Welding spot detection method and system, intelligent terminal and storage medium

The solder joint image is identified and analyzed through the solder joint detection model, and the problem of low accuracy of solder joint detection in the prior art is solved, achieving higher detection accuracy and efficiency.

CN119090823AInactive Publication Date: 2024-12-06NINGBO DONGHAO PHOTOVOLTAIC TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411120787.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing solder joint detection method is difficult to accurately judge the solder joint quality through image detection, especially when the solder joint image has low similarity to the preset good or defective image, resulting in low detection accuracy and misjudgment.

Method used

Use the solder joint detection model to identify the solder joint image. By acquiring and analyzing the solder joint image detection information, it is judged as a good product or a defective product, and clean and adjust it when necessary to ensure the accuracy and quality of image acquisition, thereby improving the Detection accuracy.

Benefits of technology

It improves the accuracy of solder joint detection, reduces misjudgment, ensures the clarity and quality of solder joint images, and enhances the detection efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090823A_ABST
    Figure CN119090823A_ABST
Patent Text Reader

Abstract

The invention relates to a welding spot detection method and system, an intelligent terminal and a storage medium, and relates to the technical field of welding detection, and the method comprises the following steps: obtaining welding spot image detection information of a preset workpiece; inputting the welding spot image detection information into a preset welding spot detection model for welding spot detection to generate welding spot detection result information; judging whether the welding spot detection result information is preset non-defective product detection information or preset defective product detection information; if the detection information is good product detection information, acquiring image detection information of a next welding spot to continuously detect the welding spot; and if the detection information is the defective detection information, performing defective marking on the workpiece, and obtaining the image detection information of the next welding spot to continuously detect the welding spot. The method and the device have the effect of improving the accuracy of welding spot detection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of welding detection technology, and in particular to a welding point detection method, system, intelligent terminal and storage medium. Background Art

[0002] Currently, in hardware processing plants, after the product welding process is completed, the product's solder joints need to be inspected to ensure the quality of the solder joints, thereby providing product quality control.

[0003] In the related art, after the metal connection terminal is spot welded, it is usually inspected by image detection method, controlling the camera to capture images of the welding points of the metal connection terminal, and then comparing the captured images with pre-set good product images and defective product images to determine whether the quality of the welding points of the metal connection terminal is qualified.

[0004] With respect to the above-mentioned related technologies, there is great uncertainty in solder joints. When using image detection methods to identify whether the quality of solder joints of metal connection terminals is qualified, only solder joint images with high similarity to good product images and defective product images can be judged, while solder joint images with low similarity to good product images and defective product images cannot be judged, resulting in misjudgment of solder joint quality and low detection accuracy of solder joints. There is still room for improvement. Summary of the invention

[0005] In order to improve the accuracy of solder joint detection, the present application provides a solder joint detection method, system, intelligent terminal and storage medium.

[0006] In a first aspect, the present application provides a solder joint detection method, which adopts the following technical solution: A solder joint detection method, comprising: Obtaining the welding spot image detection information of the preset workpiece; Inputting the solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information; Determine whether the solder joint inspection result information is preset good product inspection information or preset defective product inspection information; If it is good product detection information, then obtain the next solder joint image detection information to continuously detect the solder joints; If it is defective detection information, the workpiece is marked as defective, and the next solder joint image detection information is obtained to continuously detect the solder joints.

[0007] By adopting the above technical solution, the solder joint detection model is controlled to identify the solder joint image detection information, thereby obtaining the solder joint detection result information, and it will not be impossible to identify whether the solder joint is qualified because the solder joint image does not exist in the preset image, thereby improving the accuracy of solder joint detection.

[0008] Optionally, the step of obtaining the welding spot image detection information of a preset workpiece includes: Obtain the welding spot detection trigger information of the workpiece; Controlling a preset conveying device to convey the workpiece into a preset image detection area according to the welding spot detection trigger information; Get area arrival trigger information; According to the area arrival trigger information, a preset image acquisition device is controlled to acquire the welding point image detection information of the workpiece.

[0009] By adopting the above technical solution, after receiving the weld spot detection trigger information, the conveying device is controlled to convey the workpiece into the image detection area, and the image acquisition device is controlled to acquire the weld spot image detection information of the workpiece, thereby improving the convenience of obtaining the weld spot image detection information.

[0010] Optionally, the step of controlling a preset conveying device to convey the workpiece into a preset image detection area according to the welding spot detection trigger information includes: Control the conveying device to convey the workpiece according to the welding spot detection trigger information, and obtain the cleaning trigger information; Acquire cleaning image information of the workpiece according to the cleaning trigger information; Determining whether the cleaned image information meets the preset requirements of the front cleaned image information; If it meets the requirements, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method, and the preset cleaning completion trigger information is output; If not, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset adjustment cleaning method, and the preset cleaning completion trigger information is output; The conveying device is controlled to convey the workpiece into the image detection area according to the cleaning completion trigger information.

[0011] By adopting the above technical solution, upon receiving the cleaning trigger information, the air blowing cleaning device is controlled to blow air on the surface of the workpiece, thereby preventing stains on the workpiece surface from causing the quality of the collected workpiece image to be unqualified, thereby affecting the subsequent image recognition, thereby improving the quality of the collected image.

