Weld detection method, device, industrial robot, and storage medium

Abstract

The embodiment of the present invention discloses a weld detection method, device, industrial robot, and storage medium, the method comprising: obtaining a weld image of a weld segment to be detected; inputting the weld image into a pre-trained weld detection model, obtaining output result information of the weld detection model, and determining the standard compliance of the weld displayed by the weld image. The technical solution of the embodiment of the present invention can improve the efficiency of weld detection.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of processing and manufacturing technology, and specifically to a weld detection method, device, industrial robot, and storage medium. Background Art

[0002] With the development of electronic technology, computer technology, numerical control and robotics technology, automatic welding robots have been used in production since the 1960s. Their technology has become increasingly mature and has been widely used in all walks of life.

[0003] In terms of welding operation, welding robots replace manual welding by workers, which frees workers from smoke, radiation and dangerous working environment, and improves welding efficiency. However, in terms of welding quality inspection, the welding result inspection of the thermal processing industry still relies on the workers' intuition and experience to judge whether it is qualified, which is inefficient. Summary of the invention

[0004] In view of this, embodiments of the present invention provide a weld detection method, device, industrial robot, and storage medium to improve weld detection efficiency.

[0005] Other features and advantages of the embodiments of the present invention will become apparent from the following detailed description, or may be learned in part by the practice of the embodiments of the present invention.

[0006] In a first aspect of the present disclosure, an embodiment of the present invention provides a weld detection method, comprising:

[0007] Acquire a weld image of a weld segment to be inspected;

[0008] The weld image is input into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0009] In one embodiment, acquiring a weld image of a weld segment to be detected includes: during a welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, acquiring a weld image of the weld segment to be detected.

[0010] In one embodiment, after obtaining the output result information of the weld detection model, it also includes, if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, issuing an alarm and / or correcting the weld segment to be detected.

[0011] In one embodiment, obtaining a weld image of a weld segment to be detected includes: obtaining a surface contour image of a product to be detected, performing image cutting on the surface contour image to obtain multiple weld images corresponding to multiple weld segments; inputting the weld image into a pre-trained weld detection model, and obtaining output result information of the weld detection model includes: respectively inputting the multiple weld images into the pre-trained weld detection model, and respectively obtaining output result information of the weld detection model.

[0012] In one embodiment, after inputting any weld image among the multiple weld images into the pre-trained weld detection model to obtain the output result information of the weld detection model, it also includes: if the output result information of the weld image determines that the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, the weld segment corresponding to the weld image is recorded as unqualified.

[0013] In one embodiment, the method further includes, if there is an unqualified record in a weld segment corresponding to any weld image cut from the surface contour image of the product to be inspected, determining to perform a repair process on the product to be inspected.

[0014] In one embodiment, obtaining the surface contour image of the product to be inspected includes: scanning the product to be inspected by an industrial robot equipped with a 3D laser scanner to obtain the surface contour image of the product to be inspected;

[0015] Performing image cutting on the surface contour image to obtain a plurality of weld images corresponding to a plurality of weld segments includes: segmentally intercepting the surface contour image of the product to be inspected to obtain a plurality of weld images corresponding to a plurality of weld segments.

[0016] In one embodiment, the weld detection model is trained by the following steps:

[0017] Acquire a training sample set, wherein the training sample includes a weld sample image and a label used to indicate whether the weld displayed by the weld sample image is qualified;

[0018] Determining an initialized weld inspection model, wherein the initialized weld inspection model includes a target layer for outputting a standard compliance of a weld displayed in a weld image;

[0019] By using a machine learning method, the weld sample images in the training samples in the training sample set are used as the input of the initialized weld detection model, and the annotation information corresponding to the input weld sample images is used as the expected output of the initialized weld detection model, so as to train and obtain the weld detection model.

[0020] In one embodiment, determining the initialized weld detection model includes: determining that the initialized weld detection model is a three-layer neural network model.

