A rubber product defect detection device and a defect detection method thereof

Abstract

This invention discloses a rubber product defect detection device and method, including a detection and sorting device and a processor. The detection and sorting device includes a loading tray, a belt feeder, and a worktable. The worktable includes a glass plate and multiple sets of detection cameras. The processor integrates a pre-trained defect detection model. The rubber product to be tested is placed on the loading tray; the belt feeder transports the rubber product from the loading tray to the glass plate; the rubber product passes sequentially through the multiple sets of detection cameras; the multiple sets of detection cameras acquire three views of the rubber product; the defect detection model performs defect detection on the three views to obtain the defect detection result; wherein the three views include a top view, a bottom view, and a side view. This invention can realize automated defect detection of rubber products, reduce manual operation, thereby improving production efficiency and defect detection accuracy, and can be widely applied in the field of automated inspection technology.

Description

Technical Field

[0001] This application relates to the field of automated testing technology, and in particular to a defect detection device for rubber products and a defect detection method thereof. Background Technology

[0002] In the field of traditional rubber product quality inspection, defect identification mainly relies on manual operation. While this method is fundamental, it has significant drawbacks. First, for large-scale production of rubber products, manual inspection requires repeatedly handling each product, which is not only labor-intensive but also demands a large investment of human resources, thus significantly increasing production costs and resulting in low production efficiency. Second, prolonged continuous visual inspection can easily lead to visual fatigue among operators, affecting the accuracy and reliability of rubber product inspection, increasing the risk of missed detections and misjudgments, and ultimately potentially causing customer complaints due to product quality issues. Summary of the Invention

[0003] To solve the above-mentioned technical problems, the purpose of this invention is to provide a rubber product defect detection device and defect detection method with high efficiency and accuracy.

[0004] To achieve the above objectives, one aspect of this application proposes a rubber product defect detection device, including a detection and sorting device and a processor. The detection and sorting device includes a loading tray, a belt feeder, and a worktable. The worktable includes a glass plate and multiple sets of detection cameras. One end of the belt feeder is connected to the loading tray, and the other end of the belt feeder is connected to the glass plate. Each set of detection cameras is located above or below the glass plate, and each set of detection cameras is also connected to the processor. The processor integrates a pre-trained defect detection model. The loading tray is used to place the rubber product to be tested. The belt feeder is used to transport the rubber product to be tested from the loading tray to the glass plate. The glass plate is used to allow the rubber product to be tested to pass through each set of detection cameras in sequence. Each set of detection cameras is used to acquire three views of the rubber product to be tested. The defect detection model is used to perform defect detection on the three views to obtain defect detection results. The three views include a top view, a bottom view, and a side view.

[0005] In some embodiments, the detection and sorting device further includes a plurality of sorting boxes, and the workbench further includes a plurality of air inlets. Each air inlet is located on the glass plate, and each sorting box is located below each air inlet. Each air inlet is also connected to the processor. The air inlets are used to blow the rubber product to be tested into the corresponding sorting box according to the defect detection result.

[0006] In some embodiments, the detection camera includes a first detection camera, a second detection camera, and a third detection camera. The first detection camera is located above the glass plate, the second detection camera is located below the glass plate, and the third detection camera is placed on the glass plate. The first detection camera is used to capture the top view of the rubber product to be tested, the second detection camera is used to capture the bottom view of the rubber product to be tested, and the third detection camera is used to capture the side view of the rubber product to be tested.

[0007] In some embodiments, the belt feeder includes a lifting belt and a feeding belt. One end of the lifting belt is connected to the bottom of the loading tray, and the other end of the lifting belt is connected to one end of the feeding belt. The other end of the feeding belt is connected to the edge of the glass tray. The lifting belt is used to transport the rubber product to be tested from the loading tray to the feeding belt, and the feeding belt is used to transport the rubber product to be tested from the lifting belt to the glass tray.

[0008] In some embodiments, the air inlet includes a first air inlet, a second air inlet, and a third air inlet; the sorting box includes a qualified product box, a waste product box, and a pending processing box; the qualified product box is located below the first air inlet; the waste product box is located below the second air inlet; and the pending processing box is located below the third air inlet. The first air inlet, the second air inlet, and the third air inlet are used to blow the rubber product to be tested into the qualified product box, the waste product box, or the pending processing box according to the defect detection result.

[0009] To achieve the above objectives, another aspect of this application proposes a defect detection method for a rubber product defect detection device, comprising the following steps:

[0010] Place the rubber product to be tested onto the loading tray;

[0011] The rubber product to be tested is conveyed from the loading tray to the glass tray by a belt feeder;

[0012] The rubber product to be tested is passed sequentially through multiple sets of detection cameras via the glass disk;

[0013] The three views of the rubber product under test are acquired by multiple sets of detection cameras, including a top view, a bottom view, and a side view.

[0014] The three views are subjected to defect detection using a pre-trained defect detection model within the processor, and the defect detection results are obtained.

