A tire defect X-ray image detection and grading system and method

Abstract

The present invention relates to a tire defect detection and grading system and method, the detection and grading system includes an X-ray machine, an AI automatic grading system based on deep learning, an automatic grading system based on pattern recognition, a central grading system, a manual review terminal, a local display, an MES system, and a PLC system. The industrial control hosts of multiple X-ray machines are connected together via a local area network. Each X-ray machine is electrically connected to the AI automatic grading system based on deep learning, the automatic grading system based on pattern recognition, and the central grading system. The AI automatic grading system based on deep learning and the automatic grading system based on pattern recognition are respectively connected to the central grading system. The central grading system is electrically connected to the manual review terminal, the local display, the MES system, and the PLC system. The present invention solves the problems of inconsistent detection standards, large differences in grading results, and low recognition rates in existing methods.

Description

Technical Field

[0001] The present invention relates to the technical field of nondestructive testing in tire factories, and in particular to a tire defect detection and grading system and method. Background Art

[0002] Tire defect detection is crucial to the development of the national tire industry, particularly in light of national development strategies such as "Made in China 2025" and the creation of Industry 4.0 intelligent factories. Currently, X-ray transmission imaging is commonly used to detect tire defects, generating X-ray images of tires at varying grayscale levels. After obtaining tire X-ray images, manual inspection and grading are currently the primary method. This method requires systematic training before implementation. Due to factors such as visual fatigue and inconsistent manual inspection standards, the operator's responsibility, experience, and technical skills significantly influence the grading results.

[0003] Summary of the Invention

[0004] Purpose of the invention: The present invention provides a tire defect detection and grading system and method, the purpose of which is to solve the problems of existing methods such as inconsistent detection standards, large differences in grading results, and low recognition rate.

[0005] Technical solution:

[0006] A tire defect X-ray image detection and grading system includes an X-ray machine, an AI automatic grading system based on deep learning, an automatic grading system based on pattern recognition, a central grading system, a manual review terminal, a local display, an MES system, and a PLC system. The industrial control hosts of multiple X-ray machines are connected together via a local area network. Each X-ray machine is electrically connected to the AI automatic grading system based on deep learning, the automatic grading system based on pattern recognition, and the central grading system. The AI automatic grading system based on deep learning and the automatic grading system based on pattern recognition are respectively connected to the central grading system. The central grading system is electrically connected to the manual review terminal, the local display, the MES system, and the PLC system.

[0007] A grading method of the tire defect X-ray image detection and grading system as claimed in claim 1,

[0008] Step 1: The tire passes through the scanning station of the X-ray machine and the scanner obtains the tire barcode information;

[0009] Step 2: The tire enters the X-ray machine room to obtain an X-ray image of the tire;

[0010] Step 3: The tire barcode information and X-ray image are simultaneously saved on the disk of the industrial control host of the X-ray machine;

[0011] Step 4: The tire X-ray images and barcode information detected by each X-ray machine are transmitted via the local area network to three systems: an AI automatic grading system based on deep learning, an automatic grading system based on pattern recognition, and a central grading system.

[0012] Step 5: The deep learning-based AI automatic grading system uses the AI grading process to grade the X-ray images for "bubble" and "impurity" defects. The grading results are transmitted to the central grading system. If the grading result is "unqualified", the defect location and defect code are marked on the corresponding position of the X-ray image and transmitted to the central grading system. If the grading result is qualified, the "qualified" information is transmitted to the central grading system.

[0013] Step 6: The automatic grading system based on pattern recognition uses the pattern recognition grading process to determine the defect level based on the features of the X-ray image. The grading results are transmitted to the central grading system. If the grading result is "unqualified", the defect location and defect code are marked on the corresponding position of the X-ray image and transmitted to the central grading system. If the grading result is qualified, the "qualified" information is transmitted to the central grading system.

[0014] Step 7: The central grading system performs a comprehensive grading on the corresponding X-ray images according to the grading process. If the comprehensive grading result is "qualified", the tire enters the qualified product channel. If the comprehensive grading result is "unqualified", the unqualified defect location and defect code are directly displayed on the local display, and the manual review end determines whether it is a real defect;

[0015] Step 8: If the manual review result is qualified, the central grading system directly controls the PLC to transport the tire to the qualified product channel and uploads the picture and barcode information to the MES system; if the manual review result is unqualified, the central grading system directly controls the PLC system to transport the tire to the unqualified product channel and uploads the picture, defect location and barcode information to the MES system.

