WO2023123003A1 - 机器视觉检测方法、其检测装置及其检测系统 - Google Patents

机器视觉检测方法、其检测装置及其检测系统 Download PDF

Info

Publication number
WO2023123003A1
WO2023123003A1 PCT/CN2021/142250 CN2021142250W WO2023123003A1 WO 2023123003 A1 WO2023123003 A1 WO 2023123003A1 CN 2021142250 W CN2021142250 W CN 2021142250W WO 2023123003 A1 WO2023123003 A1 WO 2023123003A1
Authority
WO
WIPO (PCT)
Prior art keywords
boundary
component
straight line
point
line
Prior art date
Application number
PCT/CN2021/142250
Other languages
English (en)
French (fr)
Inventor
屠银行
高毅飞
Original Assignee
宁德时代新能源科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 宁德时代新能源科技股份有限公司 filed Critical 宁德时代新能源科技股份有限公司
Priority to EP21969381.9A priority Critical patent/EP4286789A1/en
Priority to CN202180097091.3A priority patent/CN117203486A/zh
Priority to JP2023553313A priority patent/JP2024508331A/ja
Priority to PCT/CN2021/142250 priority patent/WO2023123003A1/zh
Priority to KR1020237029433A priority patent/KR20230134597A/ko
Publication of WO2023123003A1 publication Critical patent/WO2023123003A1/zh
Priority to US18/524,990 priority patent/US20240095949A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B5/00—Measuring arrangements characterised by the use of mechanical techniques
    • G01B5/0037—Measuring of dimensions of welds
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/14—Measuring arrangements characterised by the use of optical techniques for measuring distance or clearance between spaced objects or spaced apertures
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
    • G01B11/03—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness by measuring coordinates of points
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • G01B11/2513—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object with several lines being projected in more than one direction, e.g. grids, patterns
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • G01B11/2518—Projection by scanning of the object
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/06—Topological mapping of higher dimensional structures onto lower dimensional surfaces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds

Definitions

  • the present application relates to the field of machine vision, in particular to a machine vision detection method, a detection device and a detection system thereof.
  • Visual inspection is mainly based on naked eye observation, combined with the use of auxiliary tools such as magnifying glass, measuring tools and templates to conduct a comprehensive inspection of the weld surface quality and visual dimensions.
  • the present application provides a machine vision detection method, a detection device and a detection system thereof, which can improve the efficiency and accuracy of machine vision detection.
  • the present application provides a machine vision inspection method.
  • the method includes: receiving a three-dimensional image from a line laser, the three-dimensional image comprising: at least a portion of a boundary of a first part, at least a portion of a boundary of a second part, and at least one of the boundaries of the first part and the boundary of the second part Welding points; convert the three-dimensional image into a two-dimensional grayscale image; obtain the boundary of the first component and the boundary of the second component in the two-dimensional grayscale image; determine N lines between the boundary of the first component and the boundary of the second component Vertical line: Calculate the average value of the lengths of N vertical lines as the gap between the first component and the second component, where N is a positive integer.
  • the three-dimensional image collected by the line laser can support continuous sampling of the detection component, and the camera does not need to be calibrated in advance.
  • the gap between two parts can be calculated more accurately, which improves the detection accuracy.
  • the above-mentioned step of acquiring the boundary of the first part and the boundary of the second part in the two-dimensional grayscale image specifically includes: setting N equally divided boundaries in the boundary area containing the boundary of the first part The first fitting unit of the area; sequentially connect the intersection points between each first fitting unit and the boundary of the first part to form the first straight line fitted with the boundary of the first part; in the boundary area containing the boundary of the second part Among them, N second fitting units of equally divided boundary regions are set; the intersection points between each second fitting unit and the boundary of the second part are connected in turn to form a second straight line fitted with the boundary of the second part; wherein, N is a positive integer between 20-50.
  • N is a positive integer between 20-50.
  • the step of generating N perpendicular lines between the boundary of the first component and the boundary of the second component specifically includes: determining the intersection points between the N first fitting units and the boundary of the first component as Vertical line setting points; each vertical line setting point is used as a starting point to generate N vertical lines perpendicular to the second straight line; or determine the intersection between N second fitting units and the boundary of the second component as the vertical line Line setting points; each vertical line setting point is used as a starting point to generate N vertical lines perpendicular to the first straight line.
  • the vertical lines are generated based on the intersection points between the fitting unit and the boundary of the component, so that the N vertical lines can be evenly distributed along the gap, so that the gap between the two components can be calculated more accurately.
  • the above method further includes: extracting a third straight line and a fourth straight line respectively fitted to the borders on both sides of the solder joint in the two-dimensional grayscale image; The tangent point and the second tangent point where the fourth straight line is tangent to the solder point; calculate the distance between the first tangent point and the second tangent point as the width of the solder point.
  • a method of calculating the width of the welding spot between two components based on the three-dimensional image is also provided. It can be automatically calculated by obtaining a fitting line fitted to the boundary of the solder joint to help more comprehensively evaluate the welding status between components.
  • the above-mentioned step of extracting the third straight line and the fourth straight line respectively fitted to the two side boundaries of the solder joint in the two-dimensional grayscale image specifically includes: Among them, the third fitting units of M equally divided boundary regions are respectively set; the intersection points between each third fitting unit and the first side boundary of the solder joint are connected in turn to form the third straight line, and; In the boundary area of the second side boundary of the solder joint, M fourth fitting units of equally divided boundary areas are set respectively; the intersection points between each fourth fitting unit and the second side boundary of the solder joint are connected in turn to form The fourth straight line; wherein, M is a positive integer between 30-50.
  • the above-mentioned step of determining the first tangent point of the third line tangent to the welding spot and the second tangent point of the fourth line tangent to the welding spot specifically includes: The intersection point between the fitting unit and the first side boundary of the welding point is taken as the first tangent point; and the intersection point of the last fourth fitting unit on the fourth straight line and the second side boundary of the welding spot is taken as the second tangent point.
  • the first tangent point and the second tangent point can be approximately determined simply and quickly by fitting the last fitting unit of the straight line.
  • the above step of determining the first point of tangency between the third line and the welding point and the second point of tangency between the fourth line and the welding point specifically includes: The intersection point between is taken as the first tangent point; and the intersection point between the fourth straight line and the first straight line is taken as the second tangent point.
  • the width of the solder joint can basically be considered as the width of the gap between the two components
  • the intersection point between the fitting straight line based on the component and the fitting on both sides of the solder joint is provided to determine Quickly determine the two tangent points used to calculate the solder joint width, which can help to obtain accurate solder joint width calculation results.
  • the present application provides a machine vision detection device.
  • the machine vision detection device includes: a receiving module, configured to receive a three-dimensional image from a line laser, the three-dimensional image includes: at least a part of the boundary of the first part, at least a part of the boundary of the second part and at least one part of the boundary of the first part and the welding spot on the boundary of the second part; the conversion module is used to convert the three-dimensional image into a two-dimensional grayscale image; the fitting module is used to obtain the boundary of the first part and the relationship with the second part in the two-dimensional grayscale image The boundary; the gap calculation module is used to determine N vertical lines between the first component boundary and the second component boundary; and calculate the average value of the N vertical line lengths as the gap between the first component and the second component , N is a positive integer.
  • the three-dimensional image collected by the line laser can support continuous sampling of the detection component, and the camera does not need to be calibrated in advance.
  • the device can calculate the real gap between two components by averaging multiple vertical lines, effectively eliminating the interference caused by image distortion and the like. As a result, the detection speed and detection accuracy are effectively improved.
  • the present application provides an electronic device.
  • the electronic device includes: a processor and a processor communicated with the processor; the memory stores computer program instructions, and when the computer program instructions are invoked by the processor, the processor executes the above visual detection method.
  • the three-dimensional image collected by the electronic device through the line laser can support continuous sampling of the detection component, and the camera does not need to be calibrated in advance.
  • the real gap between two components can be calculated by averaging multiple vertical lines, which effectively eliminates the interference caused by image distortion and the like. As a result, the detection speed and detection accuracy are effectively improved.
  • the present application provides a non-volatile computer storage medium.
  • the non-volatile computer storage medium stores computer program instructions, so that when the computer program instructions are invoked by the processor, the above visual detection method is executed.
  • the three-dimensional image collected by the line laser can support continuous sampling of the detection component, and the camera does not need to be calibrated in advance.
  • the real gap between two components can be calculated by averaging multiple vertical lines, which effectively eliminates the interference caused by image distortion and the like. As a result, the detection speed and detection accuracy are effectively improved.
  • the present application provides a machine vision inspection system.
  • the machine vision inspection system includes: an image acquisition device including several line lasers for acquiring three-dimensional images; a drive mechanism for causing relative movement between the image acquisition device and the component to be tested; a first communication connection with the image acquisition device A controller, the first controller is used to execute the above machine vision detection method to process the three-dimensional image, so that the processing result of the three-dimensional image can be used for detection of the component to be tested.
  • continuous sampling of the detection component can be supported, and the camera does not need to be calibrated in advance, which can effectively improve the detection accuracy and detection efficiency.
  • the line laser can intuitively display the characteristics of the detection area, which can significantly improve the detection effect of the sampled image, and can greatly improve the detection accuracy.
  • the image acquisition device includes: two line lasers, a sensor bracket and a hood; the two line lasers are respectively arranged on both sides of the sensor bracket; the light hood is fixed on the sensor bracket, and the cover covers the outside of the line laser; the sensor
  • the bracket includes: a height adjustment module and a distance adjustment module; the height adjustment module is used to adjust the height of the line laser; the distance adjustment module is used to adjust the distance between two line lasers.
  • an additional height adjustment module and a spacing adjustment module are also set up, so that the machine vision inspection system can be adaptively adjusted when the size of the component to be tested changes, and meets various requirements.
  • the additionally provided hood can prevent the laser light of the line laser from scattering to the outside, so as to prevent the eyes of the operator from being hurt.
  • the machine vision detection system further includes: a second controller, the second controller is used to control the height adjustment module and the distance adjustment module, so that the two line lasers reach the target distance and/or the target height;
  • the second controller stores several pieces of configuration information recording target distances and target heights; each configuration information corresponds to at least one component to be tested.
  • the configuration information of the components under test of different models, sizes or shapes can be pre-recorded in the second controller. Therefore, when the size, model or shape of the component to be tested changes, technicians can realize automatic switching and quick adjustment by loading the corresponding configuration information, which effectively improves the compatibility and operating efficiency of the detection system.
  • Fig. 1 is the structural representation of the machine vision detection system of some embodiments of the present application.
  • FIG. 2 is a schematic structural diagram of an image acquisition device in some embodiments of the present application.
  • FIG. 3 is a schematic structural diagram of a machine vision inspection system according to other embodiments of the present application.
  • Fig. 4 is the method flowchart of the machine vision detection method of some embodiments of the present application.