[0015] In some embodiments, the defect detection method further includes a step of pre-training the defect detection model, wherein pre-training the defect detection model specifically includes:

[0016] Obtain a training sample set, which includes multiple top view samples, bottom view samples, side view samples of the rubber products to be tested, and corresponding detection labels;

[0017] Determine the defect judgment rules, and perform multi-stage training on the pre-constructed convolutional neural network based on the training sample set, the detection labels, and the defect judgment rules to obtain the defect detection model.

[0018] In some embodiments, determining the defect determination rule involves training a pre-constructed convolutional neural network in multiple stages based on the training sample set, the detection labels, and the defect determination rule to obtain the defect detection model. Specifically, this includes:

[0019] Determine the defect determination rules, and then input the defect determination rules into the convolutional neural network;

[0020] The top view sample and the side view sample are input into the convolutional neural network for the first stage of training. Then, the top view sample and the side view sample are randomly selected to test the convolutional neural network after the first stage of training to obtain the first defect detection result.

[0021] The defect determination rule is updated based on the first defect detection result until the first defect detection result reaches the preset first test condition, and the randomly input top view sample and side view sample reach the preset first quantity threshold.

[0022] The top view sample and the side view sample at the critical point are input into the convolutional neural network for the second stage of training to obtain the second defect detection result.

[0023] The defect determination rule is updated based on the second defect detection result and the detection label until the second defect detection result reaches the preset second test condition, and the input top view sample and side view sample at the critical point reach the preset second quantity threshold.

[0024] The engraving features of the rubber product to be tested are determined, the engraving features are imported into a preset masking list, and then the bottom view sample and the masking list are input into the convolutional neural network for the third stage of training to obtain the third defect detection result.

[0025] The defect determination rules are updated based on the third defect detection result and the detection label until the third defect detection result reaches the preset third test condition and the input bottom view sample reaches the preset third quantity threshold. Then, training stops, and the trained defect detection model is obtained.

[0026] In some embodiments, the step of performing defect detection on the three views using a pre-trained defect detection model within the processor to obtain defect detection results specifically includes:

[0027] The top view, the bottom view, and the side view are input into the defect detection model, and then the top view, the bottom view, and the side view are divided into regions to obtain multiple regions to be tested.

[0028] Defect detection is performed on each of the areas to be tested to obtain the defect detection results;

[0029] The defect detection includes at least one of the following: glue deficiency detection, bubble detection, chipping detection, crack detection, inclusion detection, deformation detection, or dimensional detection.

[0030] In some embodiments, the defect detection method further includes the step of blowing the rubber product to be tested into a corresponding sorting box through multiple air outlets according to the defect detection result. The air outlets include a first air outlet, a second air outlet, and a third air outlet. The sorting box includes a qualified product box, a waste product box, and a pending processing box. Specifically, the step of blowing the rubber product to be tested into the corresponding sorting box through multiple air outlets according to the defect detection result includes:

[0031] When the defect detection result shows that the rubber product to be tested is a qualified product, the rubber product to be tested is blown into the qualified product box through the first air blowing port;

[0032] When the defect detection result shows that the rubber product to be tested is a waste product, the rubber product to be tested is blown into the waste product box through the second air blowing port;

[0033] When the defect detection result shows that the rubber product to be tested is a product to be processed, the rubber product to be tested is blown into the processing box through the third air blowing port.

[0034] The beneficial effects of this invention are as follows: The rubber product defect detection device and method of this invention include a detection and sorting device and a processor. The detection and sorting device includes a loading tray, a belt feeder, and a worktable. The worktable includes a glass plate and multiple sets of detection cameras. The processor integrates a pre-trained defect detection model. This invention automatically acquires top, bottom, and side views of the rubber product to be tested through the detection and sorting device, and then performs defect detection on the acquired top, bottom, and side views using the pre-trained defect detection model. This enables automated defect detection of rubber products, reduces manual operation, and thus improves production efficiency and the accuracy of defect detection. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the embodiments of the present invention are described below. It should be understood that the drawings described below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a structural block diagram of a rubber product defect detection device provided in an embodiment of the present invention;

[0037] Figure 2 This is a top view of the detection and sorting device provided in an embodiment of the present invention;

[0038] Figure 3 This is a front view of the detection and sorting device provided in an embodiment of the present invention;

[0039] Figure 4 A flowchart illustrating the steps of a defect detection method using a rubber product defect detection device provided in this embodiment of the invention;

[0040] Figure 5 A schematic diagram of crack defects in a rubber product provided in an embodiment of the present invention;

[0041] Figure 6 This is a schematic diagram of inclusion defects in a rubber product provided in an embodiment of the present invention;

[0042] Figure 7 This is a schematic diagram of inclusion defects in a rubber product provided in an embodiment of the present invention;

[0043] Figure 8 A schematic diagram of the first-stage training process provided for an embodiment of the present invention;

[0044] Figure 9 A schematic diagram of the second-stage training process provided in an embodiment of the present invention;

[0045] Figure 10 This is a schematic diagram of the bottom of a rubber product with cracks, provided in an embodiment of the present invention.