[0016] Furthermore, the AI grading process of the AI automatic grading system based on deep learning is as follows:

[0017] a. The AI automatic grading system based on deep learning obtains tire X-ray images and corresponding barcode information from the X-ray machine's industrial control host via the local area network;

[0018] b. The AI host computer of the deep learning-based AI automatic grading system automatically divides the tire X-ray image into recognizable small images;

[0019] c. The AI host computer distributes these small images to the corresponding computing unit groups;

[0020] d. Each operation unit identifies the small image through the defect recognition method;

[0021] e. Each computing unit transmits the recognition results back to the AI host computer;

[0022] f. The AI host computer summarizes the results sent back by all computing units and generates a grading result for the entire image based on the bubble defect judgment standard and the impurity defect judgment standard;

[0023] g. The AI host computer transmits the grading results to the central grading system via the local area network.

[0024] Furthermore, the grading process of the automatic grading system based on pattern recognition is as follows:

[0025] a. The automatic grading system based on pattern recognition obtains the tire X-ray image and corresponding barcode information from the X-ray machine's industrial control host through the local area network;

[0026] b. The automatic grading system based on pattern recognition automatically selects the detection parameters according to the acquired barcode information;

[0027] c. Pre-process the images using an automatic grading system based on pattern recognition;

[0028] e. Extract relevant features from the preprocessed results and make selections;

[0029] f. Classify defects by characteristics;

[0030] g. Generate grading results based on bubble defect determination criteria and impurity defect determination criteria; the grading results are "qualified" or "unqualified"; "unqualified" also includes defect code and defect location information;

[0031] e. The grading results are transmitted to the central grading system.

[0032] Furthermore, the characteristics in step e include "cord sparseness", "joint open", "cord bending", "cross-cord paralleling", "folding", "cord breakage", and "scattered cords".

[0033] Furthermore, the bubble defect judgment standard is that if bright spots with bubble characteristics can be detected by X-ray, it is judged as unqualified, otherwise it is qualified; the impurity defect judgment standard is that the number of impurities in the bead and crown parts is less than or equal to 4, and the number of impurities in the sidewall parts is less than or equal to 3, and there is no obvious deformation, folding and other characteristics of the cord arrangement in the bead, crown and sidewall parts, then it is judged as qualified, otherwise it is unqualified.

[0034] Furthermore, the comprehensive grading process of the central grading system is as follows:

[0035] a. The central grading system uses the received barcode information to sequentially query and analyze the tire grading results of the pattern recognition-based automatic grading system and the deep learning-based AI automatic grading system;

[0036] b. If the automatic grading system based on pattern recognition and the deep learning-based AI automatic grading system both give a qualified result, the tire is considered qualified. The central grading system then controls the PLC system to route the tire to the qualified product channel.

[0037] c. If the automatic grading system based on pattern recognition determines that the tire is qualified, but the automatic grading system based on deep learning determines that the tire is unqualified, the location and code of the unqualified defect will be transmitted to the central grading system. The central grading system will then transmit the location and code of the unqualified defect to a local display. A manual reviewer will visually determine whether the tire is qualified. Finally, the central grading system will control the PLC system to send the tire to the appropriate channel.

[0038] d. If the pattern recognition-based automatic grading system 3 determines that the tire is unqualified, but the deep learning-based AI automatic grading system determines that the tire is qualified, the location and defect code of the unqualified defect determined by pattern recognition will be transmitted to the central grading system. A technician will visually determine whether the tire is qualified. Ultimately, the central grading system will control the PLC system to send the tire to the appropriate channel.

[0039] e. If the automatic grading system based on pattern recognition determines that the tire is unqualified and the AI automatic grading system based on deep learning also determines that the tire is unqualified, the unqualified defect location and defect code determined by the automatic grading system based on pattern recognition and the AI system will be transmitted to the central grading system. The central grading system will transmit the unqualified defect location and defect code to the local display, and the manual review end will visually determine whether the tire is qualified. Finally, the central grading system will control the PLC system to send the tire to the corresponding channel.