  • FIG. 5 is a schematic diagram of a two-dimensional grayscale image of some embodiments of the present application.
  • Fig. 6 is a method flowchart of step S403 in some embodiments of the present application.
  • Fig. 7a is a schematic diagram of acquiring component boundaries in some embodiments of the present application, showing a fitting unit for equally dividing the boundary area;
  • Figure 7b is a schematic diagram of the boundary of some embodiments of the present application, showing the display form of the component boundary obtained in Figure 7a;
  • Fig. 8 is a method flowchart of a machine vision detection method according to another embodiment of the present application.
  • Fig. 9 is a method flowchart of step S801 in some embodiments.
  • Fig. 10 is a method flow chart of the machine vision inspection method in some embodiments of the present application, showing that the parts to be tested are the cell top cover and the cell aluminum shell after the pre-welding process is completed;
  • Fig. 11 is a schematic diagram of the parts to be tested in some embodiments of the present application, showing the top cover of the cell and the aluminum case of the cell tested in Fig. 9;
  • Fig. 12 is a schematic diagram of a machine vision inspection device according to some embodiments of the present application.
  • Fig. 13 is a schematic diagram of a machine vision detection device according to another embodiment of the present application.
  • Fig. 14 is a schematic diagram of an electronic device according to some embodiments of the present application.
  • multiple refers to more than two (including two), similarly, “multiple groups” refers to more than two groups (including two), and “multiple pieces” refers to More than two pieces (including two pieces).
  • the top cover of the cell and the aluminum shell need to be welded.
  • a pre-welding process for preliminary welding of the top cover and the aluminum case to realize the initial positioning between the two.
  • a typical machine vision inspection method is to use a two-dimensional camera to collect image information of the welding part of the top cover and the aluminum shell after the pre-welding process, and then process and analyze the image information to determine whether the gap between the two meets the requirements.
  • the applicant provides a component gap detection method based on line lasers, which can effectively solve the problem that traditional two-dimensional cameras need to be calibrated in advance, and there are many interferences in collecting image information The defects that make the detection accuracy and detection efficiency low.
  • the components to be tested in the embodiments of the present application are the top cover and the aluminum shell after the pre-welding process is taken as an example.
  • the machine vision inspection system of the embodiment of the present application can also be applied to other components to be tested with similar structural shape characteristics for inspection.
  • FIG. 1 is a schematic structural diagram of a machine vision inspection system according to some embodiments of the present application.
  • the machine vision inspection system includes: an image acquisition device 110 , a driving mechanism 120 and a first controller 130 .
  • the image acquisition device 110 is a device for acquiring three-dimensional image signals. Specifically, any suitable type and quantity of line lasers can be selected and used, and a support structure suitable for the line lasers can be used.
  • the driving mechanism 120 is an action unit for driving relative movement between the component to be tested (for example, the cell top cover and the cell aluminum case after the pre-soldering process) and the image acquisition device 110 .
  • the component to be tested can be clamped and fixed on the driving mechanism 120 , driven by the driving mechanism 120 , it moves relative to the line laser of the image acquisition device 110 , so as to complete the image sampling of the component to be tested.
  • the first controller 130 may be an electronic computing device with a logic operation function, including but not limited to a server or an industrial computer. It can establish a communication connection with the image acquisition device in a wired or wireless manner, so as to receive the three-dimensional image signal acquired by the image acquisition device.
  • the cell top cover A1 and the cell aluminum case A2 can be driven by a motor or other suitable type of driving mechanism 120, and the image acquisition device 110 can be connected at a set speed. relatively mobile.
  • the line laser of the image acquisition device 110 can continuously acquire the image composed of the pre-welded cell top cover A1 and the cell aluminum shell A2 through an encoder and other similar sensor devices according to the acquisition frequency adapted to the relative moving speed. Three-dimensional image signal of the part under test on both long sides.
  • the 3D image signal acquired by the image acquisition device 110 is provided to the first controller 130, and after a series of machine vision detection method steps such as image processing are performed by the first controller 130, the detection result is output and provided to an external device.
  • a series of machine vision detection method steps such as image processing are performed by the first controller 130
  • the detection result is output and provided to an external device.
  • the image acquisition device can acquire the three-dimensional image signals of the component to be tested in a sampling and continuous acquisition manner.
  • the continuous scanning method of the line laser can effectively reduce the frequency of start and stop actions, thus greatly improving the detection speed.
  • the image acquisition device 110 may include: two line lasers 111 , a sensor bracket 112 , a height adjustment module 113 , a spacing adjustment module 114 and a light shield 115 .
  • two line lasers 111 are arranged on both sides of the sensor bracket 112 respectively, and can be used to simultaneously collect three-dimensional image signals of two symmetrical long sides of the cell top cover and the cell aluminum shell.
  • the line laser can have a suitable field of view and pixel precision according to the needs of the actual situation.
  • the resolution of the line laser 111 in the scanning direction can be set to be smaller than the gap detection threshold (eg, 0.08 mm) to meet the detection requirements.
  • the scanning line speed is set above 130mm/s, and the scanning frequency is around 5kHz.
  • Both the height adjustment module 113 and the distance adjustment module 114 are arranged on the sensor bracket 112 .
  • it can be implemented using any suitable type of mechanical structure, including but not limited to based on screws, cylinders, or gears.
  • the height adjustment module 113 and the distance adjustment module 114 can be changed within a certain range to meet the detection requirements of different types or sizes of batteries.
  • the light shield 115 may be disposed on the sensor bracket 112 so as to cover the line laser 111 therein.
  • a cover of any suitable shape, size or material can be used, as long as the cover can cover the line laser 111 .
  • Such a design can avoid the leakage or reflection of the laser light generated by the line laser to the eyes of the operator, and achieve the effect of protecting the human body.
  • the data acquisition of the component to be tested is completed by the line laser, which can effectively avoid the unclear image acquired by the traditional camera, the image is distorted, or the gap between the image and the actual value is biased due to the setting of the light source. Minor flaws.
  • the line laser provides a three-dimensional image signal, which can realize multi-angle and multi-direction measurement, avoid measurement misjudgment caused by blind areas blocked by vision, and can also provide more accurate and intuitive image information.
  • FIG. 3 is a schematic structural diagram of a machine vision inspection system provided by other embodiments of the present application.
  • the machine vision inspection system may also include a second controller 140 .
  • the second controller 140 stores several pieces of configuration information recording target distances and target heights.
  • the configuration information is data information corresponding to the component to be tested, and can be preset by technicians according to actual product production conditions.
  • the second controller when the parts to be tested entering the machine vision inspection system change, technicians or operators can choose to determine the configuration information corresponding to the parts to be tested that need to be tested at present, and then the second controller according to the selected configuration information, automatically control the height adjustment module 113 and the spacing adjustment module 114 to move the line laser to the target spacing and target height recorded in the configuration information, so as to complete the three-dimensional image signal acquisition of the component to be tested.
  • first controller and “second controller” according to different functions to be performed by the controllers.
  • first controller and “second controller” according to different functions to be performed by the controllers.
  • the description of the first controller and the second controller is not intended to limit the specific implementation of the controllers, which may be different functional modules in the same electronic computing device, or may be separate Functional modules arranged in different electronic computing devices.
  • One of the advantages of the embodiment of the present application is: through the pre-stored configuration information, when the size or type of the part to be tested changes (for example, when the size of the battery cell to be detected changes), the operator can simply and quickly adjust the machine vision The detection system is adjusted so that it can adapt to the changed parts to be tested, which effectively improves the detection efficiency and compatibility.
  • FIG. 4 is a method flowchart of a machine vision inspection method in some embodiments of the present application.
  • the machine vision detection method can be executed by the above-mentioned first controller.
  • the machine vision inspection method includes:
  • S401 Receive a three-dimensional image from a line laser.
  • the three-dimensional image is an image signal containing depth information collected and obtained by the line laser moving relative to the component to be measured.
  • the three-dimensional image includes: at least a part of the first component, at least a part of the second component, and at least one welding spot between the first component and the second component.
  • the above three-dimensional image is specifically determined by the shooting area of the line laser.
  • the line laser can also photograph and collect all the parts to be tested, as long as it can include the solder joints formed after the pre-soldering process, and can meet the needs of the inspection, and there is no limitation here.
  • the three-dimensional image collected by the line laser may be a color image marked with depth information.
  • an appropriate type of pixel conversion method can be used to convert it into a corresponding grayscale image.
  • a part of the border of the first part 510 (such as the aluminum case of the electric core) may be included in the two-dimensional grayscale image
  • a part of the border of the second part 520 and a part of the border between the first part 510 and the second part may be included.
  • Solder spot 530 on the border of 520 The welding spot 530 is located on the first component 510 and the second component 520 at the same time, so as to fix the first component 510 and the second component 520 .
  • obtaining refers to distinguishing the boundaries of the first part and the second part from other parts in the two-dimensional grayscale image, and identifying them in any suitable form.
  • boundary extraction methods such as edge detection algorithms based on autocorrelation functions, edge detection algorithms based on gray level co-occurrence matrix, or boundary fitting methods based on differential thinking.
  • N is a positive integer, indicating the number of vertical lines that need to be set. It can be set by technicians according to actual needs, for example, 20 to 50.
  • a "perpendicular line” is a line segment between the boundary of the first part and the second part, perpendicular to the first straight line or the second straight line. Those skilled in the art can understand that each vertical line represents the gap between the first component and the second component at the position of the vertical line.
  • the length of each vertical line represents the gap between the first component and the second component at the position of the vertical line, and the overall situation of the gap between the two components can be obtained by taking the average value of these vertical lines, which is used to help judge the expected Whether the gap between the first part and the second part after the welding process meets the quality requirements for subsequent laser welding.
  • One of the advantages of embodiments of the present application is that no pre-calibrated data need to be utilized when calculating the gap between two components. Moreover, taking the average value of the lengths of multiple vertical lines as the result of gap detection between two components can eliminate interference well and improve detection accuracy.
  • FIG. 6 is a flowchart of a method for acquiring a first component boundary and a second component boundary in a two-dimensional grayscale image according to an embodiment of the present application.
  • the step S403 of obtaining the boundary of the first component and the boundary of the second component specifically includes:
  • the "boundary area” is an image area including the boundary of the first component. It is a preliminarily demarcated image area, which can be obtained by dividing some marks in the image. For example, the gap between the first part and the second part in a simple two-dimensional grayscale image can be used as a boundary for division.
  • the "first fitting unit” is the sampling window used for fitting, which represents the step size in the fitting process. It can be understood that, for the same first component boundary, the more first fitting units are set, the smaller the length of each first fitting unit is, and the higher the fitting degree is, and vice versa.
  • the first fitting unit is used as a sampling window, which may be a rectangular frame with a certain width in the boundary area.
  • the boundary of the first component extending to the entire boundary area will sequentially pass through N first fitting units to form an intersection with the first fitting unit.
  • a boundary 711 is included in the boundary area 710 .