[0012] Optionally, the step of controlling a preset air blowing cleaning device to perform air blowing cleaning on the workpiece according to a preset adjustment cleaning method includes: Analyze the cleaned image information to determine the orientation information of the solder joint; Analyze the solder joint orientation information and the preset front orientation information to determine the adjustment direction information; Controlling a preset adjustment component to adjust the welding point orientation of the workpiece according to the adjustment direction information, and obtaining adjustment orientation image information in real time; Determine whether the adjusted orientation image information meets the requirements of the front cleanup image information; If it does not meet the requirements, the adjustment component is controlled to adjust the welding point orientation of the workpiece according to the adjustment direction information, and the adjustment orientation image information is obtained in real time for cyclic judgment; If it meets the requirements, the adjustment component is controlled to stop, and the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method.

[0013] By adopting the above technical solution, the adjustment component is controlled to adjust the orientation of the welding point of the workpiece, so that when it is determined that the adjustment orientation image information meets the requirements of the front cleaning image information, the adjustment component is controlled to stop, and the air blowing cleaning device is controlled to blow air to clean the welding point on the workpiece, thereby preventing the welding point from being unable to be cleaned when the welding point is not facing up, thereby improving the convenience of stain cleaning.

[0014] Optionally, the step of controlling a preset image acquisition device to acquire weld spot image detection information of a workpiece according to the area arrival trigger information includes: According to the area arrival trigger information, the preset lighting components are controlled to illuminate the workpiece, and the acquisition trigger information is output; Controlling the image acquisition device to photograph the workpiece to obtain basic image information according to the acquisition trigger information; Analyze the basic image information and preset workpiece feature information to determine the precise image information; Processing the precise image information according to a preset random processing method to generate verification image information; The verification image information and the precise image information are associated to generate solder joint image inspection information.

[0015] By adopting the above technical solution, the image acquisition device is controlled to shoot the workpiece to obtain basic image information, and then the basic image information is cropped according to the workpiece feature information to obtain precise image information, thereby reducing picture information and facilitating model recognition. The precise image information is processed according to the random processing method to obtain verification image information, and the verification image information and the precise image information are associated to generate solder joint image detection information, so as to prevent the model from being unable to accurately give detection results due to a single image, thereby improving the efficiency and accuracy of solder joint detection.

[0016] Optionally, the step of controlling a preset lighting component to illuminate the workpiece according to the area arrival trigger information includes: Acquire regional image information according to regional arrival trigger information; Analyze the regional image information, the preset solder joint feature information and the preset coordinate system to determine the solder joint position information; Analyze the welding spot location information and preset related distance information to determine the surrounding area location information; The lighting assembly is controlled to illuminate the welding spot according to the preset point light source information and welding spot position information, and the lighting assembly is controlled to illuminate the workpiece according to the preset ring light source information and surrounding area position information.

[0017] By adopting the above technical solution, the lighting component is controlled to illuminate the welding point with point light source information, and the workpiece is illuminated according to the ring light source information, so that the workpiece and the welding point are clearer during image acquisition, thereby improving the image quality.

[0018] Optionally, the step of inputting the solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information includes: Analyze the solder joint image detection information to determine input feature image information and verification feature image information; Controlling the solder joint detection model to perform solder joint detection on the input feature image information to determine basic detection result information; Determine whether the basic test result information meets the preset clear result information requirements; If it is in compliance, the basic test result information is defined as the solder joint test result information; If it does not meet the requirements, the solder joint detection model is controlled to perform solder joint detection on the verification feature image information to determine the verification detection result information; Analyze the basic test result information and the verification test result information to determine the result score information; The solder joint inspection result information is determined by analyzing the result score information, the preset good product score information and the preset defective product score information.

[0019] By adopting the above technical scheme, when the basic test result information does not meet the requirements of clear result information, the solder joint detection model is controlled to perform solder joint detection according to the verification feature image information to determine the verification test result information, thereby determining the result score information according to the basic test result information and the verification test result information, and after determining that the result score information is closer to the good product score information or the defective product score information, the solder joint detection result information is determined, thereby improving the accuracy of the solder joint detection result information.

[0020] In a second aspect, the present application provides a solder joint detection system, which adopts the following technical solution: A solder joint detection system, comprising: An acquisition module is used to acquire solder joint image detection information; A memory, used to store a program of a solder joint detection method as described in any one of the above items; The program in the memory can be loaded and executed by the processor to implement a solder joint detection method as described in any one of the above items.

[0021] By adopting the above technical solution, the processor loads and executes a program of a solder joint detection method stored in the memory, so that the acquisition module obtains a series of data related to solder joint detection, and the solder joint detection model identifies whether the quality of the solder joint is qualified based on the solder joint image detection information, and the problem of being unable to identify the solder joint quality due to the image not existing in the preset image will not occur, thereby improving the accuracy of solder joint detection.

[0022] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution: An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and executes a solder joint detection method as described in any one of the above items.