[0040] Beneficial effects:

[0041] This grading method greatly reduces manual workload. Instead of having to visually judge all the information in the entire image, humans only need to judge the defect location and defect code transmitted by the "deep learning-based AI automatic grading system" and the "pattern recognition-based automatic grading system." BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 System overall architecture diagram;

[0043] Figure 2 This is a flowchart of the AI automatic grading system based on deep learning;

[0044] Figure 3 This is a flow chart of the automatic grading system based on pattern recognition;

[0045] Figure 4 This is the flow chart of the central grading system.

[0046] Labels in the figure: 1. X-ray machine, 2. AI automatic grading system, 3. Automatic grading system based on pattern recognition, 4. Central grading system, 5. Manual compound terminal, 6. Local display, 7. MES system, 8. PLC system. DETAILED DESCRIPTION

[0047] The present invention will be described in more detail below with reference to the accompanying drawings.

[0048] like Figure 1 As shown, the tire defect detection and grading system of the present invention includes an X-ray machine 1, an AI automatic grading system based on deep learning 2, an automatic grading system based on pattern recognition 3, a central grading system 4, a manual review terminal 5, a local display 6, an MES system 7, and a PLC system 8. The industrial control host computers of all multiple X-ray machines 1 in the factory are connected together via a local area network. The X-ray images and barcode information generated by each X-ray machine 1 when inspecting a tire are simultaneously transmitted via the network to the AI automatic grading system based on deep learning 2, the automatic grading system based on pattern recognition 3, and the central grading system 4. The AI automatic grading system based on deep learning 2 and the automatic grading system based on pattern recognition 3 also input their output results to the central grading system 4, which then outputs signals to the local display 6, the MES system 7, and the PLC system 8 respectively. The manual review terminal 5, based on the information on the local display 6, then inputs the information from the manual review terminal into the central grading system 4.

[0049] There are many types of tire defects, which can be generally divided into two categories: bubbles and impurities, and steel wire skeletons. Since the deep learning-based AI automatic grading system 2 requires a large number of samples for training, not all defects have a large number of samples to support it. However, bubbles and impurities are defects with very high incidence rates, so the deep learning-based AI automatic grading system 2 is used to automatically grade bubbles and impurities. The pattern recognition-based system can accurately grade steel wire skeleton defects, so the pattern recognition-based automatic grading system 3 is used to automatically grade steel wire skeleton defects.

[0050] like Figure 1 As shown, a tire defect detection and grading method process is as follows:

[0051] Step 1: The tire passes through the scanning station of the X-ray machine 1 and the barcode information of the tire is obtained by the scanner;

[0052] Step 2: The tire enters the X-ray machine room 1 to obtain a tire X-ray image;

[0053] Step 3: The barcode information and X-ray image of the tire are simultaneously saved on the disk of the industrial control host of the X-ray machine 1;

[0054] Step 4: The tire X-ray images and barcode information generated by each X-ray machine 1 are transmitted via the local area network to three systems: the deep learning-based AI automatic grading system 2, the pattern recognition-based automatic grading system 3, and the central grading system 4.

[0055] Step 5: The deep learning-based AI automatic grading system 2 uses the AI grading process to grade the X-ray image for "bubble" and "impurity" defects. The grading results are transmitted to the central grading system 4. If the grading result is "unqualified," the defect location and defect code are marked at the corresponding position on the X-ray image and transmitted to the central grading system 4. If the grading result is qualified, the "qualified" information is transmitted to the central grading system 4.

[0056] like Figure 2 As shown, the grading process of the AI automatic grading system 2 based on deep learning is as follows:

[0057] a. The deep learning-based AI automatic grading system 2 obtains the tire X-ray image and corresponding barcode from the X-ray machine's industrial control host via the local area network;

[0058] b. The AI host computer automatically divides the tire X-ray image into recognizable small images; the pixels of the small images are 512x512, which can match the algorithm model under these conditions and has high computational efficiency;

[0059] c. The AI host computer distributes these small images to the corresponding computing unit groups;

[0060] d. Each computing unit identifies the small image using a defect recognition method, which uses the defect recognition method in patent CN108711148;

[0061] e. Each computing unit transmits the recognition results back to the AI host computer;

[0062] f. The AI host computer summarizes the results sent back by all computing units and generates a grading result for the entire image based on the bubble defect judgment standard and the impurity defect judgment standard;

[0063] The criterion for determining bubble defects is that if a bright spot with bubble characteristics can be detected by X-ray, the product is deemed unqualified, otherwise it is qualified.