  • each fitting unit 720 image processing analysis is performed on each fitting unit 720 in turn, and the intersection point 730 between the boundary 711 of the first part and the fitting unit 720 can be found. It can be understood that the shorter the length of the fitting unit, the closer the line segment formed by the connecting line between adjacent intersection points 730 and the first component boundary 711 is to the line segment belonging to the fitting unit. Correspondingly, the degree of fit is higher.
  • fitting unit 720 when presenting the final fitting result, only the formed intersection point 730 and the line segment 740 connecting the two intersection points 730 may be displayed, in a form similar to the line segment connection shown in FIG. 7 b , which is similar to a caliper.
  • fitting unit 720 may also be referred to as a "caliper" in some embodiments.
  • the method of obtaining the boundary of the second component is the same as the method of obtaining the boundary of the first component in the above steps S4031 and S4032, for details, reference may be made to the fitting process shown in Fig. 7a and Fig. 7b.
  • the use of "first" and “second” is only used to distinguish the sampling windows provided on the first component and the second component, and is not used to specifically limit the sampling windows.
  • the method of obtaining the second straight line is similar to the method of obtaining the first straight line, and also the intersection points formed by the fitting units are sequentially connected.
  • N can be set as a positive integer between 20-50.
  • Such a numerical range can also balance the required calculation amount under the condition that the normal detection accuracy requirement is met.
  • Such a fitting straight line generation method can conveniently obtain the required fitting straight line by adjusting the number of sampling units (such as the number of calipers), so as to meet the detection requirement for the gap between the first part and the second part.
  • the step of generating N vertical lines may specifically include:
  • each vertical line setting point is used as a starting point to generate N vertical lines perpendicular to the second straight line.
  • setting N fitting units generally forms N intersection points 730 .
  • the vertical line setting point ie, the intersection point 730
  • N vertical lines can be formed.
  • the caliper or sampling unit used when generating the second straight line is used to set the vertical line accordingly, so that the vertical line can also achieve an even distribution.
  • the spacing between the two vertical lines can have a suitable distance.
  • the vertical line can also be set based on the sampling unit used when generating the second straight line, That is, determine the intersection points between the N second fitting units and the boundary of the second component as the vertical line setting points; respectively use each vertical line setting point as a starting point to generate N vertical lines perpendicular to the first straight line.
  • the vertical lines thus arranged have the same number as the sampling units. Multiple vertical lines can be evenly distributed in the component gap of the two-dimensional image, so that the calculation result of the average value of the vertical line length is closer to the real gap between two components.
  • FIG. 8 is a machine vision detection method provided by another embodiment of the present application. Please refer to FIG. 8.
  • the detection method also The following steps can be performed on the basis of a two-dimensional grayscale image:
  • the "soldering point" may be a welding position used to realize the initial positioning between the first component and the second component after the pre-soldering process is completed.
  • solder joints 530 are located on the first component 510 and the second component 520 as shown in FIG. 5 .
  • the welding site may appear in any type of shape and area in the two-dimensional grayscale image based on the actual welding situation.
  • tangent point refers to the tangent point between the fitted boundary line and the area occupied by the solder point in the two-dimensional grayscale image, and also indicates the intersection point of the solder point between the boundary on this side and the gap. Location.
  • first tangent point and second tangent point are used to distinguish the tangent points located on both sides of the welding point, and are not used to limit the positions of the two tangent points.
  • the width of the solder joint can be regarded as the width of the solder joint above the gap of the two components in the direction of gap extension, which can be represented by the distance between the two tangent points located at the farthest ends. Therefore, the distance between the two tangent points can be calculated as the width of the welding spot to help judge the welding quality of the pre-welding process.
  • One of the advantages of the embodiment of the present application is: on the basis of the two-dimensional grayscale image, a detection method for the width of the solder joint is further provided, which can better ensure the accuracy of the detection result of the pre-soldering process and avoid full welding bad.
  • the step S801 of extracting the third straight line and the fourth straight line may specifically include the following steps:
  • M is an empirical value, which can be set by technicians according to the needs of the actual situation. More fitting units may have a smoother fitting line, and fewer fitting units may have less calculation. quantity.
  • M can be set as a positive integer between 30-50.
  • Such a numerical range can also balance the required calculation amount under the condition that the normal detection accuracy requirement is met.
  • the method for obtaining the boundaries on both sides of the solder joint is the same as the method for obtaining the boundary of the first component and the boundary of the second component described above.
  • the specific implementation process can be referred to as shown in FIG. 7a and FIG. 7b , and will not be repeated here.
  • the following steps can be adopted to help determine the first tangent point and the second tangent point:
  • the intersection point between the third fitting unit and the boundary on one side of the solder joint is used as the first tangent point.
  • the intersection point between the last fitting unit and the boundary among the fitting lines can be selected as the tangent point for calculating the width of the solder joint.
  • the intersection point of the last fitting unit can basically be considered to be located at the extreme end of a segment boundary to be obtained. Therefore, the position of the last intersection point on the third straight line and the fourth straight line is basically the junction of the solder joint and the gap, and then determined as the first tangent point and the second tangent point.
  • the fitted straight line is obtained based on the method shown in FIG.
  • the intersection point between the first straight lines is taken as the first tangent point.
  • the position where the first straight line intersects with the fitted straight lines on both sides of the solder joint can be used as the tangent point for calculating the solder joint width.
  • the above method can also determine the junction position between the welding spot and the gap, and then determine it as the first tangent point and the second tangent point.
  • FIG. 10 is a method flow chart of a method for detecting the gap between pre-welded components and the width of solder joints provided by the embodiment of the present application.
  • Fig. 11 is a schematic diagram of the battery cell after the pre-soldering process provided by the embodiment of the present application.
  • the steps of the component gap and solder joint width detection method include:
  • the single cell flows to the sampling area where the image acquisition device is located along with the fixture.
  • the component to be tested after the pre-soldering process is mainly composed of an aluminum cell shell 910 and a cell top cover 920 .
  • the aluminum cell shell 910 is rectangular and symmetrical on both sides.
  • the cell top cover 920 is surrounded by the cell aluminum case 910 and has an outline close to that of the cell aluminum case.
  • a certain gap 930 exists between the two. Overlying the gap 930 is a plurality of solder joints 940 .
  • the controller After the cell to be detected enters the detection starting position, the controller sends a scanning signal to the image acquisition device.
  • the controller may specifically use any suitable type of sensor (such as an infrared sensor) to determine whether the battery cell has entered the detection starting position.
  • the controller may be a Programmable Logic Controller (PLC) or any other suitable type of electronic processing device.
  • PLC Programmable Logic Controller
  • the driving mechanism drives the cell to move relative to the image acquisition device at a set speed
  • the image acquisition device that receives the scanning signal scans according to the output frequency of the encoder to acquire a three-dimensional image signal.
  • the encoder is a component that feeds back the relative transfer speed of the cell to be tested. Therefore, according to the output frequency of the encoder, the line laser can use a scanning frequency adapted to the relative moving speed of the battery to scan to obtain a three-dimensional image signal of the battery.
  • the line lasers can be arranged in pairs to form the two symmetrical long sides of the aluminum shell 910 and the top cover 920 of the battery as shown in the dotted box 950 in the figure. shooting area. In the photographing area, there are a plurality of welding spots 940 for realizing the initial positioning of the aluminum cell shell 910 and the cell top cover 920 .
  • the controller receives the three-dimensional image signal acquired by the image acquisition device and generates a corresponding two-dimensional grayscale image.
  • the processing operation on the three-dimensional image in the above step S904 can be executed by calling one or more algorithms in the corresponding image software system.
  • a coordinate system can be established according to the positional relationship between the long side and the short side of the aluminum battery case, so as to facilitate subsequent calculation and operation.
  • the long side and short side of the battery aluminum shell can be obtained, and then the intersection point between the long side and the short side is used as the positioning point of the coordinate system, and the rotation angle of the long side and the short side relative to the coordinate system is used as the reference angle, so that Establish a coordinate system in which the y-axis of the coordinate system is parallel to the long side and the x-axis of the coordinate system is parallel to the short side.
  • the controller extracts, from the preprocessed two-dimensional grayscale image, a first straight line fitted to the boundary of the aluminum case of the battery cell and a second straight line fitted to the boundary of the top cover of the second battery cell.
  • the controller can be deployed in the production line or testing site, any suitable type of computing equipment with logic computing capabilities. It runs corresponding image processing software to realize a series of image processing operations on two-dimensional grayscale images.
  • the number of vertical lines to be calculated may be determined by the number of calipers selected to be used when generating the fitted straight line.
  • the intersection point of each sampling unit (ie, the caliper) and the boundary of the part is used as the starting point of the vertical line, and the distance to the fitting line on the other side is calculated.
  • the gap threshold can be set according to actual needs, for example, it is set to 0.08mm.
  • the controller extracts a third straight line and a fourth straight line fitting with boundaries on both sides of the solder joints from the two-dimensional grayscale image.
  • a similar edge extraction algorithm can also be used to obtain a straight line fitted to the boundaries on both sides of the solder joint.
  • the two sides of the solder joint refer to the two sides through which the extending direction of the gap passes.
  • the first tangent point and the second tangent point at the farthest end can be found in various ways to calculate the width of the welding spot. Similar to the gap between the above two components, the width of the welding spot usually needs to be within a certain range to avoid poor welding. In some embodiments, a standard range of weld joint widths may be 3-5mm.
  • the points where the last sampling unit (ie, the caliper) of the third straight line and the fourth straight line intersect with the edge of the welding spot can be used as the two tangent points.
  • two intersection points where the third straight line and the fourth straight line intersect the first straight line may also be used as tangent points respectively.
  • the detection result refers to data information such as calculated component gaps and/or solder joint widths. It can be fed back to the manufacturing execution system, and displayed in any suitable form on a display device such as a display to show the operator in real time.
  • One of the advantages of the embodiments of the present application is that continuous sampling can be supported without stopping at each solder joint position, which improves the detection speed. Moreover, when detecting the components to be tested, the real component gap and solder joint width can be detected in the two-dimensional grayscale image, which is not easily affected by external light sources, etc., and the detection accuracy has been effectively improved.
  • FIG. 12 is a machine vision detection device according to an embodiment of the present application.
  • the machine vision detection device 1100 includes: a receiving module 1110 , a conversion module 1120 , a fitting module 1130 and a gap calculation module 1140 .
  • the receiving module 1110 is used for receiving the three-dimensional image from the line laser.
  • the three-dimensional image includes: at least a portion of the boundary of the first component, at least a portion of the boundary of the second component, and at least one welding point located on the boundary of the first component and the second component.
  • the conversion module 1120 is used to convert the 3D image into a 2D grayscale image.
  • the fitting module 1130 is used to obtain the boundary of the first component and the boundary with the second component in the two-dimensional grayscale image.