[0023] By adopting the above technical solution and operating the intelligent terminal, the processor is made to load and execute a computer program of a solder joint detection method stored in the memory, so that the solder joint detection model can identify whether the solder joint quality is qualified according to the solder joint image detection information, and the problem of being unable to identify the solder joint quality due to the image not existing in the preset image will not occur, thereby improving the accuracy of solder joint detection.

[0024] In a fourth aspect, the present application provides a computer storage medium capable of storing corresponding programs, which is convenient for improving the accuracy of solder joint detection, and adopts the following technical solutions: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned solder joint detection methods.

[0025] By adopting the above technical solution, a computer program of a weld spot detection method is stored in a computer-readable storage medium, so that the processor loads and executes the computer program in the storage medium, and the weld spot detection model identifies whether the weld spot quality is qualified according to the weld spot image detection information, and the problem of being unable to identify the weld spot quality due to the image not existing in the preset image will not occur, thereby improving the accuracy of weld spot detection.

[0026] In summary, the present application includes at least one of the following beneficial technical effects: 1. By controlling the solder joint detection model to identify the solder joint image detection information, the solder joint detection result information is obtained, and the solder joint image does not exist in the preset image, which will lead to the inability to identify whether the solder joint is qualified, thereby improving the accuracy of solder joint detection; 2. Upon receiving the cleaning trigger information, the air blowing cleaning device is controlled to blow air on the surface of the workpiece, thereby preventing stains on the workpiece surface from causing the quality of the collected workpiece image to be unqualified, thereby affecting the subsequent image recognition, thereby improving the quality of the collected image; 3. The basic image information is obtained by controlling the image acquisition device to shoot the workpiece, and then the basic image information is cropped according to the feature information of the workpiece to obtain the precise image information, thereby reducing the picture information and facilitating the model to recognize it. The precise image information is processed according to the random processing method to obtain the verification image information, and the verification image information and the precise image information are associated to generate the weld image detection information to prevent the model from being unable to accurately give the detection result due to a single image, thereby improving the efficiency and accuracy of the weld detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of a solder joint detection method in an embodiment of the present application.

[0028] Figure 2 It is a flow chart of the steps of obtaining the welding spot image detection information of a preset workpiece in an embodiment of the present application.

[0029] Figure 3 It is a flowchart of the steps of controlling a preset conveying device to convey a workpiece into a preset image detection area according to welding spot detection trigger information in an embodiment of the present application.

[0030] Figure 4 It is a flow chart of the steps of controlling a preset air blowing cleaning device to perform air blowing cleaning on a workpiece according to a preset adjustment cleaning method in an embodiment of the present application.

[0031] Figure 5 It is a flowchart of the steps of controlling a preset image acquisition device to acquire weld spot image detection information of a workpiece according to area arrival trigger information in an embodiment of the present application.

[0032] Figure 6 It is a flowchart of the steps of controlling a preset lighting component to illuminate a workpiece according to area arrival trigger information in an embodiment of the present application.

[0033] Figure 7 It is a flowchart of the steps of inputting solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-Figure 7 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0035] The embodiment of the present application discloses a method for quickly detecting welding points on metal connection terminals, and specifically discloses a processing terminal and an image acquisition device. The processing terminal and the image acquisition device are connected through a data wire to realize data interaction and control. After the processing terminal controls the image acquisition device to acquire welding point image detection information and sends the welding point image detection information to the processing terminal, the welding point detection model in the processing terminal identifies the welding point detection result information of the welding point according to the welding point image detection information. Therefore, when the welding point detection result information is good product detection information, the welding point image detection information of the next workpiece is continuously obtained for continuous identification and detection. When the welding point detection result is defective product detection information, the workpiece is marked and the welding point image detection information of the next workpiece is continuously obtained for continuous identification and detection.

[0036] Reference Figure 1 The present application embodiment discloses a solder joint detection method, comprising the following steps: Step S100: obtaining welding spot image detection information of a preset workpiece.

[0037] The workpiece refers to a part that has been welded. In this application, a metal connection terminal that has been laser spot welded is taken as an example. Each metal connection terminal has two circular welding spots arranged side by side. The welding spot image detection information refers to the welding spot image of the workpiece, which is obtained by the processing terminal controlling the image acquisition device to shoot and process. For the specific method, refer to Figure 2 steps.

[0038] Step S101: inputting solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information.

[0039] Among them, the solder joint detection model refers to a machine learning model that identifies whether the quality of the solder joint is qualified based on the solder joint image. The network structure of the machine learning model in this application includes: an input layer, which is used to input the image; an initial convolution layer, which uses a standard convolution layer to perform preliminary processing on the input image to reduce the input dimension and extract basic features; multiple repeated depth-separable convolution blocks, each block includes depth convolution, using a smaller convolution kernel to independently apply convolution to each input channel, batch normalization, improving training stability and accelerating convergence, HSwish activation function, introducing nonlinearity, while avoiding expensive exponential operations, point-by-point convolution, using 1x1 convolution kernel to combine features from depth convolution, and adjusting the number of channels; SE module, at the end of the depth-separable convolution block, an SE module is added to learn channel weights to enhance the model's attention to important features; a global average pooling layer, which reduces the final feature map to a vector of a fixed length, usually used to prepare the output of a classification task; an output layer, a fully connected layer, used for classification tasks, and the number of output nodes is equal to the number of categories. A large number of solder joint images are used as feature inputs and the corresponding results and result scores are used as labels to train the model. After evaluating the model, the model is quantized into int8 and deployed in the processing terminal to improve the running speed of the model.