[0046] Figure 11 A schematic diagram of the bottom of a qualified rubber product provided for an embodiment of the present invention;

[0047] Figure 12 A schematic diagram of the third-stage training process provided in an embodiment of the present invention;

[0048] Figure 13 A schematic diagram of a region division method for a top view of a shock-absorbing pad provided in an embodiment of the present invention;

[0049] Figure 14This is a schematic diagram of another method for dividing the area of ​​the shock-absorbing pad as provided in an embodiment of the present invention;

[0050] Figure 15 A comparative schematic diagram of the top view, bottom view and side view of rubber shock-absorbing pad A and rubber shock-absorbing pad B provided in the embodiments of the present invention;

[0051] Figure 16(a) is a schematic diagram of the defect detection results of the top view of the shock-absorbing pad A provided in the embodiment of the present invention;

[0052] Figure 16(b) is a schematic diagram of the defect detection results of the top view of the shock-absorbing pad B provided in the embodiment of the present invention;

[0053] Figure 17(a) is a schematic diagram of the defect detection results of the shock-absorbing pad A from the bottom view provided in the embodiment of the present invention;

[0054] Figure 17(b) is a schematic diagram of the defect detection results of the shock-absorbing pad B in the top view provided in the embodiment of the present invention;

[0055] Figure 18(a) is a schematic diagram of the defect detection results of the side view of the shock-absorbing pad A provided in the embodiment of the present invention;

[0056] Figure 18(b) is a schematic diagram of the defect detection results of the side view of the shock-absorbing pad provided in the embodiment of the present invention;

[0057] Reference numerals in the attached diagram: 1. Workbench; 2. Glass tray; 3. First air inlet; 4. Second air inlet; 5. Third air inlet; 6. Feed conveyor belt; 7. Lifting conveyor belt; 8. Loading tray; 9. First inspection camera; 10. Second inspection camera; 11. Third inspection camera; 12. Qualified product box; 13. Scrap product box; 14. Box to be processed. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0059] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0060] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0061] In the field of traditional rubber product quality inspection, defect identification mainly relies on manual operation. While this method is fundamental, it has significant drawbacks. First, for large-scale production of rubber products, manual inspection requires repeatedly handling each product, which is not only labor-intensive but also demands a large investment of human resources, thus significantly increasing production costs and resulting in low production efficiency. Second, prolonged continuous visual inspection can easily lead to visual fatigue among operators, affecting the accuracy and reliability of rubber product inspection, increasing the risk of missed detections and misjudgments, and ultimately potentially causing customer complaints due to product quality issues.

[0062] To address this, this invention provides a rubber product defect detection device, comprising a detection and sorting device and a processor. The detection and sorting device includes a loading tray, a belt feeder, and a worktable. The worktable includes a glass plate and multiple sets of detection cameras. The processor integrates a pre-trained defect detection model. This invention automatically acquires top, bottom, and side views of the rubber product to be tested through the detection and sorting device, and then performs defect detection on the acquired top, bottom, and side views using the pre-trained defect detection model. This enables automated defect detection of rubber products, reduces manual operation, and thus improves production efficiency and the accuracy of defect detection.

[0063] Reference Figure 1 , Figure 1This is a structural block diagram of a rubber product defect detection device provided in an embodiment of the present invention. The device includes a detection and sorting device and a processor. The detection and sorting device includes a loading tray 8, a belt feeder, and a workbench 1. The workbench 1 includes a glass plate 2 and multiple sets of detection cameras. One end of the belt feeder is connected to the loading tray 8, and the other end is connected to the glass plate 2. Each set of detection cameras is located above or below the glass plate 2, and each set of detection cameras is also connected to the processor. The processor integrates a pre-trained defect detection model. The loading tray 8 is used to place the rubber product to be tested. The belt feeder is used to transport the rubber product to be tested from the loading tray 8 to the glass plate 2. The glass plate 2 is used to allow the rubber product to be tested to pass through each set of detection cameras in sequence. Each set of detection cameras is used to acquire three views of the rubber product to be tested. The defect detection model is used to perform defect detection on the three views to obtain the defect detection result. The three views include a top view, a bottom view, and a side view.

[0064] Specifically, the inspection and sorting equipment is used to automatically collect top, bottom, and side views of the rubber products to be tested;

[0065] The processor, as the core component of the entire detection device, integrates a defect detection model trained based on deep learning algorithms. It is used to perform various defect detections on the top view, bottom view, and side view of the rubber product to be tested, and obtain the defect detection results.

[0066] Reference Figure 2 and Figure 3 , Figure 2 To inspect the top view of the sorting equipment, Figure 3 To inspect the front view of the sorting equipment, as an optional implementation, the belt feeder includes a lifting belt 7 and a feeding belt 6. One end of the lifting belt 7 is connected to the bottom of the loading tray 8, and the other end of the lifting belt 7 is connected to one end of the feeding belt 6. The other end of the feeding belt 6 is connected to the edge of the glass tray 2. The lifting belt 7 is used to transport the rubber product to be tested from the loading tray 8 to the feeding belt 6, and the feeding belt 6 is used to transport the rubber product to be tested from the lifting belt 7 to the glass tray 2.