  • the gap calculation module 1140 is used to determine N vertical lines between the boundary of the first component and the boundary of the second component; and calculate the average value of the lengths of the N vertical lines as the gap between the first component and the second component, N is positive integer.
  • the receiving module 1110 receives and provides to the converting module 1120 a three-dimensional image comprising two parts and the solder joints overlaid between the parts.
  • the conversion module 1120 converts the 3D image into a 2D grayscale image.
  • the fitting module 1130 performs edge extraction in the two-dimensional grayscale image generated by the conversion module 1120 to obtain the boundaries of the two components.
  • the gap calculation module 1140 obtains the gap between the two components by calculating the length of the vertical line between the boundaries of the two components multiple times and taking the average value.
  • One of the advantages of the embodiment of the present application is that when detecting the gap between components, a more accurate measurement result of the gap can be obtained by adopting the method of averaging multiple times of detection. Moreover, image acquisition based on line lasers can achieve continuous sampling while effectively eliminating a series of interference caused by traditional cameras due to light source occlusion and other factors.
  • the fitting module 1130 may specifically include: a first fitting unit 1131 and a second fitting unit 1132 .
  • the first sampling unit 1131 is used to set 20-50 first fitting units that equally divide the boundary area in the boundary area including the boundary of the first component, and connect each first fitting unit with the first component in turn.
  • the points of intersection between the boundaries form the first straight line.
  • the second fitting unit 1132 is used to set 20-50 second fitting units that equally divide the boundary area in the boundary area including the boundary of the second component; The point of intersection between them forms the second straight line.
  • the gap calculation module 1140 is specifically configured to: determine the N distance between the N first fitting units and the boundary of the first part
  • the intersection points are vertical line setting points, each of the vertical line setting points is used as a starting point to generate N vertical lines perpendicular to the second straight line or determine N between the N second fitting units and the second component boundary
  • the intersection points are vertical line setting points, and each of the vertical line setting points is used as a starting point to generate N vertical lines perpendicular to the first straight line.
  • Such a vertical line setting method is based on the fitting unit that generates a fitting straight line, and generates a plurality of vertical lines with the same number as the fitting unit and evenly distributed, which can be used to realize a gap detection method for multiple detection and averaging.
  • the machine vision inspection device further includes: an edge extraction module 1150 and a solder joint width calculation module 1160 .
  • the edge extraction module 1150 is used for extracting the third straight line and the fourth straight line respectively fitted to the two side boundaries of the solder joint in the two-dimensional grayscale image.
  • Solder joint width calculation module 1160 is used to determine the first tangent point of the 3rd straight line tangent to the solder joint and the second tangent point of the fourth straight line tangent to the solder joint; calculate the distance between the first tangent point and the second tangent point as the width of the solder joint.
  • Such a technical solution further performs automatic detection of the width of solder joints on the basis of two-dimensional grayscale image detection of component gaps, which is conducive to comprehensively evaluating the quality of the pre-soldering process.
  • the edge extraction module 1150 is specifically configured to: respectively set 30-50 third fitting units that equally divide the boundary area on one side of the welding point in the boundary area on one side of the welding point; Connect the intersection points between each third fitting unit and the boundary on one side of the solder joint in turn to form a third straight line, and set 30-50 equally divided solder joints in the boundary area on the other side of the solder joint.
  • the fourth fitting unit in the boundary area sequentially connect the intersection points between each fourth fitting unit and the boundary on the other side of the solder joint to form a fourth straight line.
  • Such a design uses a method similar to component edge extraction to obtain the third straight line and the fourth straight line fitted to the boundaries on both sides of the solder joint, which can help complete the automatic detection of the solder joint width.
  • the welding spot width calculation module 1160 is specifically configured to: connect the last third fitting unit on the third straight line to one side of the welding spot The intersection point between the boundaries is taken as the first tangent point; and the intersection point of the last fourth fitting unit on the fourth straight line and the boundary on the other side of the solder joint is taken as the second tangent point.
  • the position of the last sampling unit of the third straight line and the fourth straight line is used as the two tangent points, and the positions of the two tangent points can be determined simply and quickly.
