Material joint overlap amount detection system, method, device and equipment and medium

Abstract

The embodiment of the invention relates to a system, a method, a device and equipment for detecting the overlap amount of a material joint, and a medium, which are used for improving the efficiency and the precision of detecting the overlap amount of the material joint. The system is applied to a tire building machine, a to-be-measured material is attached to a belt drum / building drum of the tire building machine, the system comprises a laser, a depth camera and a processing device, the laser is used for emitting laser and projecting the laser to the belt drum / building drum, and the projection direction of the laser is parallel to the axis direction of the belt drum / building drum; the depth camera is used for shooting a laser image on the belted drum / forming drum in the laminating process of the to-be-detected material and acquiring a depth image of the belted drum / forming drum in the laminating process of the to-be-detected material; the processing equipment is used for acquiring a depth image, acquired by the depth camera, of the belt drum / forming drum in the laminating process of the to-be-detected material; and determining the joint overlap amount of the to-be-measured material according to the obtained depth image.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to a system, method, device, equipment and medium for detecting the overlapping amount of material joints. Background Art

[0002] The fitting of materials is a very important link in the tire forming process, and the overlapping amount of material joints is one of the process parameters that need to be monitored keyly. In the prior art, the detection of the overlapping amount of material joints is realized by manual measurement, with low efficiency and poor accuracy. Summary of the Invention

[0003] Embodiments of this application provide a system, method, device, equipment and medium for detecting the overlapping amount of material joints, so as to improve the efficiency and accuracy of detecting the overlapping amount of material joints.

[0004] In a first aspect, embodiments of this application provide a system for detecting the overlapping amount of material joints, which is applied to a tire forming machine, and the material to be measured is fitted on the belt drum of the tire forming machine. The system includes a laser, a depth camera and a processing device, wherein: The laser is used to emit laser light and project it onto the belt drum, and the projection direction of the laser is parallel to the axis direction of the belt drum; The depth camera is used to capture the laser image on the belt drum during the fitting process of the material to be measured, and collect the depth image of the belt drum during the fitting process of the material to be measured; The processing device is used to obtain the depth image of the belt drum during the fitting process of the material to be measured collected by the depth camera; perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured; input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; determine the overlapping amount of the joints of the material to be measured according to the head edge and tail edge of the material to be measured.

[0005] In a possible implementation manner, the installation positions of the laser, the depth camera and the belt drum form a triangle, and the angle of the angle corresponding to the belt drum is greater than or equal to 20 degrees and less than or equal to 50 degrees.

[0006] In a possible implementation manner, the pre-trained lightweight model is an ONNX model.

[0007] In a possible implementation manner, the ONNX model is obtained in the following manner: Replace the backbone network of the EdgeNAT network to obtain a lightweight model to be trained; Use training samples to train the lightweight model to be trained to obtain an ONNX model.

[0008] In a possible implementation manner, the training samples are obtained in the following manner: Obtain an image of the sample joint area of the material to be measured; Use the labelme annotation tool to annotate the head edge and tail edge of the material to be measured in the image of the sample joint area of the material to be measured according to the gradient change of the depth information; Perform data cleaning on the annotated image of the sample joint area to obtain training samples.

[0009] In a second aspect, an embodiment of the present application provides a material joint overlap amount detection system, which is applied to a tire building machine. The material to be measured is bonded on the building drum of the tire building machine. The system includes a laser, a depth camera, and a processing device, where: The laser is used to emit laser light projected onto the building drum, and the projection direction of the laser is parallel to the axis direction of the building drum; The depth camera is used to capture the laser image on the building drum during the bonding process of the material to be measured, and collect the depth image of the building drum during the bonding process of the material to be measured; The processing device is used to obtain the depth image of the building drum during the bonding process of the material to be measured collected by the depth camera; perform image segmentation on the obtained depth image to obtain the image of the joint area of the material to be measured; input the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

[0010] In a possible implementation manner, the installation positions of the laser, the depth camera, and the building drum form a triangle, and the angle of the angle corresponding to the building drum is greater than or equal to 20 degrees and less than or equal to 50 degrees.