[0067] Specifically, the detection and sorting equipment proposed in this embodiment of the invention includes a loading tray 8, a belt feeder, and a worktable 1. The worktable 1 includes a rotatable glass disk 2 and multiple sets of detection cameras. A belt feeder is installed on one side of the glass disk 2. The belt feeder includes a lifting belt 7 and a feeding belt 6. The lifting belt 7 connects the loading tray 8 and the feeding belt 6, and is used to transport the rubber products to be tested from the loading tray 8 to the feeding belt 6. The feeding belt 6 connects the lifting belt 7 and the glass disk 2, and is used to transport the rubber products to be tested from the lifting belt 7 to the glass disk 2.

[0068] Reference Figure 2and Figure 3 As an optional implementation, the detection camera includes a first detection camera 9, a second detection camera 10, and a third detection camera 11. The first detection camera 9 is located above the glass plate 2, the second detection camera 10 is located below the glass plate 2, and the third detection camera 11 is placed on the glass plate 2. The first detection camera 9 is used to capture a top view of the rubber product to be tested, the second detection camera 10 is used to capture a bottom view of the rubber product to be tested, and the third detection camera 11 is used to capture a side view of the rubber product to be tested.

[0069] In some optional embodiments, the inspection camera includes a first inspection camera 9, a second inspection camera 10, and a third inspection camera 11. The first inspection camera 9 is located above the glass disk 2 and is used to take photos of the rubber product to be tested from top to bottom, i.e., a top view of the rubber product to be tested. The second inspection camera 10 is located below the glass disk 2 and is used to take photos of the rubber product to be tested from bottom to top, i.e., a bottom view of the rubber product to be tested. The third inspection camera 11 is placed on the glass disk 2 and is used to take photos of the rubber product to be tested from the side, i.e., a side view of the rubber product to be tested. The third inspection camera 11 is equipped with at least four cameras to capture images of the side of the rubber product to be tested from all angles, and the inspection cameras respectively acquire at least six photos of the rubber product to be tested.

[0070] It should be noted that photos taken from a single angle may lead to false alarms or missed defects due to lighting, shadows, occlusions, etc. The embodiments of the present invention, by taking pictures of the rubber product under test from different angles, can provide more data points for the defect detection module, show the various parts and details of the product, thereby improving the accuracy and reliability of defect detection.

[0071] Furthermore, in this embodiment of the invention, by setting a rotatable and transparent glass disk 2, the rubber product to be tested passes sequentially through the first detection camera 9, the second detection camera 10, and the third detection camera 11, and the top view, bottom view, and side view of the rubber product to be tested are collected. During the process, there is no need for operators to manually flip the product multiple times, which can reduce the input of human resources and improve production efficiency.

[0072] Reference Figure 2 and Figure 3 In some optional embodiments, the inspection and sorting equipment also includes multiple sorting boxes, and the workbench 1 also includes multiple air blowing ports. Each air blowing port is located on the glass plate 2, and each sorting box is located below each air blowing port. Each air blowing port is also connected to the processor. The air blowing ports are used to blow the rubber products to be tested into the corresponding sorting boxes according to the defect detection results.

[0073] Reference Figure 2 and Figure 3In some optional embodiments, the air inlet includes a first air inlet 3, a second air inlet 4, and a third air inlet 5. The sorting box includes a qualified product box 12, a waste product box 13, and a waiting-to-be-processed box 14. The qualified product box 12 is located below the first air inlet 3, the waste product box 13 is located below the second air inlet 4, and the waiting-to-be-processed box 14 is located below the third air inlet 5. The first air inlet 3, the second air inlet 4, and the third air inlet 5 are used to blow the rubber products to be tested into the qualified product box 12, the waste product box 13, or the waiting-to-be-processed box 14 according to the defect detection results.

[0074] In summary, the workflow of a rubber product defect detection device according to an embodiment of the present invention is as follows: During operation, the lifting belt 7 carries the rubber product to be tested from the loading tray 8 to the feeding belt 6, and the feeding belt 6 carries the rubber product to be tested into the glass tray 2. When the glass tray 2 rotates, the rubber product to be tested passes through the first detection camera 9 in sequence, which takes a top view of the rubber product from top to bottom; the rubber product then passes through the second detection camera 10, which takes a bottom view of the rubber product from bottom to top; the rubber product then passes through the third detection camera 11, which takes a side view of the rubber product from the side. Then, the defect detection model in the processor comprehensively determines whether the rubber product to be tested is a qualified product, a defective product, or a product to be processed based on the top view, bottom view, and side view of the rubber product. Qualified products are blown into the qualified product box 12 through the first air inlet 3, defective products are blown into the defective product box 13 through the second air inlet 4, and products to be processed are blown into the processing box 14 through the third air inlet 5.

[0075] The structure and operation of the rubber product defect detection device according to embodiments of the present invention have been described above. It can be recognized that, compared with existing rubber product defect detection methods, embodiments of the present invention have the following advantages:

[0076] 1. By setting a rotatable and transparent glass disk, the rubber product to be tested passes through the first detection camera, the second detection camera and the third detection camera in sequence, and the top view, bottom view and side view of the rubber product to be tested are collected. During the process, there is no need for the operator to manually flip the product multiple times, which can reduce the input of human resources and improve production efficiency.