[0040] The solder joint inspection result information refers to the inspection result data of the solder joint quality. The processing terminal inputs the solder joint image inspection information into the solder joint inspection model for identification. The specific method is as follows: Figure 7 steps.

[0041] Step S102: determining whether the solder joint inspection result information is preset good product inspection information or preset defective product inspection information.

[0042] Among them, the good product detection information refers to the trigger data of the solder joint quality being good, and the defective product detection information refers to the trigger data of the solder joint quality being defective, both of which are stored in the processing terminal by the operator.

[0043] The processing terminal determines whether the inspection result corresponding to the solder joint inspection result information is the result data corresponding to the good product inspection information or the result data corresponding to the defective product inspection information, thereby determining whether the workpiece needs to be processed.

[0044] Step S1021: If it is good product detection information, then obtain the next solder joint image detection information to continuously detect the solder joints.

[0045] Among them, if the processing terminal determines that the detection result corresponding to the solder joint detection result information is the result data corresponding to the good product detection information, it indicates that the quality of the solder joint on the workpiece is good, so the next solder joint image detection information continues to be collected, thereby performing continuous solder joint detection.

[0046] Step S1022: If the information is defective, the workpiece is marked as defective, and the next solder joint image detection information is obtained to continuously detect the solder joints.

[0047] Among them, if the processing terminal determines that the test result corresponding to the weld point detection result information is the result data corresponding to the defective detection information, it indicates that the weld point quality of the workpiece is unqualified. Therefore, the workpiece is marked as defective using light, and then the next weld point image detection information is detected, so as to perform continuous weld point detection.

[0048] Reference Figure 2 , the steps of obtaining the welding spot image detection information of the preset workpiece include: Step S200: obtaining the welding spot detection trigger information of the workpiece.

[0049] The solder joint detection trigger information refers to the trigger data for starting to detect the solder joints on the workpiece, and the operator operates the start button to send the solder joint detection trigger information to the processing terminal.

[0050] Step S201: controlling a preset conveying device to convey the workpiece into a preset image detection area according to the welding spot detection trigger information.

[0051] After the processing terminal receives the solder joint detection trigger information, the processing terminal controls the conveying device to convey the workpiece into the image detection area. The specific method is as follows Figure 3 steps, thereby providing basic support for the subsequent image acquisition device to detect the solder joint image.

[0052] The conveying device refers to a device for conveying workpieces. In this application, a chain conveyor belt is used. A plurality of clamping cavities are evenly spaced on the conveyor belt. One end of the metal connection terminal can be inserted into the clamping cavity. When the conveyor belt is conveying, the metal connection terminals can enter the image detection area one by one. A micro motor is installed at the bottom of the clamping cavity. The micro motor is used to rotate the metal connection terminal. The image detection area refers to the area where the image of the welding point on the workpiece is collected. It is located at a certain position in the conveying path of the conveying device. The specific position can be selected by the operator according to the actual situation.

[0053] Step S202: Acquire area arrival trigger information.

[0054] Among them, the area arrival trigger information refers to the trigger data when the conveying device conveys the workpiece into the image detection area. When the photoelectric sensor installed in the image detection area senses that the workpiece enters the area, the area arrival trigger information is sent to the processing terminal.

[0055] Step S203: controlling a preset image acquisition device to acquire weld spot image detection information of the workpiece according to the area arrival trigger information.

[0056] Among them, after the processing terminal receives the area arrival trigger information, it indicates that the conveying device has conveyed the workpiece to the image detection area, so the processing terminal controls the image acquisition device to collect the welding point image detection information of the workpiece. The specific method is referred to Figure 5 steps.

[0057] The image acquisition device refers to a device for acquiring images of welding points on a workpiece, and a fixed camera may be used.

[0058] Reference Figure 3 The step of controlling a preset conveying device to convey a workpiece into a preset image detection area according to the welding spot detection trigger information includes: Step S300: controlling the conveying device to convey the workpiece according to the welding spot detection trigger information, and obtaining the cleaning trigger information.

[0059] Among them, after the processing terminal receives the welding point detection trigger information, the processing terminal controls the conveying device to convey the workpiece in a preset direction and speed, and obtains the cleaning trigger information to provide data support for the subsequent cleaning adjustment of the workpiece.

[0060] The cleaning trigger information refers to the trigger data for cleaning the surface of the workpiece. When the conveying device transports the workpiece to the cleaning area, the photoelectric sensor installed in the cleaning area senses the workpiece and sends the cleaning trigger information to the processing terminal.

[0061] Step S301: obtaining cleaning image information of the workpiece according to cleaning trigger information.

[0062] Among them, the cleaning image information refers to the image collected of the workpiece in the cleaning area. After the processing terminal receives the cleaning trigger information, the camera in the cleaning area is controlled to shoot the workpiece and send it to the processing terminal. The cameras in the cleaning area include cameras in four directions of up, down, left and right. Similarly, the cleaning image information includes images of the workpiece in four directions.