[0011] In a possible implementation manner, the pre-trained lightweight model is an ONNX model.

[0012] In a possible implementation manner, the ONNX model is obtained in the following manner: Replace the backbone network of the EdgeNAT network to obtain a lightweight model to be trained; Use the training samples to train the lightweight model to be trained to obtain the ONNX model.

[0013] In a possible implementation manner, the training samples are obtained in the following manner: Obtain an image of the sample joint area of the material to be measured; Use the labelme annotation tool to annotate the head edge and tail edge of the material to be measured in the image of the sample joint area of the material to be measured according to the gradient change of the depth information; Perform data cleaning on the annotated image of the sample joint area to obtain training samples.

[0014] In a third aspect, an embodiment of the present application provides a method for detecting the overlapping amount of a material joint, which is applied to a tire building machine. The material to be measured is bonded on the belt drum of the tire building machine. The method includes: Obtaining a depth image of the belt drum during the bonding process of the material to be measured; Performing image segmentation on the obtained depth image to obtain an image of the joint area of the material to be measured; Inputting the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; Determining the overlapping amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0015] In a fourth aspect, an embodiment of the present application provides a method for detecting the overlapping amount of a material joint, which is applied to a tire building machine. The material to be measured is bonded on the building drum of the tire building machine. The method includes: Obtaining a depth image of the building drum during the bonding process of the material to be measured; Performing image segmentation on the obtained depth image to obtain an image of the joint area of the material to be measured; Inputting the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; Determining the overlapping amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0016] In a fifth aspect, an embodiment of the present application provides a device for detecting the overlapping amount of a material joint, which is applied to a tire building machine. The material to be measured is bonded on the belt drum of the tire building machine. The device includes: An acquisition unit for obtaining a depth image of the belt drum during the bonding process of the material to be measured; A determination unit for performing image segmentation on the obtained depth image to obtain an image of the joint area of the material to be measured; inputting the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determining the overlapping amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0017] In a sixth aspect, an embodiment of the present application provides a device for detecting the overlapping amount of a material joint, which is applied to a tire building machine. The material to be measured is bonded on the building drum of the tire building machine. The device includes: An acquisition unit for obtaining a depth image of the belt drum during the bonding process of the material to be measured; A determination unit is configured to perform image segmentation on the acquired depth image to obtain an image of the joint area of the material to be measured; input the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

[0018] In a seventh aspect, an embodiment of the present application provides a processing device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which when executed by the processor implement the method provided in the third aspect or the fourth aspect of the embodiment of the present application.

[0019] In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored, which when executed by the processor implement the method provided in the third aspect or the fourth aspect of the embodiment of the present application.

[0020] For the material joint overlap amount detection system, method, device, equipment and medium provided in the embodiments of the present application, if the material to be measured is attached to the belt drum of the tire building machine, the depth image of the belt drum during the attachment of the material to be measured is acquired; if the material to be measured is attached to the forming drum of the tire building machine, the depth image of the forming drum during the attachment of the material to be measured is acquired, and then the joint overlap amount of the material to be measured is determined according to the acquired depth image, realizing the visual detection of the joint overlap amount of the material, and improving the detection efficiency and accuracy.

[0021] Other features and advantages of the present application will be described in the following specification, and part of them will become obvious from the specification, or will be understood by implementing the present application. The purpose and other advantages of the present application can be realized and obtained through the structures specifically pointed out in the written specification, claims, and drawings. Description of the Drawings

[0022] Figure 1 It is a schematic diagram of a material joint overlap amount detection system provided by an embodiment of the present application; Figure 2 It is one of the schematic diagrams of the laser projection direction on the belt drum provided by an embodiment of the present application; Figure 3 It is another schematic diagram of the laser projection direction on the belt drum provided by an embodiment of the present application; Figure 4 It is one of the schematic diagrams of the depth image provided by an embodiment of the present application; Figure 5 It is another schematic diagram of the depth image provided by an embodiment of the present application; Figure 6 It is another schematic diagram of the depth image provided by an embodiment of the present application; Figure 7 Schematic diagram of another material joint overlap amount detection system provided by an embodiment of the present application; Figure 8 Flow schematic diagram of a material joint overlap amount detection method provided by an embodiment of the present application; Figure 9 Flow schematic diagram of another material joint overlap amount detection method provided by an embodiment of the present application; Figure 10 Structural schematic diagram of a material joint overlap amount detection device provided by an embodiment of the present application; Figure 11 Structural schematic diagram of a processing device provided by an embodiment of the present application. Detailed implementation manners

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, rather than all, of the embodiments of the technical solutions of the present application. Based on the embodiments described in this application document, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the technical solutions of the present application.