[0077] Second, by collecting top, bottom, and side views of the rubber product under test from different angles, more data points can be provided for the defect detection module, showing the various parts and details of the product, thereby improving the accuracy and reliability of defect detection.

[0078] Third, by automatically collecting top, bottom, and side views of the rubber products to be tested through the detection and sorting equipment, and then using a pre-trained defect detection model to perform defect detection on the collected top, bottom, and side views, the automatic defect detection of rubber products can be realized, reducing the manual operation process and thus improving production efficiency and the accuracy of defect detection.

[0079] Reference Figure 4 , Figure 4 This invention provides a flowchart of a defect detection method using a rubber product defect detection device. The method, executed by the aforementioned rubber product defect detection device, includes steps S101 to S105.

[0080] S101. Place the rubber product to be tested through the loading tray;

[0081] S102. The rubber product to be tested is conveyed from the loading tray to the glass tray by a belt feeder;

[0082] S103. The rubber product to be tested is passed through multiple sets of detection cameras in sequence via a glass disk;

[0083] S104. Acquire three views of the rubber product to be tested through multiple sets of inspection cameras. The three views include top view, bottom view and side view.

[0084] Furthermore, the defect detection effect of images taken under different light source conditions varies for different areas of the rubber product. Therefore, in order to improve the accuracy and reliability of model defect detection, the top view, bottom view and side view of the rubber product under test should be taken using at least red light source, green light source, blue light source and combined light source.

[0085] S105. Defect detection is performed on the three views using a pre-trained defect detection model within the processor to obtain the defect detection results.

[0086] As an optional implementation, the defect detection method further includes a step of pre-training a defect detection model. This step of pre-training the defect detection model can be further divided into the following steps A101 and A102:

[0087] A101. Obtain the training sample set, which includes top view samples, bottom view samples, side view samples of multiple rubber products to be tested, and corresponding detection labels.

[0088] Specifically, a certain number of top view samples, bottom view samples, and side view samples of rubber products to be tested can be collected from the production line or inventory. Half of each type of sample is a qualified product, and the other half is a clearly marked defective product. Moreover, each type of sample includes at least images taken under a red light source, a green light source, a blue light source, and a composite light source. Then, these samples are labeled to determine whether they are qualified products or defective products. Further, the defect types of the defective products can also be labeled, enabling the model to learn how to classify the defect types of defective products. In addition to determining whether the rubber product to be tested is a qualified product or a defective product, the defect type corresponding to the defective product is output to improve the defect detection accuracy of the system.

[0089] A102. Determine the defect judgment rule, and perform multi-stage training on the pre-constructed convolutional neural network according to the training sample set, detection labels, and the defect judgment rule to obtain a defect detection model.

[0090] Further, as an optional implementation manner, the step of determining the defect judgment rule and performing multi-stage training on the pre-constructed convolutional neural network according to the training sample set, detection labels, and the defect judgment rule to obtain a defect detection model can be specifically further divided into the following steps A1021 to A1027:

[0091] A1021. Determine the defect judgment rule, and then input the defect judgment rule into the convolutional neural network;

[0092] In some optional embodiments, the defects of the rubber product to be tested include, but are not limited to, lack of glue (i.e., the absence of material on the surface or inside of the product), air bubbles (i.e., the formation of cavities inside the product), chipping (i.e., the shedding or breakage of the surface material), cracks (i.e., the generation of fine cracks on the surface or inside of the product, as Figure 5 shown), inclusions (i.e., foreign substances mixed into the inside of the product, as Figure 6 and Figure 7 shown), deformation (i.e., the shape of the product deviates from the design standard), or dimensions (i.e., the dimensions of the product do not meet the specification requirements), and other various defect types.

[0093] For quantifiable features (such as area, gray-scale contrast), clear thresholds can be set as the defect judgment criteria. Exemplarily, set the area critical value of a certain area of the rubber product to be tested as 500, and the gray-scale contrast critical value as 40. Then, when the total area of this area exceeds 500 and the gray-scale contrast exceeds 40, it is defined as a defective product;

[0094] For products with specific shape and size requirements, clear geometric parameter standards can be set, and algorithms can be designed to measure and compare these parameters. For example, the critical value for the roundness of a shock-absorbing pad is defined as 0.4, the upper limit of the inner diameter of a certain area is 9.9, the lower limit of the inner diameter is 9.3, the upper limit of the outer diameter is 27, and the lower limit of the outer diameter is 25. Then, if the maximum and minimum values ​​of the roundness and diameter of the shock-absorbing pad, as determined by comprehensive evaluation, exceed the specified range, it is defined as a defective product.

[0095] For defects that are difficult to determine based on a single condition, multispectral imaging technology can be used. For example, the defect can be determined by taking a picture of area A of the rubber product under test with a red light source and taking a picture of area B with a green light source. When the pictures under different light sources show different image characteristics from qualified products, they are defined as defective products.

[0096] For rubber products whose defects are difficult to determine, such as those containing residual rubber inclusions, they are defined as products awaiting treatment.

[0097] A1022. Input the top view samples and side view samples into the convolutional neural network for the first stage of training, and then randomly select the top view samples and side view samples to test the convolutional neural network after the first stage of training to obtain the first defect detection result.