[0064] The standard for judging impurity defects is that the number of impurities in the tire bead and crown is less than or equal to 4 spacings (the steel wires of a full steel radial tire are evenly distributed, and the distance between every two steel wires is the spacing), and the number of impurities in the sidewall is less than or equal to 3 spacings, and there is no obvious deformation, folding or other characteristics of the cord arrangement in the tire bead, crown and sidewall. If so, it is judged as qualified, otherwise it is unqualified.

[0065] g. The AI host computer transmits the grading results to the central grading system 4 via the local area network.

[0066] Step 6, the "automatic grading system 3 based on pattern recognition" uses the pattern recognition grading process to identify wire skeleton defects such as "cord sparseness", "joint openness", "cord bending", "cross-line paralleling", "folding", "broken wire", and "loose wire" based on the characteristics of the X-ray image. The grading results are transmitted to the central grading system 4. If the grading result is "unqualified", the defect location and defect code are marked at the corresponding position of the X-ray image and transmitted to the central grading system 4. If the grading result is qualified, the "qualified" information is transmitted to the central grading system 4. Figure 3 As shown, the grading process of the automatic grading system 3 based on pattern recognition is as follows:

[0067] a. Automatic grading system 3 based on pattern recognition obtains tire X-ray images and corresponding barcodes from the X-ray machine industrial control host via a local area network;

[0068] b. The system automatically selects detection parameters based on the acquired barcode information; the parameters include: tire specifications, models, matching templates, etc.

[0069] c. Preprocess the image; preprocessing includes image processing methods such as layering, enhancement, and binarization of tire images.

[0070] e. Extract relevant features from the processed results and make selections;

[0071] f. Classify defects by characteristics;

[0072] g. Generate grading results based on bubble defect determination criteria and impurity defect determination criteria; the grading results are "qualified" or "unqualified"; "unqualified" also includes defect code and defect location information;

[0073] The criterion for determining bubble defects is that if a bright spot with bubble characteristics can be detected by X-ray, the product is deemed unqualified, otherwise it is qualified.

[0074] The criteria for determining impurity defects are that the number of impurities in the bead and crown is less than or equal to 4, and the number of impurities in the sidewall is less than or equal to 3, and there is no obvious deformation, folding or other characteristics of the cord arrangement in the bead, crown and sidewall. If so, it is considered qualified, otherwise it is unqualified.

[0075] e. The grading results are transmitted to the central grading system 4.

[0076] In step seven, the central grading system 4 grades the corresponding X-ray images according to the grading process. If the comprehensive grading result is "pass," the tire enters the qualified product channel. If the comprehensive grading result is "fail," the defect location and defect code are directly displayed on the local display 6, and the technicians at the manual review terminal 5 determine whether it is a genuine defect.

[0077] like Figure 4 As shown, the grading process of the central grading system 4 is as follows:

[0078] a. The central grading system 4 sequentially queries and analyzes the received barcodes to obtain the tire judgment results of the automatic grading system 3 based on pattern recognition and the AI automatic grading system 2 based on deep learning;

[0079] b. If the automatic grading system 3 based on pattern recognition has a qualified grading result, and the automatic grading system 2 based on deep learning also has a qualified grading result, then the tire is qualified. The central grading system 4 then controls the PLC system 8 to move the tire to the qualified product channel.

[0080] c. If the pattern recognition-based automatic grading system 3 determines a tire is acceptable, but the deep learning-based AI-based automatic grading system 2 determines a tire is unacceptable, the AI-based defect location and defect code are transmitted to the central grading system 4. The central grading system 4 then transmits the defect location and defect code to the local display 6. A technician at the manual review terminal 5 visually determines whether the tire is acceptable. Finally, the central grading system 4 controls the PLC system 8 to send the tire to the appropriate channel.

[0081] d. If the pattern recognition-based automatic grading system 3 determines that the tire is unqualified, but the AI system determines that it is qualified, the location and defect code of the unqualified defect determined by pattern recognition will be transmitted to the central grading system 4, where technicians will visually determine whether the tire is qualified. Ultimately, the central grading system 4 controls the PLC system 8 to send the tire to the appropriate channel.