[0024] In the prior art, during the tire forming process, the detection of the overlap amount of the joints of the materials is realized by manual measurement, with low efficiency and poor accuracy. In view of this, the embodiments of the present application provide a material joint overlap amount detection system, method, device, equipment, and medium, which are applied to a tire forming machine and realize visual detection of the overlap amount of the joints of the materials. Compared with the prior art, it does not require manual participation and improves the detection efficiency and accuracy.

[0025] The following describes the preferred embodiments of the present application with reference to the accompanying drawings of the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application. And without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0026] As Figure 1 shown, it is a schematic diagram of a material joint overlap amount detection system in an embodiment of the present application. This system is applied to a tire forming machine and includes a laser 101, a depth camera 102, and a processing device 103.

[0027] The material joint overlap amount detection system provided by the embodiments of the present application can be specifically applied to the belt drum 100 of a tire forming machine. The material to be measured is bonded on the belt drum 100 of the tire forming machine. For example, the belt layer and the tread are bonded on the belt drum 100 of the tire forming machine. At this time: A laser 101 is used to emit a laser beam that is projected onto the belt drum 100, and the projection direction of the laser beam is parallel to the axis direction of the belt drum 100. As shown in Figure 2 the figure, the thick line is the laser beam projected from the laser 101 onto the belt drum 100; Figure 3 The figure shows a top view of the laser 101 and the belt drum 100, and the laser incident surface is as shown in the figure; A depth camera 102 is used to capture the laser image on the belt drum 100 during the fitting process of the material to be measured, and to acquire the depth image of the belt drum 100 during the fitting process of the material to be measured; A processing device 103 is used to obtain the depth image of the belt drum 100 during the fitting process of the material to be measured collected by the depth camera 102; perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured; input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

[0028] In actual implementation, as shown in Figure 1 the figure, the installation positions of the laser 101, the depth camera 102, and the belt drum 100 form a triangle. Preferably, the angle of the angle corresponding to the belt drum 100 is greater than or equal to 20 degrees and less than or equal to 50 degrees, and the horizontal distances between the belt drum 100 and the laser 101 and the depth camera 102 can be adjusted according to the actual scenario.

[0029] The material joint overlap amount detection system provided by the embodiment of the present application can also be specifically applied to the forming drum of a tire building machine. The material to be measured is fitted on the forming drum of the tire building machine. For example, a composite layer and a ply are fitted on the forming drum of the tire building machine. At this time: A laser 101 is used to emit a laser beam that is projected onto the forming drum, and the projection direction of the laser beam is parallel to the axis direction of the forming drum; the specific projection direction of the laser 101 on the forming drum can refer to the illustration of the projection direction of the laser 101 on the belt drum 100 described above; A depth camera 102 is used to capture the laser image on the forming drum during the fitting process of the material to be measured, and to acquire the depth image of the forming drum during the fitting process of the material to be measured; A processing device 103 is used to obtain the depth image of the forming drum during the fitting process of the material to be measured collected by the depth camera 102; perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured; input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

[0030] In actual implementation, the installation positions of the laser 101, the depth camera 102, and the forming drum form a triangle. Preferably, the angle of the angle corresponding to the forming drum is greater than or equal to 20 degrees and less than or equal to 50 degrees, and the horizontal distances between the forming drum and the laser 101 and the depth camera 102 can be adjusted according to the actual scenario.

[0031] In actual implementation, an encoder can be used to give a signal to control the depth camera 102 to perform image acquisition. The material fitting position is fixed, and the signal emission time for starting the acquisition of the depth camera 102 is fixed through the encoder. The floating change of the joint position of the material to be measured in the finally acquired depth image is small, which can ensure the accuracy of the subsequent joint overlap amount detection.