[0098] A1023. Update the defect judgment rules according to the first defect detection result until the first defect detection result reaches the preset first test condition and the randomly input top view sample and side view sample reach the preset first quantity threshold.

[0099] Specifically, such as Figure 8 The diagram shows the flowchart of the first stage of training. First, a large number of top view and side view samples of rubber products to be tested (half of which are qualified products and half are defective products) are input to train the convolutional neural network. When the defect detection result output by the network is inconsistent with the detection label corresponding to the sample, the defect judgment rule of the network is adjusted, and top view and side view samples are input to train the convolutional neural network. When the defect detection result output by the network is consistent with the detection label corresponding to the sample, it is then determined whether the number of input top view and side view samples has reached a first preset number. When the first preset number is reached, random photos are input to test the convolutional neural network. When the first defect detection result output by the network meets the first test condition, and the number of input random photos reaches the first threshold, the network can be considered to have initially entered the practical stage and enter the next stage of training.

[0100] Those skilled in the art will understand that the first preset quantity, the first test conditions, and the first quantity threshold can be set according to actual needs. For example, the first preset quantity can be 1500 sheets, 1000 sheets, etc., the first test conditions can be that the first defect detection result is consistent with the manual judgment standard, etc., and the first quantity threshold can be 100 sheets, 200 sheets, etc.

[0101] A1024. Input the top view samples and side view samples at the critical point into the convolutional neural network for the second stage of training to obtain the second defect detection result;

[0102] A1025. Update the defect judgment rules based on the second defect detection results and detection labels until the second defect detection results reach the preset second test conditions, and the input top view samples and side view samples at the critical point reach the preset second quantity threshold.

[0103] Specifically, such as Figure 9 The diagram shows the flowchart of the second stage of training. A large number of top view and side view samples of the rubber products to be tested that are at the critical point are input (i.e., under manual inspection, the defect points are not obvious, and inspector A judges them as qualified products while inspector B judges them as defective products). The convolutional neural network is trained. When the second defect detection result output by the network is inconsistent with the detection label corresponding to the sample, the defect judgment rule of the network is adjusted, and the top view and side view samples at the critical point are input again to train the convolutional neural network until the second defect detection result output by the network reaches the second test condition, and the number of top view and side view samples at the critical point reaches the second threshold. Then, the next stage of training begins.

[0104] Those skilled in the art will understand that the second test condition and the second quantity threshold can be set according to actual needs. For example, the second test condition can be that the second defect detection result is consistent with the corresponding detection label or that the second defect detection result is consistent with the manual judgment standard, etc., and the second quantity threshold can be 200 sheets, 100 sheets, etc.

[0105] A1026. Determine the engraving features of the rubber product to be tested, import the engraving features into a preset masking list, and then input the bottom view sample and the masking list into the convolutional neural network for the third stage of training to obtain the third defect detection result.

[0106] A1027. Update the defect judgment rules based on the third defect detection results and detection labels until the third defect detection results reach the preset third test conditions and the input bottom view samples reach the preset third quantity threshold, then stop training and obtain the trained defect detection model.

[0107] It should be noted that the bottom of rubber products usually has markings or numbers indicating the product's specifications, such as "w026", "P2", and "214". "w026" and "P2" are fixed numbers for the rubber product's specifications, while "214" is the mold cavity number for the rubber product's mold, with numbers ranging from 001 to 367. When judging defects in the model, these markings and numbers can easily be confused with defects such as chipping or cracking. Figure 10 The image shows a schematic diagram of the bottom of a rubber product that has developed cracks. Figure 11 The image shown is a schematic diagram of the bottom of a qualified rubber product. Therefore, the above-mentioned engravings and serial numbers need to be included in the model's masking list to avoid the model listing the engravings or serial numbers as defects.

[0108] Specifically, such as Figure 12 The diagram shows the flowchart of the third stage of training. First, the engraved features that need to be blocked are determined, such as the letters "W", "P" and the numbers "1-9" as graphics, and included in the pre-established blocking list in the model. Then, a large number of bottom view samples with engraved features are input to train the convolutional neural network. When the third defect detection result output by the network is inconsistent with the detection label corresponding to the sample, the defect judgment rule of the network is adjusted until the third defect detection result output by the network meets the third test condition and the input bottom view samples reach the preset third quantity threshold, thus obtaining the trained defect detection model.

[0109] Those skilled in the art will understand that the third preset quantity and the third test conditions can be set according to actual needs. For example, the third preset quantity can be 500 sheets, 200 sheets, etc., and the third test conditions can be that the third defect detection result is consistent with the corresponding detection label or that the third defect detection result is consistent with the manual judgment standard, etc.

[0110] As an optional implementation, the step of performing defect detection on the three views using a pre-trained defect detection model within the processor to obtain the defect detection results can be further divided into the following steps S1051 and S1052:

[0111] S1051. Input the top view, bottom view and side view into the defect detection model, and then divide the top view, bottom view and side view into regions to obtain multiple regions to be tested.