[0063] Step S302: determining whether the cleaned image information meets the preset requirements of the front cleaned image information.

[0064] The front cleaning image information refers to an image of the workpiece with the welding point facing upward, which is stored in the processing terminal by the operator. The requirement for the front cleaning image information is that the image captured by the front camera is consistent with the image corresponding to the front cleaning image information.

[0065] By processing the terminal, it is determined whether the image captured by the front camera in the cleaning image information is consistent with the image corresponding to the front cleaning image information, thereby determining whether the welding point on the workpiece is facing up, providing data support for subsequent cleaning of the workpiece.

[0066] Step S3021: If it is in compliance, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method, and the preset cleaning completion trigger information is output.

[0067] Among them, if the processing terminal determines that the image taken by the front camera in the cleaning image information is consistent with the image corresponding to the front cleaning image information, it indicates that the weld on the workpiece is facing up. Therefore, the air blowing cleaning device can be controlled according to the direct blowing method to perform air blowing cleaning on the workpiece, and output the cleaning completion trigger information, providing a specific moment for the subsequent delivery of the workpiece into the image detection area.

[0068] The direct blowing method refers to a method of controlling the air blowing cleaning device to directly blow air to clean the welds on the workpiece. The power and time of the air blowing are determined by the operator according to the actual situation. The air blowing cleaning device refers to a device that blows air around the welds on the workpiece to clean. It can be composed of an air pump and an air blowing nozzle installed in the cleaning area. The cleaning completion trigger information refers to the trigger data for completing the cleaning of the welds on the workpiece. The cleaning time is counted by the air blowing cleaning device. When the time equals the cleaning time, the cleaning completion trigger information is output to the processing terminal.

[0069] Step S3022: If not, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset adjustment cleaning method, and the preset cleaning completion trigger information is output.

[0070] Among them, if the processing terminal determines that the image taken by the front camera in the cleaning image information is inconsistent with the image corresponding to the front cleaning image information, it means that the weld is not facing up. Therefore, after adjusting the direction of the weld according to the cleaning method, the air blowing cleaning device is controlled to blow air to clean the workpiece, and the cleaning completion trigger information is output to provide a specific time for the subsequent delivery of the workpiece into the image detection area.

[0071] The adjustment cleaning method refers to the method of adjusting the direction of the weld point of the workpiece first, and then controlling the air blowing cleaning device to perform air blowing cleaning. The operator stores it in the processing terminal. For specific methods, refer to Figure 4 The air blowing cleaning device and the cleaning completion trigger information in this step are consistent with the air blowing cleaning device and the cleaning completion trigger information in step S3021, and are not described in detail here.

[0072] Step S303: controlling the conveying device to convey the workpiece into the image detection area according to the cleaning completion trigger information.

[0073] Among them, after the processing terminal receives the cleaning completion trigger information, it indicates that the welding points of the workpieces on the conveyor belt are facing upwards and are not affected by impurities. Therefore, the conveying device continues to be controlled to convey the workpieces into the image detection area, providing basic support for image detection on the workpiece welding points.

[0074] Reference Figure 4 The steps of controlling the preset air blowing cleaning device to perform air blowing cleaning on the workpiece according to the preset adjustment cleaning method include: Step S400: analyzing the cleaned image information to determine the solder joint orientation information.

[0075] Among them, the weld orientation information refers to the specific upward orientation of the weld on the workpiece. The processing terminal analyzes the image with the weld in the cleaned image information and determines the orientation of the weld by which camera. For example, if the weld exists in the image taken by the camera on the left side of the cleaned image information, it indicates that the weld is facing left.

[0076] Step S401: analyzing the solder joint orientation information and the preset front orientation information to determine the adjustment direction information.

[0077] The front orientation information refers to the direction in which the welding spot faces upwards, which is stored in the processing terminal by the operator. The adjustment direction information refers to the adjustment direction of adjusting the welding spot of the workpiece to face upwards, which is determined by the processing terminal after analyzing the welding spot orientation information and the front orientation information. For example, if the welding spot orientation information is facing left, with the front orientation information as a reference, the adjustment direction is to rotate rightward along the circumferential direction.

[0078] Step S402: controlling a preset adjustment component to adjust the orientation of the welding point of the workpiece according to the adjustment direction information, and acquiring adjustment orientation image information in real time.

[0079] Among them, after the processing terminal determines the adjustment direction information, the adjustment component is controlled to adjust the welding point orientation of the workpiece in the direction corresponding to the adjustment direction information, and the adjustment orientation image information is detected to provide data support for the subsequent determination of whether the adjustment is completed.

[0080] The adjustment component refers to a micro motor installed on the conveyor belt of the conveyor device. After the workpiece is inserted into the clamping cavity, the micro motor drives the workpiece to rotate to adjust the direction of the weld. The orientation adjustment image information refers to the image of the orientation adjustment of the weld on the workpiece, which is taken by a fixed camera above and sent to the processing terminal.

[0081] Step S403: determining whether the adjusted orientation image information meets the requirement of the front clearing image information.