  • the welding spot width calculation module 1160 is specifically configured to: use the intersection point between the third straight line and the first straight line as the first point of tangency; and the point of intersection between the fourth straight line and the first straight line is taken as the second point of tangency.
  • Such a design utilizes the intersection between the fitting straight line of the component boundary and the fitting straight line on both sides of the solder joint to obtain two tangent points, and can also quickly and conveniently determine the positions of the two tangent points.
  • the functional modules of the machine vision inspection device are divided according to the method steps to be executed.
  • one or more functional modules (such as a receiving module, a conversion module, a fitting module, a gap calculation module, and an edge extraction module) in the machine vision detection device in the embodiments of the present application can be combined according to the needs of the actual situation. module and the welding spot width calculation module) are split into more functional modules to perform corresponding method steps.
  • one or more functional modules in the power exchange device of the embodiments of the present application may also be integrated into fewer functional modules to perform corresponding method steps.
  • FIG. 14 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the electronic device may be a first controller, a second controller or any other suitable type of electronic computing platform for executing the above-mentioned image software system, and its specific implementation is not limited here.
  • the electronic device may include: a processor 1310 , a communication interface 1320 , a memory 1330 and a communication bus 1340 .
  • the processor 1310 , the communication interface 1320 and the memory 1330 communicate with each other through the communication bus 1340 .
  • the communication interface 1320 is used for communication connection with other devices (such as image acquisition devices).
  • the processor 1310 is used to call the program 1350 to execute one or more method steps in the machine vision inspection method in the above-mentioned embodiments or realize one or more functional modules in the machine vision inspection device in the above-mentioned embodiments.
  • the program 1350 may include program codes or computer operation instructions.
  • the processor 1310 may be a central processing unit, other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc.
  • the memory 1330 is used to store the program 1350 .
  • the memory 1330 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.
  • the embodiment of the present application also provides a computer-readable storage medium.
  • the computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • the computer readable storage medium stores a computer program.
  • a complete computer program product is embodied on one or more computer-readable storage media (including but not limited to, disk storage, CD-ROM, optical storage, etc.) containing the computer program disclosed in the embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Image Analysis (AREA)

Abstract

一种机器视觉检测方法、检测装置及检测系统。该检测方法包括：接收来自线激光器的三维图像；将三维图像转换为二维灰度图像；在二维灰度图像中获取第一部件的边界以及第二部件的边界；确定N条位于第一部件的边界和第二部件的边界之间的垂线；计算N条垂线长度的平均值作为第一部件与第二部件之间的间隙。其中，该三维图像中包含：至少一部分的第一部件的边界，第二部件的边界以及至少一个位于第一部件和第二部件的边界上的焊点，N为正整数。该方法通过多次检测求平均值的方式，能够更精确的计算两个部件之间的间隙，提升了检测准确率。而且，其可以支持对检测部件连续采样，并且不需要对相机预先进行标定。

Description

机器视觉检测方法、其检测装置及其检测系统 技术领域
本申请涉及机器视觉领域，具体涉及一种机器视觉检测方法、其检测装置及其检测系统。
背景技术
在生产过程之中，工件焊接后检测通常使用“目视检测”来完成。目视检测是以肉眼观察为主，并结合利用放大镜、量具以及样板等辅助工具对焊缝表面质量以及目视尺寸等进行全面检查的方法。
为了提升人工检测的效率，现有一些基于工业相机实现的机器视觉检测方案。但是这些机器视觉检测方案受限于结构设计、成本控制以及与实际应用场景的匹配情况等问题，仍然存在不少的缺陷。其检测准确率和检测效率上都有待提升。
发明内容
鉴于上述问题，本申请提供一种机器视觉检测方法、其检测装置及其检测系统，能够提升机器视觉检测的效率和准确率。
第一方面，本申请提供了一种机器视觉检测方法。该方法包括：接收来自线激光器的三维图像，三维图像中包含：至少一部分的第一部件的边界，至少一部分的第二部件的边界以及至少一个位于第一部件的边界和第二部件边界上的焊点；将三维图像转换为二维灰度图像；在二维灰度图像中获取第一部件的边界以及第二部件的边界；确定N条位于第一部件边界和第二部件边界之间的垂线；计算N条垂线长度的平均值作为第一部件与第二部件之间的间隙，N为正整数。
本申请实施例的技术方案中，通过线激光器采集的三维图像，可以支持对检测部件连续采样，并且不需要对相机预先进行标定。另外，通过求多个垂线长度平均值的方式，能够更精确的计算两个部件之间的间隙，提 升了检测准确率。
在一些实施例中，上述在二维灰度图像中获取第一部件的边界和第二部件的边界的步骤具体包括：在包含第一部件边界的边界区域中，设置N个等分所述边界区域的第一拟合单元；依次连接每个第一拟合单元与第一部件边界之间的交点，形成与第一部件边界拟合的第一直线；在包含第二部件边界的边界区域中，设置N个等分边界区域的第二拟合单元；依次连接每个第二拟合单元与第二部件边界之间的交点，形成与第二部件边界拟合的第二直线；其中，N为20-50之间的正整数。在上述技术方案中，通过设置多个均分边界区域的拟合单元来进行拟合获得边界直线，可以方便的通过调整拟合单元数量以调整拟合直线的拟合程度，从而满足不同实际情况的需要。
在一些实施例中，上述生成N条位于第一部件的边界和第二部件的边界之间的垂线的步骤具体包括：确定N个第一拟合单元与第一部件边界之间的交点为垂线设置点；分别以每个垂线设置点为起点，生成与第二直线垂直的N条垂线；或者确定N个第二拟合单元与所述第二部件边界之间的交点为垂线设置点；分别以每个垂线设置点为起点，生成与第一直线垂直的N条垂线。在上述技术方案中，以拟合单元与部件边界之间的交点为基础来生成垂线，可以使得N条垂线沿间隙均匀分布，从而更精确的计算两个部件之间的间隙。
在一些实施例中，上述方法还包括：在二维灰度图像中提取分别与焊点的两侧边界拟合的第三直线和第四直线；确定第三直线与焊点相切的第一切点以及第四直线与焊点相切的第二切点；计算第一切点和第二切点之间距离作为焊点的宽度。
本申请实施例的技术方案中，还提供了基于三维图像计算两个部件之间的焊点宽度的方式。其可以通过获得与焊点边界拟合的拟合直线而自动计算获得，用以帮助更全面的评估部件之间的焊接状态。
在一些实施例中，上述在二维灰度图像中提取分别与焊点的两侧边界拟合的第三直线和第四直线的步骤具体包括：在包含焊点的第一侧边界的边界区域中，分别设置M个等分边界区域的第三拟合单元；依次连接每个第三拟合单元与焊点的第一侧边界之间的交点，形成所述第三直线，并且；在包含焊点的第二侧边界的边界区域中，分别设置M个等分边界区域的第四拟合单 元；依次连接每个第四拟合单元与焊点的第二侧边界之间的交点，形成第四直线；其中，M为30-50之间的正整数。
在上述技术方案中，通过设置多个均分边界区域的拟合单元的方式来拟合获得焊点的边界，可以方便的通过调整拟合单元数量以调整拟合直线与焊点边界的拟合程度，从而满足不同实际情况的需要。
在一些实施例中，上述确定第三直线与焊点相切的第一切点以及第四直线与焊点相切的第二切点的步骤具体包括：将第三直线上最后一个第三拟合单元与焊点的第一侧边界之间的交点作为第一切点；并且将第四直线上最后一个第四拟合单元与焊点的第二侧边界的交点作为第二切点。在上述技术方案中，利用拟合单元均分边界区域的特点，可以简单快速的通过拟合直线的最后一个拟合单元来近似确定第一切点和第二切点。
在一些实施例中，上述确定第三直线与焊点相切的第一切点以及第四直线与焊点相切的第二切点的步骤具体包括：将第三直线与第一直线之间的交点作为第一切点；并且将第四直线与第一直线的之间交点作为第二切点。在上述技术方案中，考虑到焊点的宽度基本上可以被认为是处于两个部件的间隙的宽度，由此提供了基于部件的拟合直线与焊点两侧的拟合之间的交点来快速的确定用于计算焊点宽度的两个切点，能够帮助获得准确的焊点宽度计算结果。
第二方面，本申请提供了一种机器视觉检测装置。该机器视觉检测装置包括：接收模块，用于接收来自线激光器的三维图像，三维图像中包含：至少一部分的第一部件的边界，至少一部分的第二部件的边界以及至少一个位于第一部件边界和第二部件边界上的焊点；转换模块，用于将三维图像转换为二维灰度图像；拟合模块，用于在二维灰度图像中获取第一部件的边界以及与第二部件的边界；间隙计算模块，用于确定N条位于第一部件边界和第二部件边界之间的垂线；并且计算N条垂线长度的平均值作为第一部件与第二部件之间的间隙，N为正整数。