[0032] After the depth camera 102 acquires the depth image, the depth image is sent to the processing device 103. After the processing device 103 obtains the depth image, the joint overlap amount of the material to be measured is detected according to the depth image.

[0033] First, the processing device 103 needs to first determine the head edge and the tail edge of the material to be measured.

[0034] The depth image obtained by the processing device 103 includes a material area and a non-material area. The non-material area is the drum surface area, and the material area is the non-drum surface area.

[0035] In actual implementation, the processing device 103 can, in the obtained depth image, first extract the material area according to the difference between the drum surface and the material. The extracted material area is as shown by the dotted line box in Figure 4 ; then perform image segmentation on the material area image to obtain the joint area image of the material to be measured as shown by the dotted line box in Figure 5 ; specifically, based on the prior data of each on-site machine, the segmentation position of the joint area can be determined, and then image segmentation is performed to segment out the joint area image.

[0036] It should be noted that in the embodiments of the present application, the head edge of the material to be measured refers to the head edge in the joint area of the material to be measured.

[0037] In some embodiments of the present application, the processing device 103 can input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and the tail edge of the material to be measured.

[0038] After the processing device 103 determines the head edge and the tail edge of the material to be measured, the joint overlap amount of the material to be measured can be determined.

[0039] In some embodiments of the present application, such as Figure 6As shown, for each column in at least one column of the joint area of the material to be measured, the processing device 103 can determine the joint overlap amount d of this column of the material to be measured according to the determined head edge point and tail edge point of the material to be measured. That is to say, for each column in at least one column of the joint area of the material to be measured, a joint overlap amount d will be determined. The positive or negative of the joint overlap amount d can indicate whether the joint is overlapped or not properly connected.

[0040] For each column in at least one column of the joint area of the material to be measured, after determining a joint overlap amount d, according to the actual scenario requirements, the average value of the joint overlap amounts d determined for each column can be calculated to obtain the average joint overlap amount of the material to be measured.

[0041] In some embodiments of the present application, as Figure 7 shown, the system may further include an alarm device 104. After the processing device 103 determines the joint overlap amount of the material to be measured, the determined joint overlap amount of the material to be measured can be compared with the allowable overlap amount. If the comparison result indicates that the joint overlap amount of the material to be measured does not meet the requirements, the processing device 103 can give a prompt alarm through the alarm device 104.

[0042] Specifically, the joint overlap amounts of each column of the material to be measured can be compared with the allowable overlap amount. If a certain proportion of the joint overlap amounts do not meet the requirements, it is determined that the joint overlap amount of the material to be measured does not meet the requirements, and the processing device 103 gives a prompt alarm through the alarm device 104.

[0043] Specifically, the average joint overlap amount of the material to be measured can also be compared with the allowable overlap amount. If the average joint overlap amount does not meet the requirements, it is determined that the joint overlap amount of the material to be measured does not meet the requirements, and the processing device 103 gives a prompt alarm through the alarm device 104.

[0044] Next, taking the material to be measured as the second belt layer as an example, the material overlap amount detection method performed by the processing device 103 in the embodiments of the present application will be illustrated by way of example.

[0045] When the material to be measured is the second belt layer, the first belt layer is the first layer of material laminated on the belt drum 100, and the second belt layer is the second layer of material laminated on the belt drum 100, and the width of the second belt layer is narrower than that of the first belt layer. Therefore, the depth image of the belt drum 100 during the lamination process of the second belt layer collected by the depth camera 102 is as Figures 4 - 6 shown.

[0046] First, the material area is extracted from the depth image, and the extracted material area is as Figure 4 shown by the dashed box; then the material area image is segmented to obtain the joint area image of the second belt layer as Figure 5as shown by the dashed box in the figure; input the image of the joint area of the second belt layer into a pre-trained lightweight model to obtain the leading edge and trailing edge of the stock of the second belt layer.