[0112] S1052. Perform defect detection on each area to be tested and obtain the defect detection results;

[0113] Among them, defect detection includes at least one of the following: missing glue detection, bubble detection, chipping detection, crack detection, inclusion detection, deformation detection, or dimensional detection.

[0114] Specifically, the three-view diagram of the rubber product to be tested can be divided into regions according to actual testing needs and objectives. For example, for certain critical components or areas, the division can be more detailed for more stringent testing; while for non-critical areas, the testing standards can be appropriately relaxed to reduce the amount of calculation. For example, taking a top view of a rubber shock-absorbing pad as an example... Figure 13 The diagram shown is a schematic representation of a method for dividing the area of ​​a shock-absorbing pad, as illustrated in the top view. Figure 14 The diagram shows another method for dividing the area into top views of the shock-absorbing pad. After the area is divided, the model performs defect detection on each area to be tested according to the determined defect judgment rules. The output defect detection results show whether the rubber product to be tested is a qualified product, a defective product, or a product to be processed.

[0115] As an optional implementation, the defect detection method further includes the step of blowing the rubber product to be tested into the corresponding sorting box through multiple air blowing ports according to the defect detection results. The air blowing ports include a first air blowing port, a second air blowing port, and a third air blowing port. The sorting boxes include a qualified product box, a waste product box, and a pending processing box. The step of blowing the rubber product to be tested into the corresponding sorting box through multiple air blowing ports according to the defect detection results can be further divided into the following steps S1061 to S1063:

[0116] S1061. When the defect detection result shows that the rubber product to be tested is a qualified product, the rubber product to be tested is blown into the qualified product box through the first air blowing port.

[0117] S1062. When the defect detection result shows that the rubber product to be tested is a defective product, the rubber product to be tested is blown into the defective product box through the second air blowing port.

[0118] S1063. When the defect detection result shows that the rubber product to be tested is a product to be processed, the rubber product to be tested is blown into the processing box through the third air blowing port.

[0119] Taking a type of rubber shock-absorbing pad as an example, such as Figure 15 The diagram shows a comparison of the top view, bottom view, and side view of rubber shock-absorbing pad A and rubber shock-absorbing pad B. The following details the workflow of the defect detection method of the rubber product defect detection device according to an embodiment of the present invention:

[0120] First, shock-absorbing pads A and B are conveyed from the loading tray to the glass tray via a belt feeder;

[0121] Next, taking the first detection camera as having one camera, the second detection camera as having one camera, and the third detection camera as having four cameras as an example, the glass disk rotates, and the shock-absorbing pads A and B pass through the first detection camera, the second detection camera, and the third detection camera in sequence, acquiring one top view, one bottom view, and four side views of the shock-absorbing pads A and B.

[0122] Then, the collected photos are input into the pre-trained defect detection model in the processor. The model performs a comprehensive judgment on the six collected photos. Figure 16(a) shows the defect detection result of the top view of shock absorber pad A, Figure 16(b) shows the defect detection result of the top view of shock absorber pad B, Figure 17(a) shows the defect detection result of the bottom view of shock absorber pad A, Figure 17(b) shows the defect detection result of the bottom view of shock absorber pad B, Figure 18(a) shows the defect detection result of the side view of shock absorber pad A, and Figure 18(b) shows the defect detection result of the side view of shock absorber pad B. It can be seen that the model judges all three views of shock absorber pad A as qualified products and all three views of shock absorber pad B as defective products.

[0123] Finally, shock-absorbing pad A is blown into the qualified product box through the first air blowing port, and shock-absorbing pad B is blown into the waste product box through the second air blowing port.

[0124] The contents of the above embodiments of the rubber product defect detection device are all applicable to the defect detection method embodiments of this rubber product defect detection device. The specific functions implemented by the defect detection method embodiments of this rubber product defect detection device are the same as those of the above embodiments of the rubber product defect detection device, and the beneficial effects achieved are also the same as those achieved by the above embodiments of the rubber product defect detection device.

[0125] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the aforementioned blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0126] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the aforementioned functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0127] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0129] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the aforementioned program can be printed, because the aforementioned program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or, if necessary, processing in other suitable ways, and then stored in computer memory.