本申请实施例的技术方案中，通过线激光器采集的三维图像，可以支持对检测部件连续采样，并且不需要对相机预先进行标定。另外，该装置能够通过多条垂线求平均值的方式来计算两个部件之间的真实间隙，有效的排除了因图像畸变等造成的干扰。由此，有效的提升检测速度和检测准 确率。
第三方面，本申请提供了一种电子设备。该电子设备包括：处理器以及与处理器通信连接的处理器；存储器存储有计算机程序指令，计算机程序指令在被处理器调用时，以使处理器执行如上的视觉检测方法。
本申请实施例的技术方案中，电子设备通过线激光器采集的三维图像，可以支持对检测部件连续采样，并且不需要对相机预先进行标定。另外，能够通过多条垂线求平均值的方式来计算两个部件之间的真实间隙，有效的排除了因图像畸变等造成的干扰。由此，有效的提升检测速度和检测准确率。
第四方面，本申请提供了一种非易失性计算机存储介质。其中，该非易失性计算机存储介质存储有计算机程序指令，以使计算机程序指令被处理器调用时，执行如上的视觉检测方法。
本申请实施例的技术方案中，通过线激光器采集的三维图像，可以支持对检测部件连续采样，并且不需要对相机预先进行标定。另外，能够通过多条垂线求平均值的方式来计算两个部件之间的真实间隙，有效的排除了因图像畸变等造成的干扰。由此，有效的提升检测速度和检测准确率。
第五方面，本申请提供了一种机器视觉检测系统。该机器视觉检测系统包括：包括若干线激光器的图像采集设备，用于采集三维图像；驱动机构，用于使图像采集设备与待测部件之间发生相对移动；与图像采集设备通信连接的第一控制器，第一控制器用于执行如上的机器视觉检测方法处理三维图像，以使三维图像的处理结果用于待测部件的检测。
本申请实施例的技术方案中，可以支持对检测部件连续采样，并且不需要对相机预先进行标定，能够有效的提升检测的准确率和检测效率。另一方面，线激光器能够直观的显示检测区域的特征，可以显著提升采样图像的检测效果，可以很好的提升检测准确率。
在一些实施例中，图像采集设备包括：两个线激光器、传感器支架以及遮光罩；两个线激光器分别设置在传感器支架的两侧；遮光罩固定在传感器支架上，罩套在线激光器外；传感器支架包括：高度调节模组和间距调节模组；高度调节模组用于调节线激光器所在的高度；间距调节模组用于调节两个线激光器之间的间距。
本申请实施例的技术方案中，还设置了额外的高度调节模组和间距调节模组，从而使机器视觉检测系统能够在待测部件的尺寸发生变化时适应性的进行调整，满足多种待测部件的使用需要。而且，额外设置的遮光罩可阻止线激光器的激光向外部散射，避免操作人员的眼睛受到伤害。
在一些实施例中，该机器视觉检测系统还包括：第二控制器，第二控制器用于控制高度调节模组和间距调节模组，以使两个线激光器达到目标间距和/或目标高度；第二控制器存储有若干个记录目标间距和目标高度的配置信息；每个配置信息与至少一种待测部件对应。在上述技术方案中，由于在第二控制器中可以预先记录不同型号、尺寸或者外形的待测部件的配置信息。因此，可以在待测部件的尺寸、型号或者外形等发生变化时，技术人员可以通过加载相应的配置信息来实现自动切换和快速调整，有效的提升了检测系统的兼容性和运行效率。
上述说明仅是本申请技术方案的概述，为了能够更清楚了解本申请的技术手段，而可依照说明书的内容予以实施，并且为了让本申请的上述和其它目的、特征和优点能够更明显易懂，以下特举本申请的具体实施方式。
附图说明
通过阅读对下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本申请的限制。而且在全部附图中，用相同的附图标号表示相同的部件。在附图中：
图1为本申请一些实施例的机器视觉检测系统的结构示意图；
图2为本申请一些实施例的图像采集设备的结构示意图；
图3为本申请另一些实施例的机器视觉检测系统的结构示意图；
图4为本申请一些实施例的机器视觉检测方法的方法流程图；
图5为本申请一些实施例的二维灰度图的示意图；
图6为本申请一些实施例的步骤S403的方法流程图；
图7a为本申请一些实施例的获取部件边界的示意图，示出了均分边界区域的拟合单元；
图7b为本申请一些实施例的边界的示意图，示出了图7a所获取的部 件边界的展示形式；
图8为本申请另一些实施例的机器视觉检测方法的方法流程图；
图9为一些实施例的步骤S801的方法流程图；
图10为本申请一些实施例机器视觉检测方法的方法流程图，示出了待测部件为预焊工序执行完毕后的电芯顶盖和电芯铝壳；
图11为本申请一些实施例的待测部件的示意图，示出了图9进行检测的电芯顶盖和电芯铝壳；
图12为本申请一些实施例的机器视觉检测装置的示意图；
图13为本申请另一些实施例的机器视觉检测装置的示意图；
图14为本申请一些实施例的电子设备的示意图。
具体实施方式
下面将结合附图对本申请技术方案的实施例进行详细的描述。以下实施例仅用于更加清楚地说明本申请的技术方案，因此只作为示例，而不能以此来限制本申请的保护范围。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同；本文中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本申请；本申请的说明书和权利要求书及上述附图说明中的术语“包括”和“具有”以及它们的任何变形，意图在于覆盖不排他的包含。
在本申请实施例的描述中，技术术语“第一”“第二”等仅用于区别不同对象，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量、特定顺序或主次关系。在本申请实施例的描述中，“多个”的含义是两个以上，除非另有明确具体的限定。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
在本申请实施例的描述中，术语“和/或”仅仅是一种描述关联对象的 关联关系，表示可以存在三种关系，例如A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
在本申请实施例的描述中，术语“多个”指的是两个以上(包括两个)，同理，“多组”指的是两组以上(包括两组)，“多片”指的是两片以上(包括两片)。
在本申请实施例的描述中，技术术语“中心”“纵向”“横向”“长度”“宽度”“厚度”“上”“下”“前”“后”“左”“右”“竖直”“水平”“顶”“底”“内”“外”“顺时针”“逆时针”“轴向”“径向”“周向”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请实施例和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本申请实施例的限制。
在本申请实施例的描述中，除非另有明确的规定和限定，技术术语“安装”“相连”“连接”“固定”等术语应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或成一体；也可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通或两个元件的相互作用关系。对于本领域的普通技术人员而言，可以根据具体情况理解上述术语在本申请实施例中的具体含义。
目前，在锂电池电芯生产流程中，在电芯入壳后，需要将电芯的顶盖和铝壳进行焊接。而在电芯的顶盖和铝壳满焊之前，存在一个初步焊接顶盖和铝壳，实现两者之间初定位的预焊工序。而为了确保电芯的顶盖和铝壳的焊接质量，避免出现两者间隙过大等问题导致焊接不良的情况，需要对预焊工序后的顶盖和铝壳的间隙以及焊点质量等进行检测。
典型的机器视觉检测方式是采用二维相机，采集顶盖和铝壳在预焊工序后的焊接部分的图像信息，然后对这些图像信息进行处理分析从而确定两者之间的间隙是否符合要求。
本申请人注意到，在采用上述机器视觉检测方式时，需要在使用前对二维相机预先进行标定，生成像素距离到实际距离的校正文件来帮助计算。这样无法获得顶盖与铝壳之间的真实间距，在物体特征不明显时也容易导 致误判。而且，二维相机拍摄图像时会因光源设置问题而导致检测不精确和测量误判等的问题，所需要的检测间隙时间较长。
为了解决上述机器视觉检测效率较低而且检测精度不足的问题，申请人提供了基于线激光器实现的部件间隙检测方法，可以有效的解决传统二维相机需要预先进行标定，采集图像信息存在较多干扰而使得检测精度和检测效率较低的缺陷。
以下实施例为了方便说明，以本申请实施例的待测部件为预焊工序结束后的顶盖和铝壳为例进行说明。当然，本领域技术人员可以理解的是，基于相同的原理和构思，还可以将本申请实施例的机器视觉检测系统应用于其他具有相似结构形状特点的待测部件中进行检测。
请参阅图1，图1为本申请一些实施例的机器视觉检测系统的结构示意图。该机器视觉检测系统包括：图像采集设备110，驱动机构120以及第一控制器130。
其中，图像采集设备110是用于采集三维图像信号的设备。其具体可以选择使用任何合适型号和数量的线激光器，并具有与线激光器相适配支撑结构。
驱动机构120是用于驱动待测部件(例如，预焊工序结束后的电芯顶盖和电芯铝壳)与图像采集设备110之间发生相对移动的动作单元。其具体可以采用任何合适类型的动力机构，以满足图像采集设备110的连续采样需要。在一些实施例中，待测部件可以被夹持固定在驱动机构120上，在驱动机构120的带动下，相对于图像采集设备110的线激光器移动，从而完成对待测部件的图像采样。
第一控制器130可以是具有逻辑运算功能的电子计算设备，包括但不限于服务器或者工业计算机等。其可以通过有线或者无线的方式与图像采集设备之间建立通信连接，从而接收图像采集设备采集获得的三维图像信号。
在操作中，预焊工序执行完毕后的电芯顶盖A1和电芯铝壳A2可以在电机或者其他合适类型的驱动机构120的驱动下，以设定的速度与图像采集设备110之间发生相对移动。图像采集设备110的线激光器可以通过编码器等类似的传感器设备，按照与相对移动速度相适配的采集频率，连续采集获得由预焊后的电芯顶盖A1和电芯铝壳A2组成的待测部件在两侧 长边的三维图像信号。
图像采集设备110采集获得的三维图像信号被提供至第一控制器130中，由第一控制器130进行图像处理等一系列机器视觉检测方法的步骤后，输出检测结果并提供至外部设备。由此，可以及时的筛选出间隙或者焊点质量不符合要求的待测部件并进行相应的处理。
本申请实施例的其中一个有利方面是：在长条形的待测部件具有多个焊点的情况下，图像采集设备可以采样连续采集的方式获取待测部件的三维图像信号。而在采用传统相机拍摄的技术方案中，由于相机拍摄的范围有限，因此只能在拍摄完一部分区域后，重新移动到新的位置进行拍摄以完成对全部焊点的拍摄采样。相对于使用传统相机拍摄的技术方案而言，线激光器连续扫描的方式可以有效的减少启动和停止的动作频次，从而极大的提升了检测速度。
根据本申请一些实施例，可选地，请参阅图2，该图像采集设备110可以包括：两个线激光器111、传感器支架112、高度调节模组113、间距调节模组114以及遮光罩115。
其中，两个线激光器111分别设置在传感器支架112的两侧，可以用于同时采集电芯顶盖和电芯铝壳两个对称的长边的三维图像信号。线激光器可以根据实际情况的需要而具有合适的视野和像素精度。例如，线激光器111在扫描方向的分辨率可以设置为小于间隙检测阈值(如0.08mm)以满足检测需要。其扫描线速度设置在130mm/s以上，扫描频率在5kHz左右。
高度调节模组113和间距调节模组114均设置在传感器支架112上。其具体可以选择使用任何合适类型的机械结构实现，包括但不限于基于螺杆、气缸或者齿轮等。
通过高度调节模组113和间距调节模组114可以使两个线激光器111的高度和之间的间距都能够在一定范围内变动，以满足不同型号或者尺寸的电芯的检测需要。
遮光罩115可以设置在传感器支架112上，从而将线激光器111罩套于其中。其具体可以采用任何合适的形状、尺寸或者材质的罩套，只需要能够罩套遮盖线激光器111即可。这样的设计可以避免线激光器产生的激 光泄露或者反射到操作人员的眼睛，达到保护人体的效果。
应当说明的是，本申请实施例中以两个对称设置的线激光器为例进行描述。本领域技术人员可以理解，这样的设计是与获取待测部件两侧的焊接区域相适应的。在待测部件的焊接区域位置或者大小发生变化时，还可以根据实际情况的需要而选择设置不同数量或者设置不同位置的线激光器，而不限于图2所示的情形。
本申请实施例的其中一个有利方面是：通过线激光器完成对待测部件的数据采集，可以有效的避免传统相机获取图像不清晰，图像存在畸变或者因光源设置而导致图像缝隙与实际值相比偏小等的缺陷。
本申请实施例的另一个有利方面是：线激光器提供的是三维图像信号，可以实现多角度多方位的测量，避免视野遮挡盲区导致的测量误判，也能够提供更加精确和直观的图像信息。
根据本申请的一些实施例，可选地，请参阅图3，图3为本申请另一些实施例提供的机器视觉检测系统的结构示意图。该机器视觉检测系统还可以包括第二控制器140。
其中，该第二控制器140中存储有若干个记录目标间距和目标高度的配置信息。该配置信息是与待测部件对应的数据信息，可以根据实际的产品生产情况由技术人员预先设置。