[0047] For each column in the sampling columns of the joint area, determine the joint overlap amount of the second belt layer based on the leading edge point and trailing edge point of the second belt layer determined by this column, and then judge whether the joint overlap amount of the second belt layer meets the requirements. When it is determined that the joint overlap amount of the second belt layer does not meet the requirements, a prompt alarm is given.

[0048] The following gives an example of the specific training process of the lightweight model.

[0049] First, training samples need to be obtained. In actual implementation, the training samples can be obtained in the following specific way: Step 1: Obtain some depth images of the belt drum or forming drum during the fitting process of the material to be measured as sample depth images.

[0050] Step 2: Extract the sample material area in the sample depth image.

[0051] In actual implementation, the sample material area can be extracted in the sample depth image according to the difference between the drum surface and the material.

[0052] Step 3: Perform image segmentation on the sample material area image to obtain the sample joint area image of the material to be measured.

[0053] In actual implementation, based on the prior data, the segmentation position of the sample joint area can be determined, and then image segmentation is performed to segment out the sample joint area image.

[0054] Step 4: Use the labelme annotation tool to annotate the leading edge and trailing edge of the material to be measured in the sample joint area image of the material to be measured.

[0055] In actual implementation, the annotation of the leading edge and trailing edge of the material to be measured can be realized according to the gradient change of the depth information.

[0056] Generally, there are two situations for the leading edge. One is that it is not covered by the trailing edge, which is called a virtual joint; the other is that it is covered by the trailing edge, which is called an overlap. Therefore, when annotating the leading edge, different situations need to be considered for annotation. When the leading edge is not covered by the trailing edge, the position of the maximum value of the gradient change of the depth information can be directly taken as the leading edge for annotation; when the leading edge is covered by the trailing edge, the material at the covered part is relatively thick, and the corresponding depth value is relatively large. Therefore, the position where the depth information starts to change from high to low is taken as the leading edge for annotation.

[0057] The labeling of the tail edge of the material is relatively simple, without the situation of being covered, and the gradient change direction of the depth information of the tail and the head of the material is opposite. Therefore, when labeling the tail edge of the material, only the position of the maximum gradient change in different gradient directions needs to be found as the tail edge for labeling.

[0058] Step 5: Clean the data of the labeled sample joint area image to obtain training samples.

[0059] Through data cleaning, unqualified data can be removed, and only qualified data is retained as training samples.

[0060] After obtaining the training samples, model training is carried out. In actual implementation, the following method can be used to obtain a trained lightweight model: Step 1: Modify the EdgeNAT network to obtain a lightweight model to be trained.

[0061] In actual implementation, the backbone network of the EdgeNAT network can be replaced to obtain a lightweight model to be trained.

[0062] Step 2: Use the training samples to train the lightweight model to be trained to obtain a trained lightweight model.

[0063] In actual implementation, the training samples can be used to continuously iterate and optimize the lightweight model to be trained, and the optimized model can be exported as an ONNX model as the trained lightweight model.

[0064] Then, the ONNX model can be parsed using C++ DNN and deployed to the detection software.

[0065] It should be noted that the specific training process of the above lightweight model is only an example and is not used to limit the protection scope of the present invention.

[0066] In summary, the material joint overlap amount detection system provided by the embodiments of the present application realizes the visual detection of the joint overlap amount of the material, improves the detection efficiency and accuracy, ensures the material fitting quality, and thus ensures the tire production quality.

[0067] Moreover, the material joint overlap amount detection system provided by the embodiments of the present application can meet the requirements of real-time detection through the lightweight model and image preprocessing segmentation. For new application scenarios, only some data of the same type needs to be simply trained to achieve high-precision detection of the joint overlap amount.

[0068] Based on the same inventive concept, as Figure 8As shown in the figure, an embodiment of the present application provides a method for detecting the overlapping amount of a material joint, which is applied to a tire building machine, specifically to the belt drum of the tire building machine. The material to be measured is bonded on the belt drum of the tire building machine. The method includes the following steps: S801. Obtain the depth image of the belt drum during the bonding process of the material to be measured.

[0069] S802. Perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured.

[0070] S803. Input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured.