[0082] e. If the automatic grading system 3 based on pattern recognition determines that the tire is unqualified and the AI system also determines that the tire is unqualified, the location and code of the unqualified defect determined by the automatic grading system 3 based on pattern recognition and the AI system are transmitted to the central grading system 4. The central grading system 4 then transmits the location and code of the unqualified defect to the local display 6. The technician at the manual review terminal 5 visually determines whether the tire is qualified using the local display 6. Finally, the central grading system 4 controls the PLC system 8 to send the tire to the appropriate channel. Step 8: If the manual review terminal 5 reviews the tire as qualified, the central grading system 4 directly controls the PLC system 8 to send the tire to the qualified channel and simultaneously uploads the image and barcode information to the MES system 7. If the manual review terminal 5 reviews the tire as unqualified, the central grading system 4 directly controls the PLC system 8 to send the tire to the unqualified channel and simultaneously uploads the image, defect location, and barcode information to the MES system 7.

[0083] Currently, there are three tire X-ray image grading methods in use in China: manual grading, template matching grading, and AI intelligent grading. Each has its own advantages and disadvantages. As described in the table below, this method effectively combines the three grading methods to maximize their strengths and minimize their weaknesses.

[0084]

[0085] Existing manual visual inspection methods limit each operator to inspecting images from only one X-ray machine, resulting in high labor costs and labor intensity, and the grading results are significantly affected by subjective factors. This method allows a single operator to simultaneously monitor images from three X-ray machines, saving 60% of labor costs in the factory's X-ray inspection process. It also significantly reduces operator workload and improves inspection efficiency, exceeding the accuracy standard for factory inspection.

Claims

1. A method for detecting and grading tire defect X-ray images using a tire defect X-ray image detection and grading system, characterized by: The detection and grading system comprises an X-ray machine (1), an AI automatic grading system based on deep learning (2), an automatic grading system based on pattern recognition (3), a central grading system (4), a manual review terminal (5), a local display (6), an MES system (7) and a PLC system (8), the industrial control hosts of multiple X-ray machines (1) are connected together through a local area network, each X-ray machine (1) is electrically connected to the AI automatic grading system based on deep learning (2), the automatic grading system based on pattern recognition (3) and the central grading system (4), the AI automatic grading system based on deep learning (2) and the automatic grading system based on pattern recognition (3) are respectively connected to the central grading system (4), and the central grading system (4) is electrically connected to the manual review terminal (5), the local display (6), the MES system (7) and the PLC system (8); the method comprises: Step 1: The tire passes through the scanning station of the X-ray machine (1) and the barcode information of the tire is obtained by the scanner; Step 2: The tire enters the X-ray machine (1) to obtain an X-ray image of the tire; Step 3: The tire barcode information and X-ray image are simultaneously saved on the disk of the industrial control host of the X-ray machine (1); Step 4: The tire X-ray images and barcode information detected by each X-ray machine (1) are transmitted via the local area network to three systems: the AI automatic grading system based on deep learning (2), the automatic grading system based on pattern recognition (3), and the central grading system (4); Step 5: The AI automatic grading system (2) based on deep learning uses the AI grading process to grade the X-ray image for "bubble" and "impurity" defects; the grading results are transmitted to the central grading system (4); if the grading result is "unqualified", the defect location and defect code are marked at the corresponding position of the X-ray image and transmitted to the central grading system (4); if the grading result is qualified, the "qualified" information is transmitted to the central grading system (4); Step 6: The automatic grading system based on pattern recognition (3) uses the pattern recognition grading process to grade defects based on the features of the X-ray image and transmits the grading results to the central grading system (4). If the grading result is "unqualified", the defect location and defect code are marked at the corresponding position of the X-ray image and transmitted to the central grading system (4); if the grading result is qualified, the "qualified" information is transmitted to the central grading system (4); Step 7: The central grading system (4) performs a comprehensive grading on the corresponding X-ray images according to the grading process. If the comprehensive grading result is "qualified", the tire enters the qualified product channel. If the comprehensive grading result is "unqualified", the unqualified defect location and defect code are directly displayed on the local display (6), and the manual review terminal (5) determines whether it is a real defect; Step 8: If the manual review end (5) reviews the tire as qualified, the central grading system (4) directly controls the PLC to transport the tire to the qualified product channel, and uploads the image and barcode information to the MES system (7); if the manual review end (5) reviews the tire as unqualified, the central grading system (4) directly controls the PLC system (8) to transport the tire to the unqualified product channel, and uploads the image, defect location and barcode information to the MES system (7).