[0082] Among them, by judging whether the image corresponding to the adjusted orientation image information is consistent with the image corresponding to the front cleaning image information, it is determined whether there are welding spots in the image taken by the fixed camera above, providing data support for whether to continue to adjust the welding spot orientation of the workpiece.

[0083] Step S4031: If not, continue to control the adjustment component to adjust the welding point orientation of the workpiece according to the adjustment direction information, and obtain the adjustment orientation image information in real time for cyclic judgment.

[0084] Among them, if the processing terminal determines that the image corresponding to the adjustment direction image information is inconsistent with the image corresponding to the front cleaning image information, it indicates that there is no weld in the image taken by the fixed camera above, and the direction of the weld has not been adjusted upward, so the adjustment component continues to be controlled to adjust the weld direction of the workpiece according to the adjustment direction information, and the adjustment direction image information continues to be detected to continue to pay attention to the adjustment status of the workpiece.

[0085] Step S4032: If it meets the requirements, the adjustment component is controlled to stop, and the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method.

[0086] Among them, if the processing terminal determines that the image corresponding to the adjustment direction image information is consistent with the image corresponding to the front cleaning image information, it indicates that there is a weld point in the image taken by the fixed camera above, and the direction of the weld point is upward. Therefore, the adjustment component is controlled to stop, and the air blowing cleaning device is controlled to blow and clean the workpiece according to the direct blowing method.

[0087] The direct blowing method in this step is consistent with the direct blowing method in step S3021, and will not be described in detail here.

[0088] Reference Figure 5 The step of controlling a preset image acquisition device to acquire the welding spot image detection information of a workpiece according to the area arrival trigger information comprises: Step S500: controlling a preset lighting component to illuminate the workpiece according to the area arrival trigger information, and outputting acquisition trigger information.

[0089] After the processing terminal receives the area arrival trigger information, the processing terminal controls the lighting component to illuminate the workpiece. The specific method is as follows Figure 6 The steps are performed to make the collected workpiece image clearer, and the collection trigger information is output to provide a specific time for the subsequent control of the image collection device to collect the image.

[0090] The lighting component refers to a device used to illuminate the workpiece and the welding point to make the image of the workpiece and the welding point clearer. It can be composed of a point light source and a ring light source controlled by a manipulator. The acquisition trigger information refers to the trigger data that controls the image acquisition device to acquire the image. After the lighting component is started, the lighting component sends the acquisition trigger information to the processing terminal.

[0091] Step S501: controlling an image acquisition device to photograph a workpiece according to acquisition trigger information to obtain basic image information.

[0092] Among them, after the processing terminal receives the acquisition trigger information, the processing terminal controls the camera in the image acquisition device to shoot the workpiece, so as to obtain basic image information, providing data support for the subsequent determination of the welding point image detection information.

[0093] The basic image information refers to the unprocessed image of the workpiece welding spot, which is directly captured by the camera in the image acquisition device and sent to the processing terminal.

[0094] Step S502: Analyze the basic image information and preset workpiece feature information to determine accurate image information.

[0095] The workpiece feature information refers to the features of the workpiece, which are stored in the processing terminal by the operator. The precise image information refers to the cropped weld image. The processing terminal identifies the workpiece feature information in the image corresponding to the basic image information, thereby determining the specific location of the workpiece feature in the image corresponding to the basic image information, and crops the image at that location to obtain the precise image information.

[0096] Step S503: Processing the precise image information according to a preset random processing method to generate verification image information.

[0097] Among them, the random processing method refers to the random upside-down flipping, random horizontal flipping, random cropping, random brightness, contrast, saturation, hue angle, and channel order changes of the precise image information, which is stored in the processing terminal by the operator. The verification image information refers to the image used to assist in determining the solder joint detection result, which is obtained by the processing terminal by randomly processing the image corresponding to the precise image information according to the random processing method.

[0098] Step S504: Associating the verification image information with the precise image information to generate solder joint image detection information.

[0099] Among them, the solder joint image detection information in this step is obtained by the processing terminal after associating the precise image information and the verification image information, so that the precise image information is preferentially used as input for solder joint detection. When the result is uncertain, the verification image information is used as input for solder joint detection, and the final solder joint detection result is determined by combining the two detection results.

[0100] Reference Figure 6 The steps of controlling the preset lighting component to illuminate the workpiece according to the area arrival trigger information include: Step S600: Acquire regional image information according to regional arrival trigger information.

[0101] The regional image information refers to the image of the image detection area, which is received by the processing terminal and responds to the regional arrival trigger information, thereby controlling the image acquisition device to acquire the image of the image detection area.

[0102] Step S601: Analyze the regional image information, the preset welding point feature information and the preset coordinate system to determine the welding point position information.

[0103] Among them, the solder joint feature information refers to the image data representing the solder joint features, which is stored in the processing terminal by the operator. The coordinate system refers to the plane rectangular coordinate system established in the image detection area. The solder joint position information refers to the specific position coordinates of the solder joint in the image. The processing terminal maps the image corresponding to the regional image information into the coordinate system, and identifies the specific position of the features corresponding to the solder joint feature information in the image in the coordinate system to obtain the solder joint position information.