[0130] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0131] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0132] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0133] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A defect detection method for a rubber product defect detection device, used to be executed by the rubber product defect detection device, characterized in that, The rubber product defect detection device includes a detection and sorting device and a processor. The detection and sorting device includes a loading tray, a belt feeder, and a worktable. The worktable includes a glass plate and multiple sets of detection cameras. One end of the belt feeder is connected to the loading tray, and the other end is connected to the glass plate. Each set of detection cameras is located above or below the glass plate, and each set of detection cameras is also connected to the processor. The processor integrates a pre-trained defect detection model. The loading tray is used to place the rubber product to be tested. The belt feeder is used to transport the rubber product to be tested from the loading tray to the glass plate. The glass plate is used to allow the rubber product to be tested to pass through each set of detection cameras in sequence. Each set of detection cameras is used to acquire three views of the rubber product to be tested. The defect detection model is used to perform defect detection on the three views to obtain defect detection results. The three views include a top view, a bottom view, and a side view. The defect detection method includes the following steps: Place the rubber product to be tested onto the loading tray; The rubber product to be tested is conveyed from the loading tray to the glass tray by a belt feeder; The rubber product to be tested is passed sequentially through multiple sets of detection cameras via the glass disk; The three views of the rubber product under test are acquired by multiple sets of detection cameras, including a top view, a bottom view, and a side view. Defect detection results are obtained by performing defect detection on the three views using a pre-trained defect detection model within the processor. The defect detection method further includes a step of pre-training the defect detection model, wherein the pre-training of the defect detection model specifically includes: Obtain a training sample set, which includes multiple top view samples, bottom view samples, side view samples of the rubber products to be tested, and corresponding detection labels; Determine the defect judgment rules, and then input the defect judgment rules into the convolutional neural network; The top view sample and the side view sample are input into the convolutional neural network for the first stage of training. Then, the top view sample and the side view sample are randomly selected to test the convolutional neural network after the first stage of training to obtain the first defect detection result. The defect determination rule is updated based on the first defect detection result until the first defect detection result reaches the preset first test condition, and the randomly input top view sample and side view sample reach the preset first quantity threshold. The top view sample and the side view sample at the critical point are input into the convolutional neural network for the second stage of training to obtain the second defect detection result. The defect determination rule is updated based on the second defect detection result and the detection label until the second defect detection result reaches the preset second test condition, and the input top view sample and side view sample at the critical point reach the preset second quantity threshold. The engraving features of the rubber product to be tested are determined, the engraving features are imported into a preset masking list, and then the bottom view sample and the masking list are input into the convolutional neural network for the third stage of training to obtain the third defect detection result. The defect determination rules are updated based on the third defect detection result and the detection label until the third defect detection result reaches the preset third test condition and the input bottom view sample reaches the preset third quantity threshold. Then, training stops, and the trained defect detection model is obtained.

2. The defect detection method of the rubber product defect detection device according to claim 1, characterized in that, The detection and sorting equipment also includes multiple sorting boxes, and the workbench also includes multiple air inlets. Each air inlet is located on the glass plate, and each sorting box is located below each air inlet. Each air inlet is also connected to the processor. The air inlets are used to blow the rubber product to be tested into the corresponding sorting box according to the defect detection result.

3. The defect detection method of the rubber product defect detection device according to claim 1, characterized in that, The detection camera includes a first detection camera, a second detection camera, and a third detection camera. The first detection camera is located above the glass plate, the second detection camera is located below the glass plate, and the third detection camera is placed on the glass plate. The first detection camera is used to capture the top view of the rubber product to be tested, the second detection camera is used to capture the bottom view of the rubber product to be tested, and the third detection camera is used to capture the side view of the rubber product to be tested.

4. The defect detection method of the rubber product defect detection device according to claim 1, characterized in that, The belt feeder includes a lifting belt and a feeding belt. One end of the lifting belt is connected to the bottom of the loading tray, and the other end of the lifting belt is connected to one end of the feeding belt. The other end of the feeding belt is connected to the edge of the glass tray. The lifting belt is used to transport the rubber product to be tested from the loading tray to the feeding belt, and the feeding belt is used to transport the rubber product to be tested from the lifting belt to the glass tray.

5. The defect detection method of the rubber product defect detection device according to claim 2, characterized in that, The air inlet includes a first air inlet, a second air inlet, and a third air inlet. The sorting box includes a qualified product box, a waste product box, and a pending processing box. The qualified product box is located below the first air inlet, the waste product box is located below the second air inlet, and the pending processing box is located below the third air inlet. The first air inlet, the second air inlet, and the third air inlet are used to blow the rubber product to be tested into the qualified product box, the waste product box, or the pending processing box according to the defect detection result.

6. The defect detection method of the rubber product defect detection device according to claim 1, characterized in that, The step of performing defect detection on the three views using a pre-trained defect detection model within the processor to obtain defect detection results specifically includes: The top view, the bottom view, and the side view are input into the defect detection model, and then the top view, the bottom view, and the side view are divided into regions to obtain multiple regions to be tested. Defect detection is performed on each of the areas to be tested to obtain the defect detection results; The defect detection includes at least one of the following: glue deficiency detection, bubble detection, chipping detection, crack detection, inclusion detection, deformation detection, or dimensional detection.

7. The defect detection method of the rubber product defect detection device according to claim 1, characterized in that, The defect detection method further includes the step of blowing the rubber product to be tested into a corresponding sorting box through multiple air outlets based on the defect detection results. The air outlets include a first air outlet, a second air outlet, and a third air outlet. The sorting box includes a qualified product box, a waste product box, and a pending processing box. Specifically, the step of blowing the rubber product to be tested into the corresponding sorting box through multiple air outlets based on the defect detection results includes: When the defect detection result shows that the rubber product to be tested is a qualified product, the rubber product to be tested is blown into the qualified product box through the first air blowing port; When the defect detection result shows that the rubber product to be tested is a waste product, the rubber product to be tested is blown into the waste product box through the second air blowing port; When the defect detection result shows that the rubber product to be tested is a product to be processed, the rubber product to be tested is blown into the processing box through the third air blowing port.

Images