[0071] S804. Determine the overlapping amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0072] S805. Judge whether the overlapping amount of the joint of the material to be measured meets the requirements according to the determined overlapping amount of the joint of the material to be measured.

[0073] Further, when it is determined that the overlapping amount of the joint of the material to be measured does not meet the requirements, a prompt alarm is given.

[0074] Based on the same inventive concept, as Figure 9 shown in the figure, an embodiment of the present application provides a method for detecting the overlapping amount of a material joint, which is applied to a tire building machine, specifically to the building drum of the tire building machine. The material to be measured is bonded on the building drum of the tire building machine. The method includes the following steps: S901. Obtain the depth image of the building drum during the bonding process of the material to be measured.

[0075] S902. Perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured.

[0076] S903. Input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured.

[0077] S904. Determine the overlapping amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0078] S905. Judge whether the overlapping amount of the joint of the material to be measured meets the requirements according to the determined overlapping amount of the joint of the material to be measured.

[0079] Further, when it is determined that the overlapping amount of the joint of the material to be measured does not meet the requirements, a prompt alarm is given.

[0080] For the specific implementation of each of the above steps, reference can be made to the foregoing system embodiment, which will not be elaborated here.

[0081] It should be noted that although the steps of the method of the present application are described in a specific order in the accompanying drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0082] Based on the same inventive concept, as Figure 10 shown, an embodiment of the present application provides a device for detecting the lap amount of a material joint, which is applied to a tire building machine. The device includes an acquisition unit 1001 and a determination unit 1002.

[0083] In some embodiments of the present application, the device is specifically applied to the belt drum of a tire building machine, and the material to be measured is applied on the belt drum of the tire building machine. At this time: The acquisition unit 1001 is configured to acquire a depth image of the belt drum during the application process of the material to be measured; The determination unit 1002 is configured to perform image segmentation on the acquired depth image to obtain an image of the joint area of the material to be measured; input the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the lap amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0084] In some embodiments of the present application, the device is specifically applied to the building drum of a tire building machine, and the material to be measured is applied on the building drum of the tire building machine. At this time: The acquisition unit 1001 is configured to acquire a depth image of the building drum during the application process of the material to be measured; The determination unit 1002 is configured to perform image segmentation on the acquired depth image to obtain an image of the joint area of the material to be measured; input the image of the joint area of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the lap amount of the joint of the material to be measured according to the head edge and tail edge of the material to be measured.

[0085] Wherein, the above-mentioned pre-trained lightweight model is an ONNX model.

[0086] In some embodiments of the present application, the ONNX model is obtained in the following manner: Replace the backbone network of the EdgeNAT network to obtain a lightweight model to be trained; Use training samples to train the lightweight model to be trained to obtain an ONNX model.

[0087] In some embodiments of the present application, the training samples are obtained in the following manner: Obtain an image of the sample joint area of the material to be measured; Using the labelme annotation tool, according to the gradient change of the depth information, annotate the head edge and tail edge of the material to be measured in the image of the sample joint area of the material to be measured; Perform data cleaning on the annotated image of the sample joint area to obtain training samples.

[0088] For the specific implementation of each of the above units, reference may be made to the foregoing system embodiments, which will not be elaborated herein.

[0089] It should be noted that the unit division of the above device is merely exemplary and not mandatory. In fact, according to the implementation manners of the present application, the features and functions of the above two or more units may be embodied in one unit. Conversely, the features and functions of the above one unit may be further divided and embodied by multiple units.

[0090] Based on the same inventive concept as the above method embodiment, an embodiment of the present application also provides a processing device. In this embodiment, the structure of the processing device may be as Figure 11 shown, including at least one memory 1101, a communication module 1103, and at least one processor 1102.

[0091] The memory 1101 is used to store computer program instructions executed by the processor 1102. The memory 1101 may be a volatile memory, such as a random-access memory (RAM); the memory 1101 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD), but not limited thereto. The memory 1201 may be a combination of the above memories.

[0092] The processor 1102 may include one or more central processing units (CPUs) or be a digital processing unit, etc. The processor 1102 is used to implement the above method for detecting the overlapping amount of the material joints when calling the computer program instructions stored in the memory 1101.