[0021] In a second aspect of the present disclosure, an embodiment of the present invention further provides a weld detection device, comprising:

[0022] A weld image acquisition unit, used to acquire a weld image of a weld segment to be detected;

[0023] The model detection unit is used to input the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0024] In one embodiment, the weld image acquisition unit is used to: during the welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, acquire a weld image of the weld segment to be detected.

[0025] In one embodiment, the device also includes a welding correction unit, which is used to, after obtaining the output result information of the weld detection model, issue an alarm and / or correct the weld segment to be detected if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold.

[0026] In one embodiment, the weld image acquisition unit is used to: acquire a surface contour image of the product to be inspected, and perform image segmentation on the surface contour image to acquire a plurality of weld images corresponding to a plurality of weld segments;

[0027] The model detection unit is used to: input the multiple weld images into the pre-trained weld detection model respectively, and obtain output result information of the weld detection model respectively.

[0028] In one embodiment, the device also includes a detection recording unit, which is used to input any weld image among the multiple weld images into the pre-trained weld detection model to obtain the output result information of the weld detection model. If the output result information of the weld image determines that the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, the weld segment corresponding to the weld image is recorded as unqualified.

[0029] In one embodiment, the device further includes a repair determination unit for determining to repair the product to be inspected if there is an unqualified record in the weld segment corresponding to any weld image cut from the surface contour image of the product to be inspected.

[0030] In one embodiment, the weld image acquisition unit is used to: scan the product to be inspected by an industrial robot equipped with a 3D laser scanner to obtain a surface contour image of the product to be inspected; and segmentally intercept the surface contour image of the product to be inspected to obtain multiple weld images corresponding to multiple weld segments.

[0031] In one embodiment, the weld detection model is obtained by training the following modules:

[0032] A sample acquisition module, used to acquire a training sample set, wherein the training sample includes a weld sample image and a label used to indicate whether the weld displayed by the weld sample image is qualified;

[0033] A model determination module, used for determining an initialized weld detection model, wherein the initialized weld detection model includes a target layer for outputting a standard compliance of a weld displayed in a weld image;

[0034] The model training module is used to utilize a machine learning device to use the weld sample image in the training sample in the training sample set as the input of the initialized weld detection model, and to use the annotation information corresponding to the input weld sample image as the expected output of the initialized weld detection model to train the weld detection model.

[0035] In one embodiment, determining the initialized weld detection model includes: determining that the initialized weld detection model is a three-layer neural network model.

[0036] In a third aspect of the present disclosure, an industrial robot is provided, comprising: a processor; and a memory for storing executable instructions, wherein when the executable instructions are executed by the processor, the industrial robot executes the method in the first aspect.

[0037] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method in the first aspect is implemented.

[0038] The beneficial technical effects of the technical solution proposed in the embodiment of the present invention are:

[0039] The embodiment of the present invention obtains a weld image of a weld segment to be detected, inputs the weld image into a pre-trained weld detection model, and determines the standard compliance of the weld displayed in the weld image according to the output result of the weld detection model. The technical solution of the embodiment of the present invention can improve the efficiency of weld detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only part of the embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the contents of the embodiments of the present invention and these drawings without creative work.

[0041] Figure 1 is a schematic flow chart of a weld detection method provided according to an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a flow chart of a method for training a weld detection model provided by an embodiment of the present disclosure is shown;

[0043] Figure 3 is a schematic flow chart of another weld detection method provided according to an embodiment of the present invention;

[0044] Figure 4 is a schematic flow chart of another weld detection method provided according to an embodiment of the present invention;

[0045] Figure 5 is a structural schematic diagram of a weld detection device provided according to an embodiment of the present invention;

[0046] Figure 6 A schematic structural diagram of a training device for a weld detection model provided by an embodiment of the present disclosure is shown;

[0047] Figure 7 is a schematic structural diagram of another weld detection device provided according to an embodiment of the present invention;

[0048] Figure 8 A schematic structural diagram of an industrial robot suitable for implementing an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0049] In order to make the technical problems solved by the embodiments of the present invention, the technical solutions adopted and the technical effects achieved more clearly, the technical solutions of the embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the embodiments of the present invention.