[0104] Step S602: Analyze the welding point location information and preset related distance information to determine the surrounding area location information.

[0105] The relevant distance information refers to the distance value of the area around the weld, and the specific value is determined by the operator according to the actual situation. The surrounding area position information refers to the position coordinates of the area around the weld. The coordinates in the range obtained by the processing terminal with the coordinates corresponding to the weld position information as the center and the distance corresponding to the relevant distance information as the radius extending outward are all the surrounding area position information.

[0106] Step S603: controlling the lighting assembly to illuminate the welding spot according to the preset point light source information and welding spot position information, and controlling the lighting assembly to illuminate the workpiece according to the preset ring light source information and surrounding area position information.

[0107] Among them, after the processing terminal determines the weld point position information and the surrounding area position information, the lighting component is controlled to illuminate the position corresponding to the weld point position information with the light source corresponding to the point light source information, and the lighting component is controlled to illuminate the position corresponding to the surrounding area position information with the light source corresponding to the ring light source information, thereby improving the clarity of the workpiece and weld point imaging.

[0108] Point light source information refers to data that the light irradiated by the light source appears in the shape of dots, and ring light source information refers to data that the light irradiated by the light source appears in the shape of rings. Both are stored in the processing terminal by the operator.

[0109] Reference Figure 7 The step of inputting the solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information includes: Step S700: Analyze the solder joint image detection information to determine input feature image information and verification feature image information.

[0110] The input feature image information refers to an image used to directly identify the quality of the solder joint, and the processing terminal defines the precise image information in the image corresponding to the solder joint image detection information as the input feature image information. The verification feature image information refers to an image used to assist in determining the quality of the solder joint, and the processing terminal defines the verification image information in the image corresponding to the solder joint image detection information as the verification feature image information.

[0111] Step S701: Controlling the solder joint detection model to perform solder joint detection on the input feature image information to determine basic detection result information.

[0112] Among them, the basic detection result information refers to the solder joint quality result first recognized by the solder joint detection model, which is directly detected by the solder joint detection model based on the input feature image information.

[0113] Step S702: Determine whether the basic test result information meets the preset clear result information requirements.

[0114] The clear result information refers to the trigger data of whether the test result is a good product or a defective product, which is stored in the processing terminal by the operator. The requirement of the clear result information refers to the consistency of the result corresponding to the clear result information.

[0115] The processing terminal determines whether the result corresponding to the basic detection result information is consistent with the result corresponding to the clear result information, thereby determining whether the result obtained by the solder joint detection model based on the input feature image information is clear, and providing data support for whether to assist in determining the detection result in the subsequent process.

[0116] Step S7021: If it is in compliance, the basic test result information is defined as solder joint test result information.

[0117] Among them, if the processing terminal determines that the result corresponding to the basic detection result information is consistent with the result corresponding to the clear result information, it means that the result corresponding to the basic detection result information is clear and there is no uncertain result. Therefore, the basic detection result information is defined as the solder joint detection result information.

[0118] Step S7022: If not, the solder joint detection model is controlled to perform solder joint detection on the verification feature image information to determine the verification detection result information.

[0119] Among them, if the processing terminal determines that the result corresponding to the basic detection result information is inconsistent with the result corresponding to the clear result information, it indicates that the result corresponding to the basic detection result information is ambiguous. Therefore, the solder joint detection model information is controlled to perform solder joint detection on the verification feature image information to obtain the verification detection result information, thereby providing data support for subsequent auxiliary determination of the detection results.

[0120] The verification test result information refers to the solder joint quality result obtained by the solder joint detection model using the image corresponding to the verification feature image information as input.

[0121] Step S70221: Analyze the basic test result information and the verification test result information to determine the result score information.

[0122] Among them, the result score information refers to the score value for the quality of the solder joint, which is determined by the processing terminal after analysis based on the basic test result information and the verification test result information. The specific analysis method is: the solder joint detection model gives result scores for the basic test result information and the verification test result information respectively, and then calculates the average value to obtain the result score information.

[0123] Step S70222: Analyze the result score information, the preset good product score information, and the preset defective product score information to determine the solder joint detection result information.

[0124] Among them, the good product score information refers to the score interval when the quality of the solder joint is good, and the defective product score information refers to the score interval when the quality of the solder joint is defective, both of which are determined by the operator according to the actual situation. The solder joint detection result information in this step is compared by the processing terminal with the good product score information and the defective product score information according to the result score information. If the result score information is closer to the good product score information, the solder joint detection result information is determined to be good, and if it is closer to the defective product score information, the solder joint detection result information is determined to be defective.

[0125] Based on the same inventive concept, an embodiment of the present invention provides a solder joint detection system, comprising: An acquisition module, used to acquire solder joint image detection information, solder joint detection trigger information, area arrival trigger information, cleaning trigger information, cleaning image information, orientation adjustment image information, basic image information and area image information; A memory for storing a program of a solder joint detection method; The program in the memory can be loaded and executed by the processor to implement a solder joint detection method.

[0126] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0127] An embodiment of the present invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute a solder joint detection method.

[0128] Computer storage media include, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0129] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute a solder joint detection method.