在操作中，当进入机器视觉检测系统的待测部件发生变化时，技术人员或者操作者可以选择确定与目前需要进行检测的待测部件相对应的配置信息，然后由第二控制器依据选定的配置信息，自动控制高度调节模组113和间距调节模组114将线激光器移动到配置信息记录的目标间距和目标高度，以便于完成对待测部件的三维图像信号采集。
应当说明的是，为陈述方便，在本申请实施例中按照控制器所要执行的功能的不同而分别以“第一控制器”和“第二控制器”进行描述。本领域技术人员可以理解，该第一控制器和第二控制器的描述并不用于对控制器的具体实现进行限定，其既可以是同一个电子计算设备中的不同功能模块，也可以是分别布置在不同电子计算设备中的功能模块。
本申请实施例的其中一个有利方面是：通过预先存储的配置信息，操作者可以在待测部件的尺寸或者类型发生变化时(如需要检测的电芯尺寸 改变时)，简单快捷的对机器视觉检测系统进行调整，使其能够与变化后的待测部件相适配，有效的提升了检测效率和兼容性。
根据本申请的一些实施例，图4为本申请一些实施例的机器视觉检测方法的方法流程图。该机器视觉检测方法可以由上述第一控制器所执行。请参阅图4，该机器视觉检测方法包括：
S401、接收来自线激光器的三维图像。
其中，该三维图像是由线激光器相对于待测部件移动而采集获得的包含深度信息的图像信号。在本实施例中，该三维图像中之中包含了：至少一部分的第一部件，至少一部分的第二部件以及至少一个位于第一部件和第二部件之间的焊点。
当然，上述三维图像具体由线激光器的拍摄区域所决定。在另一些实施例中，线激光器也可以拍摄采集全部的待测部件，只需要能够将预焊工序后形成焊点包括在内，能够满足检测的需要即可，在此不作限定。
S402、将三维图像转换为二维灰度图像。
其中，线激光器采集得到的三维图像可以是标记了深度信息的彩色图像。为了便于后续处理操作，可以采用合适类型的像素转换方式，将其转换为对应的灰度图像。
例如，请参阅图5，在二维灰度图像中可以包括第一部件510(如电芯铝壳)边界的一部分，第二部件520边界的一部分以及位于在第一部件510边界和第二部件520边界上的焊点530。焊点530同时位于第一部件510和第二部件520之上，以起到对第一部件510和第二部件520的固定作用。
S403、在二维灰度图像中获取第一部件的边界和第二部件的边界。
其中，“获取”是指在二维灰度图像中，将第一部件和第二部件的边界与其他部分相区分，通过任何合适的形式标识。在实际操作中，具体可以采用多种不同的边界提取方法，例如基于自相关函数的边缘检测算法，基于灰度共生矩阵的边缘检测算法或者基于微分思想的边界拟合方法。
S404、确定N条位于第一部件边界和第二部件边界之间的垂线。
其中，N为正整数，表示了需要设置的垂线数量。其可以根据实际情况的需要而由技术人员设置，例如20到50。
“垂线”是位于第一部件和第二部件的边界之间，与第一直线或者第二 直线垂直的线段。本领域技术人员可以理解，每个垂线表示了第一部件和第二部件在垂线所在位置的间隙。
S405、计算N条垂线长度的平均值作为第一部件与第二部件之间的间隙。
其中，每条垂线长度表示了第一部件和第二部件在垂线所在位置的间隙，通过取这些垂线的平均值就可以得到两个部件之间间隙的总体情况，用以帮助判断预焊工序后的第一部件和第二部件之间的间隙是否满足后续激光焊接的质量要求。
本申请实施例的其中一个有利方面是：在计算两个部件之间的间隙时，不需要利用预先标定的数据。而且，以对多条垂线长度的平均值作为两个部件的间隙检测结果，可以很好的排除干扰以提高检测精度。
根据本申请一些实施例，请参阅图6，图6为本申请实施例提供的在二维灰度图像中获取第一部件边界和第二部件边界的方法流程图。该获取第一部件边界和第二部件边界的步骤S403具体包括：
S4031、在包含第一部件边界的边界区域中，设置N个等分边界区域的第一拟合单元。
其中，“边界区域”是将第一部件边界包含在内的图像区域。其是一个初步划定的图像区域，可以通过一些图像中的标识进行划分而获得。例如，可以简单的二维灰度图像中第一部件和第二部件之间的间隙为界进行划分。
“第一拟合单元”是用于拟合的取样窗，表示了拟合过程中的步长。可以理解，对于同一个第一部件边界而言，设置的第一拟合单元越多，每一个第一拟合单元的长度就越小，拟合程度也越高，反之亦然。
S4032、依次连接每个第一拟合单元与第一部件的边界之间的交点，形成与第一部件边界拟合的第一直线。
其中，第一拟合单元作为取样窗，其在边界区域中可以是一个具有一定宽度的矩形框。延伸至整个边界区域的第一部件边界会依次穿过N个第一拟合单元，形成与第一拟合单元之间的交点。
为充分说明上述基于微分思想的边界拟合过程，以下结合图7a和图7b对基于拟合单元获取在二维灰度图像中特定一段边界的具体过程进行描述。请参阅图7a，在边界区域710中包含了一段边界711。
首先，基于微分的思路，边界区域710中可以设置有多个将边界区域均 分的拟合单元720。
然后，依次对每一个拟合单元720进行图像处理分析，都可以找到第一部件的边界711与拟合单元720之间的交点730。可以理解，拟合单元的长度越短，相邻交点730之间的连线形成的线段与第一部件边界711在属于该段拟合单元的线段也就越接近。相应地，拟合程度也就越高。
最后，将这些交点730依次连接起来，就可以获得与第一部件边界拟合的第一直线。
在一些实施例中，在呈现最终拟合结果时，可以只显示形成的交点730以及连接两个交点730之间的线段740，呈现类似图7b所示的线段连接的形式，与卡尺相近。由此，拟合单元720也可以在一些实施例中被称为“卡尺”。
S4033、在包含第二部件边界的边界区域中，设置N个等分边界区域的第二拟合单元。
其中，获取第二部件边界的方式与上述步骤S4031和S4032获取第一部件边界的方式相同，具体可以参考图7a和图7b所示的拟合过程。在本实施例中，使用“第一”和“第二”仅用于区分在第一部件和第二部件上设置的取样窗，而不用于对取样窗进行具体的限定。
S4034、依次连接每个第二拟合单元与第二部件的边界之间的交点，形成与第二部件的边界拟合的第二直线。
其中，获取第二直线的方式与获取第一直线的方式类似，同样也是将拟合单元形成的交点依次连接即可。
应当说明的是，图6中为方便展示而对获取第一部件边界和第二部件边界之间的步骤进行了编号，但本领域技术人员可以理解，该编号的顺序并不用于限定步骤的执行次序。
在一些实施例中，令人惊喜的发现，可以将N设置为20-50之间的正整数。这样的数值范围在满足正常检测精度要求的情况下，也可以平衡所需要的计算量。
这样的拟合直线生成方式可以通过调整取样单元的数量(如卡尺数)来方便的获得所需要的拟合直线，满足对第一部件和第二部件之间间隙的检测需要。
根据本申请实施例，可选地，在基于N个拟合单元拟合生成的第一直 线和第二直线的基础上，生成N条垂线的步骤具体可以包括：
首先，确定N个第一拟合单元与第一部件边界之间的交点为垂线设置点。然后，分别以每个垂线设置点为起点，生成与第二直线垂直的N条垂线。
其中，请参阅图7a和图7b，由于拟合单元720是用于均分边界区域的取样窗，在相邻的拟合单元中其中一个交点重合。因此，设置N个拟合单元通常形成了N个交点730。垂线设置点(即交点730)是每条垂线起始点，最终就可以形成N条垂线。
在本申请实施例中利用生成第二直线时所使用到的卡尺或者取样单元来相应的设置垂线，以使得垂线也能够实现平均分布。两条垂线之间的间隔能够具有合适的距离。
当然，除了以生成第二直线时所使用到的卡尺或者取样单元作为垂线设置的基准以外，在另一些实施例中也可以基于生成第二直线时所使用到的取样单元来设置垂线，即确定N个第二拟合单元与第二部件的边界之间的交点为垂线设置点；分别以每个垂线设置点为起点，生成与第一直线垂直的N条垂线。
这样设置的垂线具有与取样单元相同的数量。多条垂线能够均匀的分布在二维图像的部件间隙之中，使得垂线长度平均值的计算结果更贴近两个部件之间的真实间隙。
根据本申请一些实施例，图8为本申请另一实施例提供的机器视觉检测方法，请参阅图8，除了能够实现对第一部件和第二部件之间的间隙检测以外，该检测方法还可以在二维灰度图像的基础上执行如下的步骤：
S801、在二维灰度图像中提取分别与焊点的两侧边界拟合的第三直线和第四直线。
其中，“焊点”可以是预焊工序结束后，用于实现第一部件和第二部件之间初定位的焊接位点。例如，如图5所示位于第一部件510和第二部件520上焊点530。当然，该焊接位点基于实际的焊接情况，在二维灰度图像中可能呈现出任何类型的形状和面积。
基于与部件边缘提取相类似的方式，也可以采用合适的图像处理步骤对焊点进行边缘提取来获得焊点两侧的边界、。应当说明的是，“第三直线”和“第四直线”仅用于区分位于焊点两侧不同位置的拟合直线，而不用于对直 线的长度或者拟合方式等具体实现进行限定。
S802、确定第三直线与焊点相切的第一切点以及第四直线与焊点相切的第二切点。
其中，“切点”是指拟合的边界直线与焊点在二维灰度图像中所占据区域之间的切点，也表示了焊点在这一侧边界之中与间隙之间相交的位置。在本实施例中，使用“第一切点”和“第二切点”来区分表示位于焊点两侧的切点，而不用于对两个切点所处的位置进行限定。
S803、计算第一切点和第二切点之间距离作为焊点的宽度。
其中，焊点的宽度可以认为是在两个部件的间隙之上的焊点在间隙延伸方向上的宽度，其可以通过位于最远端的两个切点之间的距离来表示。由此，可以通过计算两个切点之间的距离作为焊点宽度，用以帮助判断预焊工序的焊接质量。
本申请实施例的其中一个有利方面是：在二维灰度图像的基础上，还进一步提供了焊点宽度的检测方法，可以更好的确保预焊工序的检测结果的准确性，避免满焊不良。
根据本申请一些实施例，可选地，请参阅图9，提取第三直线和第四直线的步骤S801具体可以包括如下步骤：
S8011、在包含焊点的第一侧边界的边界区域中，分别设置M个等分边界区域的第三拟合单元。
S8012、依次连接每个第三拟合单元与焊点的第一侧边界之间的交点，形成第三直线。
其中，M是一个经验性数值，可以由技术人员根据实际情况的需要而设置，更多的拟合单元数量可能具有更平滑的拟合直线，更少的拟合单元数量可能具有更少的计算量。
在一些实施例中，令人惊喜的发现，可以将M设置为30-50之间的正整数。这样的数值范围在满足正常检测精度要求的情况下，也可以平衡所需要的计算量。
S8013、在包含焊点的第二侧边界的边界区域中，分别设置M个等分边界区域的第四拟合单元。
S8014、依次连接每个第四拟合单元与焊点的第二侧边界之间的交点，形 成第四直线。
其中，获取焊点两侧边界的方法与上述获取第一部件边界和第二部件边界的方法相同，具体实现过程可以参考图7a和图7b所示，在此不作赘述。
在本申请实施例中，同样也可以通过调整拟合单元的数量(如卡尺数)来方便的调整焊点两侧边界的拟合程度，以获得所需要的拟合直线，满足对焊点宽度的检测需要。
在一些实施例中，可选地，在基于图9所示的方法获得拟合直线时，可以采用如下步骤来帮助确定第一切点和第二切点:首先，将第三直线上最后一个第三拟合单元与焊点一侧边界之间的交点作为第一切点。其次，将第四直线上最后一个第四拟合单元与焊点另一侧边界的交点作为第二切点。换言之，可以选择拟合直线之中最后一个拟合单元与边界之间的交点作为计算焊点宽度的切点。
例如图7b所示的，最后一个拟合单元的交点基本上可以被认为位于需要获取的一段边界的最末端。由此，第三直线和第四直线上最后一个交点的位置基本上就是焊点与间隙的交界处，进而确定为第一切点和第二切点。
在另一些实施例中，可选地，在基于图9所示的方法获得拟合直线时，确定第一切点和第二切点的方法还可以采用如下步骤：首先，将第三直线与第一直线之间的交点作为第一切点。其次，将第四直线与第一直线的之间交点作为第二切点。换言之，可以将第一直线与焊点两侧的拟合直线之间相交的位置作为计算焊点宽度的切点。上述方式同样也可以确定焊点与间隙之间的交界位置，进而确定为第一切点和第二切点。
根据本申请实施例，请参阅图10和图11，图10是本申请实施例提供的预焊后部件间隙和焊点宽度检测方法的方法流程图。图11是本申请实施例提供的经过预焊工序后的电芯的示意图。该部件间隙和焊点宽度检测方法的步骤包括：
S901、经过压装工位后的单个电芯随夹具流向图像采集设备所在的采样区域。
其中，请参阅图11，经过预焊工序以后的待测部件主要由电芯铝壳910和电芯顶盖920组成。电芯铝壳910呈长矩形并且两边对称。电芯顶盖920被包围在电芯铝壳910之内，具有与电芯铝壳相接近的外形轮廓。两者之 间存在一定的间隙930。在间隙930之上覆盖有多个焊点940。
S902、在待检测的电芯进入检测起始位置后，由控制器发送扫描信号至图像采集设备。
其中，控制器具体可以通过任何合适类型的传感器(如红外传感器)来确定电芯是否进入到检测起始位置。控制器可以是可编程逻辑控制器(Programmable Logic Controller,PLC)或者其他任何合适类型的电子处理设备。
S903、在驱动机构带动电芯以设定的速度相对于图像采集设备移动的同时，接收到扫描信号的图像采集设备依据编码器的输出频率进行扫描以采集三维图像信号。
其中，编码器是反馈待测电芯的相对移送速度的部件。由此，线激光器可以根据编码器的输出频率，使用与电芯的相对移动速度相适配的扫描频率来扫描获得电芯的三维图像信号。
在一些实施例中，请继续参阅图11，线激光器可以成对的设置，从而形成图中虚线框950所示的，覆盖电芯铝壳910和电芯顶盖920对称的两个长边的拍摄区域。在拍摄区域内，包含了多个用于实现电芯铝壳910和电芯顶盖920初定位的焊点940。
