES system to ensure that a stable network connection can be established with the remote service or control center; send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, it is considered that the communication is interrupted; if feedback is received from the MES system after sending the heartbeat, the program starts image acquisition.

[0076] Step 3: Image acquisition

[0077] Image acquisition settings: The program automatically adjusts the light intensity and focal length to optimize the lighting conditions on the metal surface and reduce the effects of reflections and shadows. The program adjusts the light source for comparison and analysis. If the recognition confidence is low, the light source brightness is automatically switched for recognition until the light source is adjusted to the optimal parameters and recognition and photography begins.

[0078] Camera start: Start the high-resolution camera, synchronously capture the metal surface, especially the engraving area, and process the recognized image.

[0079] Step 4: Image Processing

[0080] Image reception: The system receives the original image data from the camera.

[0081] Denoising and enhancement: Apply digital filtering technology to remove image noise, adjust image contrast and brightness, and optimize subsequent feature recognition effects.

[0082] Grayscale conversion: Convert to grayscale image to simplify data processing flow and improve processing speed.

[0083] Edge Detection: Use edge detection techniques to extract character boundaries in an image.

[0084] Shape recognition: Analyze the extracted edges to determine the basic shape and size of the characters.

[0085] Step 5: Character Recognition:

[0086] Model application: The extracted features are input into a pre-trained deep learning model, and the model recognizes and classifies characters based on the learned data.

[0087] Result generation: Generate recognition results, including the type of characters and related confidence scores. The results that meet the confidence level are used to determine the product category name, such as Figure 4 When multiple devices are working at the same time, the data flows between the devices are intersecting. The recognition and comparison method uses database comparison, deep learning algorithm, and character feature comparison to identify and connect the data flows between the devices.

[0088] Step 6: Result processing and feedback

[0089] Result collation: The recognition results are collated according to the set format, including the recognized characters and confidence levels. The collected sensor signals are used for judgment, bound to the product character results, and the final results are transmitted to the MES system. If the result is NG, the NG result and the position of the detected product are transmitted back to the MES system.

[0090] Display output: The processing results are displayed on the user interface embedded in the integrated computer for analysis and further operation by the operator and MES system.

[0091] Step 7: Exception handling and feedback:

[0092] Status monitoring: The upper system 11 continuously monitors the status and performance indicators during the entire recognition process.

[0093] Abnormal alarm: Once any abnormal operation or performance deviation is found, the abnormal alarm system will be triggered immediately, and the device interface will prompt and send it to the MES system.

[0094] Step 8: Remote Operation and Data Management

[0095] Remote data transmission and control: It can automatically upload recognition results and operation logs to a remote server for storage, facilitating data backup and further analysis.

[0096] Remote control: allows users to remotely log in to the device to adjust settings or update operating instructions to meet different recognition needs.

[0097] Step 9: The device continuously detects different products

[0098] Product compatibility: After the system loads products in different tooling positions, it automatically triggers the loading system through the proximity switch to achieve automatic detection, and at the same time distinguishes different product categories and returns the category information and characters.

[0099] The equipment operation process is as follows:

[0100] (1) Manually double-click to start the upper interface;

[0101] (2) Place the product to be identified in the corresponding tooling position, and the equipment will automatically count down for detection;

[0102] (3) The recognition results are automatically displayed on the front-end WEB interface;

[0103] (4) Manually remove the product and replace it with a new one.

[0104] The overall operation process of the equipment is simple. The operator only needs to place the product in the corresponding position. All identification, judgment and transmission are completed automatically. The upper system is compatible with the interactive docking of multiple devices and the MES system to ensure the accuracy and real-time performance of the transmission process.

Claims

1. An OCR intelligent recognition and transmission system capable of recognizing metal laser engraving, characterized in that: include: N OCR intelligent recognition devices and host systems; When N OCR smart devices work at the same time, different devices detect different products, and a single device can recognize multiple characters at the same time; N OCR smart devices interact with the MES system through the host system; When the upper system interacts with the OCR intelligent recognition device, it has image acquisition function, image processing function, character recognition function, result processing and feedback function, exception processing and feedback function, remote operation and data management function; When the upper system interacts with the MES system, the interactive data is divided into three categories: production data, alarm data, and equipment performance data. At the same time, in order to maintain normal communication between the equipment and the MES system, heartbeat detection is also required.

2. The system according to claim 1, characterized in that Production data is requested through the MES system, and the upper system returns the production index data after receiving the request. The communication method is http; the MES system is the client, and the upper system is the server. The call is made by defining the IP address and port number; the production data is the recognized characters, processed characters and original pictures; When the device generates an alarm, the upper system reports the abnormal alarm to the MES system in real time. The communication method is http. The upper system acts as the client and the MES system acts as the server. According to the frequency of change of performance indicators, the MES system sends collection requests regularly, and the upper system responds and returns the corresponding equipment performance data. The communication method is http; the MES system acts as the client and the upper system acts as the server. To ensure normal communication between the equipment and the MES system, the upper system needs to send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, the communication is considered to be interrupted.