[0130] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0131] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A solder joint detection method, It is characterized in that include: Obtaining the welding spot image detection information of the preset workpiece; Inputting the solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information; Determine whether the solder joint inspection result information is preset good product inspection information or preset defective product inspection information; If it is good product detection information, then obtain the next solder joint image detection information to continuously detect the solder joints; If it is defective detection information, the workpiece is marked as defective, and the next solder joint image detection information is obtained to continuously detect the solder joints.

2. A solder joint detection method according to claim 1, It is characterized in that The steps of obtaining the welding spot image detection information of the preset workpiece include: Obtain the welding spot detection trigger information of the workpiece; Controlling a preset conveying device to convey the workpiece into a preset image detection area according to the welding spot detection trigger information; Get area arrival trigger information; According to the area arrival trigger information, a preset image acquisition device is controlled to acquire the welding point image detection information of the workpiece.

3. A solder joint detection method according to claim 2, It is characterized in that The step of controlling a preset conveying device to convey the workpiece into a preset image detection area according to the welding spot detection trigger information includes: Control the conveying device to convey the workpiece according to the welding spot detection trigger information, and obtain the cleaning trigger information; Acquire cleaning image information of the workpiece according to the cleaning trigger information; Determining whether the cleaned image information meets the preset requirements of the front cleaned image information; If it meets the requirements, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method, and the preset cleaning completion trigger information is output; If not, the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset adjustment cleaning method, and the preset cleaning completion trigger information is output; The conveying device is controlled to convey the workpiece into the image detection area according to the cleaning completion trigger information.

4. A solder joint detection method according to claim 3, It is characterized in that The steps of controlling a preset air blowing cleaning device to perform air blowing cleaning on a workpiece according to a preset adjustment cleaning method include: Analyze the cleaned image information to determine the orientation information of the solder joint; Analyze the solder joint orientation information and the preset front orientation information to determine the adjustment direction information; Controlling a preset adjustment component to adjust the welding point orientation of the workpiece according to the adjustment direction information, and obtaining adjustment orientation image information in real time; Determine whether the adjusted orientation image information meets the requirements of the front cleanup image information; If it does not meet the requirements, the adjustment component is controlled to adjust the welding point orientation of the workpiece according to the adjustment direction information, and the adjustment orientation image information is obtained in real time for cyclic judgment; If it meets the requirements, the adjustment component is controlled to stop, and the preset air blowing cleaning device is controlled to perform air blowing cleaning on the workpiece according to the preset direct blowing method.

5. A solder joint detection method according to claim 2, It is characterized in that The step of controlling a preset image acquisition device to acquire welding spot image detection information of a workpiece according to the area arrival trigger information comprises: According to the area arrival trigger information, the preset lighting components are controlled to illuminate the workpiece, and the acquisition trigger information is output; Controlling the image acquisition device to photograph the workpiece to obtain basic image information according to the acquisition trigger information; Analyze the basic image information and preset workpiece feature information to determine the precise image information; Processing the precise image information according to a preset random processing method to generate verification image information; The verification image information and the precise image information are associated to generate solder joint image inspection information.

6. A solder joint detection method according to claim 5, It is characterized in that The steps of controlling the preset lighting assembly to illuminate the workpiece according to the area arrival trigger information include: Acquire regional image information according to regional arrival trigger information; Analyze the regional image information, the preset solder joint feature information and the preset coordinate system to determine the solder joint position information; Analyze the welding spot location information and preset related distance information to determine the surrounding area location information; The lighting assembly is controlled to illuminate the welding spot according to the preset point light source information and welding spot position information, and the lighting assembly is controlled to illuminate the workpiece according to the preset ring light source information and surrounding area position information.

7. A solder joint detection method according to claim 5, It is characterized in that The steps of inputting solder joint image detection information into a preset solder joint detection model to perform solder joint detection to generate solder joint detection result information include: Analyze the solder joint image detection information to determine input feature image information and verification feature image information; Controlling the solder joint detection model to perform solder joint detection on the input feature image information to determine basic detection result information; Determine whether the basic test result information meets the preset clear result information requirements; If it is in compliance, the basic test result information is defined as the solder joint test result information; If it does not meet the requirements, the solder joint detection model is controlled to perform solder joint detection on the verification feature image information to determine the verification detection result information; Analyze the basic test result information and the verification test result information to determine the result score information; The solder joint inspection result information is determined by analyzing the result score information, the preset good product score information and the preset defective product score information.

8. A solder joint detection system, It is characterized in that include: An acquisition module is used to acquire solder joint image detection information; A memory for storing a program of a solder joint detection method according to any one of claims 1 to 7; The program in the memory can be loaded and executed by the processor to implement a solder joint detection method as claimed in any one of claims 1 to 7.

9. A smart terminal, It is characterized in that The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute a solder joint detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, It is characterized in that A computer program capable of being loaded by a processor and executing a solder joint detection method according to any one of claims 1 to 7 is stored.

Citation Information

Patent Citations

  • Bent pipe welding flaw detection method, device and system

    CN113245238A

  • Automatic cleaning device and method for rotary probe sleeve

    CN115228842A

  • Power transformation equipment image defect identification method and system

    CN116596895A