[0050] It should be noted that the terms "system" and "network" in the embodiments of the present invention are often used interchangeably herein. The "and / or" mentioned in the embodiments of the present invention refers to any and all combinations including one or more of the related listed items. The terms "first", "second", etc. in the specification, claims and drawings of the present disclosure are used to distinguish different objects, rather than to limit a specific order.

[0051] It should also be noted that the following embodiments of the present invention may be executed separately or in combination with each other, and the present invention does not impose any specific restrictions on this.

[0052] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0053] The technical solution of the embodiment of the present invention is further explained below with reference to the accompanying drawings and through specific implementation methods.

[0054] Figure 1 The flow chart of a weld detection method provided by an embodiment of the present invention is shown. This embodiment is applicable to the case of automatically detecting the weld of a product. The method can be executed by a weld detection device configured in an industrial robot, such as Figure 1 As shown, the weld detection method described in this embodiment includes:

[0055] In step S110 , a weld image of the weld segment to be inspected is acquired.

[0056] It should be noted that the timing of obtaining the weld image of the weld segment to be inspected is related to the specific application scenario of the technical solution of this embodiment. The technical solution of this embodiment can be used for real-time weld detection during the welding process and timely correction when it does not meet the standards. It can also be used to inspect finished products during the product quality inspection stage to determine whether each product is qualified in terms of welds.

[0057] For example, if the technical solution of this embodiment can be used to perform real-time weld detection during the welding process, during the welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, a weld image of the weld segment to be detected can be obtained.

[0058] For example, if the technical solution of this embodiment is used to inspect finished products during the product quality inspection stage, a surface contour image of the product to be inspected can be obtained, and the surface contour image can be cut to obtain multiple weld images corresponding to multiple weld segments, so as to inspect the multiple weld images separately.

[0059] In step S120, the weld image is input into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0060] The weld detection model can be trained in a variety of ways. Figure 2 A flow chart of a method for training a weld detection model is shown, Figure 2 As shown, the training method of the weld detection model includes:

[0061] In step S210, a training sample set is obtained, wherein the training samples include a weld sample image and a label used to indicate whether the weld displayed by the weld sample image is qualified.

[0062] In step S220, an initialized weld detection model is determined, wherein the initialized weld detection model includes a target layer for outputting a standard compliance degree of a weld displayed in a weld image.

[0063] In step S230, using a machine learning method, the weld sample image in the training sample in the training sample set is used as the input of the initialized weld detection model, and the annotation information corresponding to the input weld sample image is used as the expected output of the initialized weld detection model to train the weld detection model.

[0064] In some embodiments, determining the initialized weld detection model includes: determining that the initialized weld detection model is a three-layer neural network model.

[0065] In this embodiment, after obtaining a weld image of a weld segment to be detected, the weld image is input into a pre-trained weld detection model, and the standard compliance of the weld displayed by the weld image is determined according to the output result of the weld detection model. A robot that controls the weld during the welding process can automatically identify the weld quality, or automatically determine whether the product is qualified in terms of the weld during the product quality inspection stage, thereby avoiding excessive manual participation and improving the efficiency of weld detection.

[0066] Figure 3 FIG. 1 is a flow chart of another weld detection method provided by an embodiment of the present invention. This embodiment is applicable to the case where weld detection is performed in real time during a welding process. The method can be performed by a weld detection device configured in an industrial robot, such as Figure 1 As shown, the weld detection method described in this embodiment includes:

[0067] In step S310, during the welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, a weld image of the weld segment to be detected is acquired.

[0068] In step S320, the weld image is input into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0069] The weld detection model can be trained in a variety of ways. Figure 2 A flow chart of a method for training a weld detection model is shown, which will not be described in detail in this embodiment.

[0070] In step S330, if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, an alarm is issued and / or the weld segment to be detected is corrected.

[0071] For example, an alarm can be used to prompt workers to make manual corrections, or the robot can be controlled to make corrections autonomously.