[0093] The communication module 1103 is used to communicate with other devices.

[0094] In the embodiments of the present application, the specific connection medium between the above memory 1101, communication module 1103, and processor 1102 is not limited. The embodiments of the present application are Figure 11The connection between the memory 1101, the communication module 1103 and the processor 1102 via the bus 1104 is only for illustrative purposes and is not limiting.

[0095] In some possible implementation manners, various aspects of the material joint overlap amount detection method provided by this application can also be implemented in the form of a program product, which includes computer program instructions that implement the above-mentioned material joint overlap amount detection method when executed by a processor.

[0096] The program product can adopt any combination of one or more computer-readable storage media. In the embodiments of this application, the computer-readable storage media can be any tangible medium that contains or stores computer program instructions.

[0097] Although the preferred embodiments of this application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be interpreted to include the preferred embodiments and all changes and modifications that fall within the scope of this application.

[0098] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. A detection system for the overlapping amount of material joints, which is applied to a tire building machine, and the material to be measured is bonded on the belt drum / forming drum of the tire building machine, and is characterized in that The system includes a laser, a depth camera, and a processing device, where: The laser is used to emit laser light and project it onto the belt drum / forming drum, and the projection direction of the laser light is parallel to the axis direction of the belt drum / forming drum; The depth camera is used to capture the laser image on the belt drum / forming drum during the fitting process of the material to be measured, and acquire the depth image of the belt drum / forming drum during the fitting process of the material to be measured; The processing device is used to obtain the depth image of the belt drum / forming drum during the fitting process of the material to be measured acquired by the depth camera; perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured; input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

2. The system according to claim 1, wherein The pre-trained lightweight model is an ONNX model.

3. The system according to claim 2, wherein The ONNX model is obtained in the following manner: Replace the backbone network of the EdgeNAT network to obtain a lightweight model to be trained; Use training samples to train the lightweight model to be trained to obtain the ONNX model.

4. The system according to claim 3, characterized in that, The training samples are obtained in the following manner: Obtain the sample joint area image of the material to be measured; Use the labelme annotation tool to annotate the head edge and tail edge of the material to be measured in the sample joint area image of the material to be measured according to the gradient change of the depth information; Perform data cleaning on the annotated sample joint area image to obtain the training samples.

5. A method for detecting the overlapping amount of a material joint, which is applied to a tire building machine, and the material to be measured is bonded on a belt drum / forming drum of the tire building machine, and is characterized in that The method includes: Obtain the depth image of the belt drum / forming drum during the fitting process of the material to be measured; Perform image segmentation on the obtained depth image to obtain the joint area image of the material to be measured; Input the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; Determine the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

6. The method according to claim 5, characterized in that, The pre-trained lightweight model is an ONNX model.

7. The method according to claim 6, characterized in that, The ONNX model is obtained in the following manner: Replace the backbone network of the EdgeNAT network to obtain a lightweight model to be trained; Use training samples to train the lightweight model to be trained to obtain the ONNX model.

8. The method according to claim 7, characterized in that, The training samples are obtained in the following manner: Obtain the sample joint area image of the material to be measured; Use the labelme annotation tool to annotate the head edge and tail edge of the material to be measured in the sample joint area image of the material to be measured according to the gradient change of the depth information; Perform data cleaning on the annotated sample joint area image to obtain the training samples.

9. A device for detecting the overlapping amount of a material joint, which is applied to a tire building machine, and the material to be measured is bonded on a belt drum / forming drum of the tire building machine, and is characterized in that The device includes: An acquisition unit for acquiring the depth image of the belt drum / forming drum during the fitting process of the material to be measured; A determination unit for performing image segmentation on the acquired depth image to obtain the joint area image of the material to be measured; inputting the joint area image of the material to be measured into a pre-trained lightweight model to obtain the head edge and tail edge of the material to be measured; and determining the joint overlap amount of the material to be measured according to the head edge and tail edge of the material to be measured.

10. A processing device, characterized in that, Includes: At least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method according to any one of claims 5-8.

11. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by a processor, the method according to any one of claims 5-8 is implemented.

Images