2. The method according to claim 1, wherein: The AI grading process of the deep learning-based AI automatic grading system (2) is as follows: a. The AI automatic grading system based on deep learning (2) obtains the tire X-ray image and the corresponding barcode information from the industrial control host of the X-ray machine (1) through the local area network; b. The AI host computer of the deep learning-based AI automatic grading system (2) automatically divides the tire X-ray image into recognizable small images; c. The AI host computer distributes these small images to the corresponding computing unit groups; d. Each operation unit identifies the small image through the defect recognition method; e. Each computing unit transmits the recognition results back to the AI host computer; f. The AI host computer summarizes the results sent back by all computing units and generates a grading result for the entire image based on the bubble defect judgment standard and the impurity defect judgment standard; g. The AI host computer transmits the grading results to the central grading system (4) via the local area network.

3. The method according to claim 1, wherein: The grading process of the automatic grading system (3) based on pattern recognition is as follows: a. Automatic grading system based on pattern recognition (3) obtains tire X-ray images and corresponding barcode information from the X-ray machine (1) industrial control host via a local area network; b. The automatic grading system based on pattern recognition (3) automatically selects detection parameters according to the acquired barcode information; c. Automatic grading system based on pattern recognition (3) pre-processes the images; e. Extract relevant features from the preprocessed results and make selections; f. Classify defects by characteristics; g. Generate grading results based on bubble defect determination criteria and impurity defect determination criteria; the grading results are "pass" or "fail"; "fail" also includes the defect code and defect location information; e. The grading results are transmitted to the central grading system (4).

4. The method according to claim 3, wherein: The characteristics in step e include "cord sparseness", "joint open", "cord bending", "cross-cord paralleling", "folding", "cord breakage" and "scattered cord".

5. The method according to claim 2 or 3, characterized in that: The standard for judging bubble defects is that if bright spots with bubble characteristics can be detected by X-ray, it is judged as unqualified, otherwise it is qualified; the standard for judging impurity defects is that the number of impurities in the bead and crown areas is less than or equal to 4, and the number of impurities in the sidewall areas is less than or equal to 3, and there is no obvious deformation or folding characteristics of the cord arrangement in the bead, crown and sidewall areas, then it is judged as qualified, otherwise it is unqualified.

6. The method according to claim 1, wherein: The comprehensive grading process of the Central Grading System (4) is as follows: a. The central grading system (4) sequentially queries and analyzes the tire judgment results of the pattern recognition-based automatic grading system (3) and the deep learning-based AI automatic grading system (2) using the received barcode information; b. If the grading result of the automatic grading system based on pattern recognition (3) is qualified, and the result of the automatic grading system based on deep learning (2) is also qualified, it indicates that the tire is qualified. Then the central grading system (4) controls the PLC system (8) to make the tire go to the qualified product channel; c. If the grading result of the automatic grading system based on pattern recognition (3) is qualified, but the result of the automatic grading system based on deep learning (2) is unqualified, the unqualified defect location and defect code are transmitted to the central grading system (4). The central grading system (4) transmits the unqualified defect location and defect code to the local display (6). The manual review end (5) judges whether it is a qualified product by naked eyes. Finally, the central grading system (4) controls the PLC system (8) to send the tire to the corresponding channel; d. If the automatic grading system (3) based on pattern recognition judges the tire as unqualified, and the AI automatic grading system (2) based on deep learning judges the tire as qualified, the location of the unqualified defect and the defect code determined by pattern recognition will be transmitted to the central grading system (4), where technicians will visually judge whether the tire is qualified. Finally, the central grading system (4) controls the PLC system (8) to send the tire to the corresponding channel; e. If the automatic grading system (3) based on pattern recognition judges that the tire is unqualified and the AI automatic grading system (2) based on deep learning also judges that the tire is unqualified, the unqualified defect location and defect code judged by the automatic grading system (3) based on pattern recognition and the AI system will be transmitted to the central grading system (4). The central grading system (4) will transmit the unqualified defect location and defect code to the local display (6). The manual review end (5) will visually judge whether the tire is qualified. Finally, the central grading system (4) will control the PLC system (8) to send the tire to the corresponding channel.

Images