[0072] This embodiment obtains a weld image during the welding process of a product and inputs the weld image into a pre-trained weld detection model. According to the output result of the weld detection model, it is determined that the weld displayed in the weld image does not meet the standard, and an alarm is issued or correction is made in time, thereby improving the authenticity rate of the product.

[0073] Figure 4 A schematic flow chart of another weld detection method provided by an embodiment of the present invention is shown. This embodiment is applicable to the situation of automatically detecting the welds of products. The method can be executed by a weld detection device configured in an industrial robot.

[0074] Before controlling the industrial robot for inspection, the user can use input devices such as mouse and keyboard to input some basic production information according to actual needs, such as the threshold for judging the quality of product workpieces, the speed of machine scanning, etc.

[0075] According to their actual needs, users use input devices such as mouse and keyboard to input some basic production information, such as the threshold for judging the quality of product workpieces and the speed of machine scanning.

[0076] like Figure 4 As shown, the weld detection method described in this embodiment includes:

[0077] In step S410, a surface contour image of a product to be inspected is acquired, and image segmentation is performed on the surface contour image to acquire a plurality of weld images corresponding to a plurality of weld segments.

[0078] It should be noted that for batches of products to be inspected, when controlling the industrial robot to perform product inspection, the algorithm in the industrial robot controller can be used to realize automatic path planning and automatic collision avoidance of the industrial robot, calculate the real-time spatial coordinates of the current workpiece, and guide the industrial robot to automatically complete the workpiece scanning task.

[0079] The welding product workpiece is scanned by an industrial robot equipped with a 3D laser scanner. Through the 3D imaging system, a large amount of surface profile data of qualified products and unqualified workpieces is obtained.

[0080] There are many methods for obtaining the surface contour image of the product to be inspected. For example, an industrial robot equipped with a 3D laser scanner can be used to scan the product to be inspected to obtain the surface contour image of the product to be inspected.

[0081] After the surface profile image of the product to be inspected is obtained, it can be segmented to obtain multiple weld images corresponding to multiple weld segments. In this way, the method of inspecting one weld segment at a time can improve the inspection accuracy.

[0082] In step S420, the plurality of weld images are respectively input into the pre-trained weld detection model to obtain output result information of the weld detection model.

[0083] In step S430, if the output result information of the weld image determines that the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, the weld segment corresponding to the weld image is recorded as unqualified.

[0084] In step S440, if there is an unqualified record in the weld segment corresponding to any weld image cut from the surface contour image of the product to be inspected, it is determined to perform a repair process on the product to be inspected.

[0085] This embodiment performs weld quality inspection on finished products during the product quality inspection stage. After determining the weld quality of each segment, it determines whether the weld of the entire product is qualified based on the recorded results of each segment to determine whether the product needs to be repaired. This allows automatic inspection of product welds, reduces labor costs, and further improves product inspection accuracy by performing segmented inspection of the product surface contour image.

[0086] As an implementation of the methods shown in the above figures, the present application provides an embodiment of a weld detection device. Figure 5 A schematic diagram of the structure of a weld detection device provided in this embodiment is shown. Figure 1-Figure 4 Corresponding to the method embodiment shown, the device can be specifically applied to various industrial robots. Figure 5As shown, the weld detection device described in this embodiment includes a weld image acquisition unit 510 and a model detection unit 520.

[0087] The weld image acquisition unit 510 is configured to acquire a weld image of a weld segment to be detected.

[0088] The model detection unit 520 is configured to input the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0089] In some embodiments, the weld image acquisition unit 510 is configured to, during the welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, acquire a weld image of the weld segment to be detected.

[0090] In some embodiments, the device further comprises a welding correction unit ( Figure 5 ), which is used to, after obtaining the output result information of the weld detection model, if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, issue an alarm and / or correct the weld segment to be detected.

[0091] In some embodiments, the weld detection model can be trained in a variety of ways. Figure 6 A structural schematic diagram of a training device for a weld detection model is shown. Figure 6 As shown, the weld detection model can be obtained through training of the following sample acquisition module 610, model determination module 620, and model training module 630.