3. The system according to claim 1, characterized in that The OCR intelligent recognition device mainly comprises a control box (1), a light source controller (2), a light source control panel (3), a start button (4), a product positioning tool (5), a light source (6), a protective cover (7), a camera lens (8), a laser indication module (9), a heat dissipation system (10), and a proximity switch (11); The control box (1), the light source controller (2), and the heat dissipation system (9) are installed in the bottom chassis of the device to realize the control of the identification process and the adjustment of the light source parameters; the product positioning tool (5) is installed on the bottom chassis of the device, and the proximity switch (11) is fixed on the product positioning tool (5) to detect whether different products are loaded or unloaded; the light source (6) and the camera lens (8) are installed on the top of the device to realize the photo collection of characters; the light source control panel (3) and the start button (4) are installed at the front end of the bottom chassis of the device to display the light source parameters in real time; the laser indication module (9) is installed and fixed on the side of the camera lens, and the indication light is projected on the product positioning tool (5) to facilitate the prompt of manual loading.

4. The system according to claim 3, characterized in that The visual real-time monitoring screen of the OCR intelligent recognition device is integrated into the MES system page, and the real-time monitoring image is displayed to the user through a unified interface.

5. The system according to claim 4, characterized in that The upper system (12) provides a web service URL address that can be connected to the MES system for calling and integrating the visual real-time monitoring screen.

6. The system according to claim 2, characterized in that OCR intelligent recognition equipment is used to identify reflective fonts on metal surfaces; The light source (6) is illuminated by two strip lights and a ring light source. The brightness of the two light sources is automatically adjusted according to the imaging effect through a program to ensure that the light and dark contrast at the metal engraving position is obvious and the light and dark contrast of different machine-processed product backgrounds is adaptive.

7. An OCR intelligent recognition and transmission method capable of recognizing metal laser engraving, characterized in that: The system according to any one of claims 1 to 6 is characterized by comprising: Step 1. System initialization: startup and self-test process: turn on the power, activate the device power, and the system starts automatically; Hardware self-check: The system performs a functional check on the camera, laser scanner, and other sensors to confirm that all hardware components are in normal condition and there are no faults. Step 2: Software loading; Receive MES system signals: The system detects whether the MES system sends a start-up instruction. If a task instruction is received, the system starts loading the program on the corresponding IP device; Software loading process: The program loads all necessary software modules, including image processing software, machine learning algorithm library and user interface; Network connection check: Verify the network interface between the device and the MES system to ensure that a stable network connection can be established with the remote service or control center; send a heartbeat to the MES system every 1 minute. If no feedback is received from the MES system for 5 consecutive times, it is considered that the communication is interrupted; if feedback is received from the MES system after sending the heartbeat, the program starts image acquisition; Step 3: Image acquisition Image acquisition settings: The program automatically adjusts the light intensity and focal length to optimize the lighting conditions on the metal surface and reduce the effects of reflections and shadows; the program adjusts the light source for comparison and analysis. If the recognition confidence is low, the light source brightness is automatically switched for recognition until the light source is adjusted to the optimal parameters and recognition and photography begins; Camera start: Start the high-resolution camera, synchronously capture the metal surface, especially the engraving area, and process the recognized image; Step 4: Image processing Image reception: The system receives the original image data from the camera; Denoising and enhancement: Apply digital filtering technology to remove image noise, adjust image contrast and brightness, and optimize subsequent feature recognition effects; Grayscale conversion: convert to grayscale image to simplify data processing flow and improve processing speed; Edge detection: Use edge detection technology to extract character boundaries in the image; Shape recognition: Analyze the extracted edges to determine the basic shape and size of the characters; Step 5: Character Recognition Model application: The extracted features are input into the pre-trained deep learning model, and the model recognizes and classifies characters based on the learned data; Result generation: Generate recognition results, including character types and related confidence scores; determine product category names for results that meet the confidence score; Step 6: Result processing and feedback Result collation: The recognition results are collated according to the set format, including the recognized characters and confidence levels; the results are judged through the collected sensor signals, bound to the product character results, and the final results are transmitted to the MES system; if the result is NG, the NG result and the position of the detected product are transmitted back to the MES system; Display output: The processing results are displayed on the user interface embedded in the integrated computer for analysis and further operation by the operator and the MES system; Step 7: Exception handling and feedback Status monitoring: The upper system 11 continuously monitors the status and performance indicators of the entire identification process; Abnormal alarm: Once any abnormal operation or performance deviation is found, the abnormal alarm system will be triggered immediately, and the device interface will prompt and send it to the MES system; Step 8: Remote Operation and Data Management Remote data transmission and control: It can automatically upload the recognition results and operation logs to the remote server for storage, which is convenient for data backup and further analysis; Remote control: allows users to remotely log in to the device to adjust settings or update operating instructions to meet different recognition needs; Step 9: The equipment continuously detects different products Product compatibility: After the system loads products in different tooling positions, it automatically triggers the loading system through the proximity switch to achieve automatic detection, and at the same time distinguishes different product categories and returns the category information and characters.

8. The method according to claim 7, characterized in that Character recognition includes: character products recognized by multiple devices have intersections, and recognition is performed through database comparison, deep learning algorithms, and character feature comparison methods. The recognized character data streams of multiple devices are interoperable.

Images