[0092] The sample acquisition module 610 is configured to acquire a training sample set, wherein the training sample includes a weld sample image and a label indicating whether the weld displayed by the weld sample image is qualified.

[0093] The model determination module 620 is configured to determine an initialized weld detection model, wherein the initialized weld detection model includes a target layer for outputting a standard compliance degree of a weld displayed in a weld image.

[0094] The model training module 630 is configured to utilize a machine learning device to use the weld sample image in the training sample in the training sample set as the input of the initialized weld detection model, and to use the annotation information corresponding to the input weld sample image as the expected output of the initialized weld detection model to train the weld detection model.

[0095] In some embodiments, the model determination module 620 is configured to determine the initialized weld detection model, which may include being configured to determine that the initialized weld detection model is a three-layer neural network model.

[0096] The weld detection device provided in this embodiment can execute the weld detection method provided in the method embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0097] Figure 7 FIG. 4 is a schematic diagram showing the structure of another weld detection device provided by an embodiment of the present invention. Figure 7 As shown, the weld detection device described in this embodiment includes: a weld image acquisition unit 710, a model detection unit 720, a detection recording unit 730 and a rework determination unit 740.

[0098] The weld image acquisition unit 710 is configured to acquire a weld image of a weld segment to be detected.

[0099] The model detection unit 720 is configured to input the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0100] The detection recording unit 730 is configured to record an unqualified weld segment corresponding to the weld image if the output result information of the weld image determines that the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold.

[0101] The repair determination unit 740 is configured to determine to repair the product to be inspected if there is an unqualified record in the weld segment corresponding to any weld image cut from the surface contour image of the product to be inspected.

[0102] In some embodiments, the weld image acquisition unit 710 is configured to acquire a surface contour image of the product to be inspected, and perform image cutting on the surface contour image to acquire a plurality of weld images corresponding to a plurality of weld segments;

[0103] The model detection unit 720 is configured to input the multiple weld images into the pre-trained weld detection model respectively, and obtain output result information of the weld detection model respectively.

[0104] In some embodiments, the weld image acquisition unit 710 is configured to scan the product to be inspected by an industrial robot equipped with a 3D laser scanner to obtain a surface contour image of the product to be inspected; and to segmentally intercept the surface contour image of the product to be inspected to obtain multiple weld images corresponding to multiple weld segments.

[0105] In some embodiments, the weld detection model can be trained in a variety of ways. Figure 6 A structural schematic diagram of a training device for a weld detection model is shown, which is not described in detail in this embodiment.

[0106] The weld detection device provided in this embodiment can execute the weld detection method provided in the method embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0107] Reference below Figure 8 , which shows a schematic diagram of the structure of an industrial robot 800 suitable for implementing the embodiment of the present invention. The terminal device in the embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The industrial robot shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0108] like Figure 8 As shown, the industrial robot 800 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the industrial robot 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0109] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the industrial robot 800 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8The industrial robot 800 having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0110] In particular, according to an embodiment of an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 809, or installed from the storage device 808, or installed from the ROM 802. When the computer program is executed by the processing device 801, the above-mentioned functions defined in the method of the embodiment of the present invention are executed.

[0111] It should be noted that the computer-readable medium in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In an embodiment of the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0112] The computer-readable medium may be included in the industrial robot; or may exist independently without being assembled into the industrial robot.

[0113] The computer-readable medium carries one or more programs. When the one or more programs are executed by the industrial robot, the industrial robot: obtains a weld image of the weld segment to be detected; inputs the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image.

[0114] Computer program code for performing the operations of embodiments of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0115] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0116] The units involved in the embodiments of the present invention may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0117] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A weld detection method, characterized in that: include: Acquire a weld image of a weld segment to be inspected; Inputting the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image; Wherein, the step of obtaining the weld image of the weld segment to be detected comprises: obtaining a surface contour image of the product to be detected, and performing image cutting on the surface contour image to obtain a plurality of weld images corresponding to a plurality of weld segments; Wherein, the step of inputting the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model comprises: inputting the plurality of weld images into the pre-trained weld detection model respectively to obtain output result information of the weld detection model respectively; Wherein, obtaining the surface profile image of the product to be inspected includes: Scanning the product to be inspected by an industrial robot equipped with a 3D laser scanner, and obtaining a surface contour image of the product to be inspected based on a 3D imaging system; Performing image cutting on the surface contour image to obtain a plurality of weld images corresponding to a plurality of weld segments includes: Segmentally intercept the surface contour image of the product to be inspected to obtain a plurality of weld images corresponding to a plurality of weld segments; Wherein, obtaining the weld image of the weld segment to be detected comprises: During the welding process, if it is detected that the length of the weld segment to be detected reaches a predetermined length, a weld image of the weld segment to be detected is obtained; Wherein, after obtaining the output result information of the weld detection model, it also includes, if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, an alarm is issued and / or the weld segment to be detected is corrected; wherein, the method of correcting the weld segment to be detected includes: manual correction by a worker or autonomous correction by a robot.

2. The method according to claim 1, characterized in that After inputting any one of the plurality of weld images into the pre-trained weld detection model to obtain output result information of the weld detection model, the method further includes: If the output result information of the weld image determines that the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, an unqualified record is made for the weld segment corresponding to the weld image.

3. The method according to claim 2, characterized in that The method further includes determining to perform a repair process on the product to be inspected if a weld segment corresponding to any weld image obtained by cutting the surface contour image of the product to be inspected has an unqualified record.

4. The method according to any one of claims 1 to 3, characterized in that: The weld detection model is trained by the following steps: Acquire a training sample set, wherein the training sample includes a weld sample image and a label used to indicate whether the weld displayed by the weld sample image is qualified; Determining an initialized weld inspection model, wherein the initialized weld inspection model includes a target layer for outputting a standard compliance of a weld displayed in a weld image; By using a machine learning method, the weld sample images in the training samples in the training sample set are used as the input of the initialized weld detection model, and the annotation information corresponding to the input weld sample images is used as the expected output of the initialized weld detection model, so as to train and obtain the weld detection model.

5. The method according to claim 4, characterized in that Determining the initialized weld detection model includes: determining that the initialized weld detection model is a three-layer neural network model.

6. A weld detection device, characterized in that: include: A weld image acquisition unit, used to acquire a weld image of a weld segment to be detected; A model detection unit, used for inputting the weld image into a pre-trained weld detection model to obtain output result information of the weld detection model, wherein the output result information is used to indicate the standard compliance of the weld displayed by the weld image; The weld image acquisition unit is used to acquire a surface profile image of the product to be inspected, and perform image cutting on the surface profile image to acquire a plurality of weld images corresponding to a plurality of weld segments; The model detection unit is used to input the plurality of weld images into the pre-trained weld detection model respectively, and obtain output result information of the weld detection model respectively; Wherein, obtaining the surface profile image of the product to be inspected includes: Scanning the product to be inspected by an industrial robot equipped with a 3D laser scanner, and obtaining a surface contour image of the product to be inspected based on a 3D imaging system; Performing image cutting on the surface contour image to obtain a plurality of weld images corresponding to a plurality of weld segments includes: Segmentally intercept the surface contour image of the product to be inspected to obtain a plurality of weld images corresponding to a plurality of weld segments; Wherein, the weld image acquisition unit is further used to acquire a weld image of the weld segment to be detected if it is detected that the length of the weld segment to be detected reaches a predetermined length during the welding process; Wherein, the device further comprises: A welding correction unit is used to, after obtaining the output result information of the weld detection model, if the standard compliance of the weld displayed by the weld image is less than a predetermined compliance threshold, to give an alarm and / or correct the weld segment to be detected; wherein the method of correcting the weld segment to be detected includes: manual correction by a worker or autonomous correction by a robot.

7. An industrial robot, characterized in that: include: processor; as well as A memory for storing executable instructions, wherein when the executable instructions are executed by the one or more processors, the industrial robot executes the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images