S904、控制器接收由图像采集设备采集获得的三维图像信号并生成对应的二维灰度图像。
其中，上述步骤S904之中对三维图像的处理操作可以在相应的图像软件系统中，通过调用一种或者多种算法来执行。在一些实施例中，可以根据电芯铝壳的长边和短边的位置关系来建立坐标系，便于后续的计算和操作。
例如，可以获取电芯铝壳的长边和短边，然后以长边和短边之间的交点作为坐标系的定位点，长边与短边相对于坐标系的旋转角度作为参考角度，从而建立一个坐标系的y轴与长边平行，坐标系的x轴与短边平行的坐标系。
S905、控制器在预处理后的二维灰度图像中提取与电芯铝壳的边界拟合的第一直线以及与第电芯顶盖的边界拟合的第二直线。
其中，控制器可以是部署在产线或者检测现场的，具有逻辑运算能力 的任何合适类型的计算设备。其运行有相应的图像处理软件以实现对二维灰度图像的一系列图像处理操作。
S906、通过第一直线和第二直线之间的多条垂线计算电芯铝壳和电芯顶盖之间的间隙。
其中，计算的垂线数量可以由生成拟合直线时选择使用的卡尺数量来决定。换言之，每个取样单元(即卡尺)与部件边界的交点均用以作为垂线的起点，计算其到另一侧拟合直线的距离。这样检测多次计算平均值的方式有利于提供更为精确的间隙检测结果。
在一些实施例中，通过间隙检测结果和预先设定的间隙阈值的比较结果，可以确定经过预焊工序后的待测部件是否符合要求。该间隙阈值可以根据实际情况的需要而设置，例如设置为0.08mm。
S907、控制器在二维灰度图像中提取与焊点两侧边界拟合的第三直线和第四直线。
其中，除了提取电芯顶盖和电芯铝壳的边缘以外，还可以采用相类似的边缘提取算法，获得焊点两侧边界拟合的直线。请参阅图11，焊点的两侧是指间隙延伸方向所经过的两侧。
S908、通过第三直线和第四直线形成的第一切点和第二切点计算焊点宽度。
其中，可以通过多种方式找到位于最远端的第一切点和第二切点从而计算焊点宽度。与上述两个部件之间的间隙相类似的，焊点宽度通常需要处于在一定的范围内，以避免造成焊接不良的情况。在一些实施例中，焊点宽度的标准范围可以是3-5mm。
在一些实施例中，可以由第三直线和第四直线最后一个取样单元(即卡尺)与焊点边缘相交的点作为两个切点。在另一些实施例中，也可以分别将第三直线和第四直线与第一直线相交的两个交点作为切点。
S909、将检测结果上传至制造执行系统(manufacturing execution system,MES)。
其中，检测结果是指经过计算得到的部件间隙和/或焊点宽度等数据信息。其可以反馈给制造执行系统，并且以任何合适的形式呈现在显示器等显示设备中实时向操作人员展示。
本申请实施例的其中一个有利方面是：可以支持连续采样而不需要在每个焊点位置停顿，提升了检测速度。而且，在检测待测部件时，能够在二维灰度图像中检测真实的部件间隙和焊点宽度，不容易受到外部光源等的影响，检测准确性得到了有效的提升。
根据本申请一些实施例，请参见图12，图12为本申请实施例的机器视觉检测装置。该机器视觉检测装置1100包括：接收模块1110，转换模块1120，拟合模块1130以及间隙计算模块1140。
其中，接收模块1110用于接收来自线激光器的三维图像。该三维图像中包含：至少一部分的第一部件的边界，至少一部分的第二部件的边界以及至少一个位于第一部件和第二部件边界上的焊点。转换模块1120用于将三维图像转换为二维灰度图像。拟合模块1130用于在二维灰度图像中获取第一部件的边界以及与第二部件的边界。间隙计算模块1140用于确定N条位于第一部件边界和第二部件边界之间的垂线；并且计算N条垂线长度的平均值作为第一部件与第二部件之间的间隙，N为正整数。
在操作中，接收模块1110接收包含了两个部件以及覆盖在部件之间的焊点的三维图像并提供至转换模块1120。转换模块1120将三维图像转为二维灰度图像。拟合模块1130在转换模块1120生成的二维灰度图像中进行边缘提取，获得两个部件的边界。间隙计算模块1140通过多次计算两个部件边界之间的垂线长度，取平均值后获得两个部件之间的间隙。
本申请实施例的其中一个有利方面是：在检测部件之间的间隙时，采用多次检测求平均值的方式，能够获得更精确的间隙测量结果。而且，基于线激光器进行图像采集，可以实现连续采样的同时，有效的排除传统相机因光源遮挡等因素造成的一系列干扰。
根据本申请一些实施例，可选地，请参阅图13，拟合模块1130具体可以包括：第一拟合单元1131和第二拟合单元1132。
其中，第一取样单元1131用于在包含第一部件边界的边界区域中，设置20-50个等分边界区域的第一拟合单元，并且依次连接每个第一拟合单元与第一部件边界之间的交点，形成第一直线。第二拟合单元1132用于在包含第二部件边界的边界区域中，设置20-50个等分边界区域的第二拟合单元；依次连接每个第二拟合单元与第二部件边界之间的交点，形成第二直线。上述基于 微分思想的边缘提取方式能够方便的通过调整拟合单元的数量来改变拟合程度，生成满足使用需要的拟合直线，用以表示和限定两个部件的边界。
根据本申请一些实施例，可选地，在选择使用拟合单元生成拟合直线的基础上，间隙计算模块1140具体用于：确定N个第一拟合单元与第一部件边界之间的N个交点为垂线设置点，分别以每个所述垂线设置点为起点，生成于第二直线垂直的N条垂线或者确定N个第二拟合单元与第二部件边界之间的N个交点为垂线设置点，分别以每个所述垂线设置点为起点，生成于第一直线垂直的N条垂线。这样的垂线设置方式以生成拟合直线的拟合单元为基础，生成与拟合单元数量相同，均匀分布的多条的垂线，可以用以实现多次检测求平均值的间隙检测方法。
根据本申请一些实施例，可选地，请继续参阅图13，该机器视觉检测装置还包括：边缘提取模块1150以及焊点宽度计算模块1160。
其中，边缘提取模块1150用于在二维灰度图像中提取分别与焊点的两侧边界拟合的第三直线和第四直线。焊点宽度计算模块1160用于确定第三直线与焊点相切的第一切点以及第四直线与焊点相切的第二切点；计算第一切点和第二切点之间距离作为焊点的宽度。这样的技术方案在二维灰度图像进行部件间隙检测的基础上，还进一步的进行了焊点宽度的自动检测，有利于全面的评估预焊工序的质量。
根据本申请一些实施例，可选地，该边缘提取模块1150具体用于：在焊点一侧的边界区域分别设置30-50个等分焊点一侧的边界区域的第三拟合单元；依次连接每个第三拟合单元与焊点一侧边界之间的交点，形成第三直线，并且；在焊点另一侧的边界区域分别设置30-50个等分焊点另一侧的边界区域的第四拟合单元；依次连接每个第四拟合单元与焊点另一侧边界之间的交点，形成第四直线。这样的设计采用了与部件边缘提取相类似的方式来获得与焊点两侧边界拟合的第三直线和第四直线，能够帮助完成对焊点宽度的自动检测。
在一些实施例中，可选地，焊点宽度计算模块1160在确定第一切点和第二切点时，具体用于：将第三直线上最后一个第三拟合单元与焊点一侧边界之间的交点作为第一切点；并且将第四直线上最后一个第四拟合单元与焊点另一侧边界的交点作为第二切点。这样的设计以第三直线和第四直线的最后 一个取样单元所在的位置作为两个切点，可以简单快捷的确定两个切点的位置。
在另一些实施例中，可选地，焊点宽度计算模块1160在确定第一切点和第二切点时，具体用于：将第三直线与第一直线之间的交点作为第一切点；并且将第四直线与第一直线的之间交点作为第二切点。这样的设计利用在先检测获得的部件边界的拟合直线与焊点两侧边缘的拟合直线之间的交点来获得两个切点，同样也能够快速方便的确定两个切点的位置。
应当说明的是，在本申请实施例中按照所要执行的方法步骤对机器视觉检测装置的功能模块进行划分。在一些实施例中，可以根据实际情况的需要，将本申请实施例中的机器视觉检测装置中的一个或者多个功能模块(如接收模块，转换模块，拟合模块，间隙计算模块，边缘提取模块以及焊点宽度计算模块)拆分成更多的功能模块，以执行相对应的方法步骤。在另一些实施例中，还可以将本申请实施例的换电装置中的一个或者多个功能模块整合为更少的功能模块，以执行相对应的方法步骤。
根据本申请一些实施例，请参阅图14，图14为本申请实施例提供的电子设备的结构示意图。该电子设备可以是第一控制器，第二控制器或者其他任何合适类型的用以执行上述图像软件系统的电子计算平台，在此不对其具体实现进行限定。
该电子设备可以包括：处理器1310、通信接口1320、存储器1330以及通信总线1340。
其中，处理器1310、通信接口1320以及存储器1330通过通信总线1340完成相互间的通信。通信接口1320用于与其它设备的通信连接(例如图像采集设备)。处理器1310用于调用程序1350，以执行上述实施例中的机器视觉检测方法中一个或者多个方法步骤或者实现上述实施例中的机器视觉检测装置中的一个或者多个功能模块。具体地，程序1350可以包括程序代码或者计算机操作指令。
在本实施例中，根据所使用的硬件的类型，处理器1310可以是中央处理单元、其他通用处理器、数字信号处理器、专用集成电路、现成可编程门阵列或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。
存储器1330用于存放程序1350。存储器1330可能包含高速RAM存储器，也可能还包括非易失性存储器，例如至少一个磁盘存储器。
本申请实施例还提供了一种计算机可读存储介质。该计算机可读存储介质可以为非易失性的计算机可读存储介质。该计算机可读存储介质存储有计算机程序。
其中，计算机程序被处理器执行时，实现上述实施例中的机器视觉检测方法中一个或者多个方法步骤或者实现上述实施例中的机器视觉检测装置中的一个或者多个功能模块。完整的计算机程序产品体现在含有本申请实施例公开的计算机程序的一个或多个计算机可读存储介质上(包括但不限于，磁盘存储器、CD-ROM、光学存储器等)。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围，其均应涵盖在本申请的权利要求和说明书的范围当中。尤其是，只要不存在结构冲突，各个实施例中所提到的各项技术特征均可以任意方式组合起来。本申请并不局限于文中公开的特定实施例，而是包括落入权利要求的范围内的所有技术方案。

Claims (13)

  1. 一种机器视觉检测方法，其特征在于，包括：
    接收来自线激光器的三维图像，所述三维图像中包含：至少一部分的第一部件的边界，至少一部分的第二部件的边界以及至少一个位于第一部件的边界和第二部件的边界上的焊点；
    将所述三维图像转换为二维灰度图像；
    在所述二维灰度图像中获取第一部件的边界和第二部件的边界；
    生成N条位于所述第一部件边界和第二部件边界之间的垂线；
    计算N条所述垂线长度的平均值作为所述第一部件和所述第二部件之间的间隙，N为正整数。
  2. 根据权利要求1所述的方法，其特征在于，所述在所述二维灰度图像中获取第一部件的边界和第二部件的边界，具体包括：
    在包含第一部件的边界的边界区域中，设置N个等分所述边界区域的第一拟合单元；
    依次连接每个所述第一拟合单元与所述第一部件的边界之间的交点，形成与所述第一部件边界拟合的第一直线；
    在包含第二部件的边界的边界区域中，设置N个等分所述边界区域的第二拟合单元；
    依次连接每个所述第二拟合单元与所述第二部件的边界之间的交点，形成与所述第二部件的边界拟合的第二直线；其中，N为20-50之间的正整数。
  3. 根据权利要求2所述的方法，其特征在于，所述生成N条位于所述第一部件的边界和第二部件的边界之间的垂线，具体包括：
    确定N个所述第一拟合单元与所述第一部件的边界之间的交点为垂线设置点；
    分别以每个所述垂线设置点为起点，生成与所述第二直线垂直的N条垂线；或者
    确定N个所述第二拟合单元与所述第二部件的边界之间的交点为垂线设 置点；
    分别以每个所述垂线设置点为起点，生成与所述第一直线垂直的N条垂线。
  4. 根据权利要求2所述的方法，其特征在于，所述方法还包括：
    在所述二维灰度图像中提取分别与所述焊点的两侧边界拟合的第三直线和第四直线；
    确定所述第三直线与所述焊点相切的第一切点以及所述第四直线与所述焊点相切的第二切点；
    计算所述第一切点和所述第二切点之间距离作为所述焊点的宽度。
  5. 根据权利要求4所述的方法，其特征在于，所述在所述二维灰度图像中提取分别与所述焊点的两侧边界拟合的第三直线和第四直线，具体包括：
    在包含所述焊点的第一侧边界的边界区域中，分别设置M个等分所述边界区域的第三拟合单元；
    依次连接每个所述第三拟合单元与所述焊点的第一侧边界之间的交点，形成所述第三直线，并且；
    在包含所述焊点的第二侧边界的边界区域中，分别设置M个等分所述边界区域的第四拟合单元；
    依次连接每个所述第四拟合单元与所述焊点的第二侧边界之间的交点，形成所述第四直线；其中，M为30-50之间的正整数。
  6. 根据权利要求5所述的方法，其特征在于，所述确定所述第三直线与所述焊点相切的第一切点以及所述第四直线与所述焊点相切的第二切点，具体包括：
    将所述第三直线上最后一个第三拟合单元与所述焊点的第一侧边界之间的交点作为所述第一切点；并且
    将所述第四直线上最后一个第四拟合单元与所述焊点的第二侧边界的交点作为所述第二切点。
  7. 根据权利要求5所述的方法，其特征在于，确定所述第三直线与所述焊点相切的第一切点以及所述第四直线与所述焊点相切的第二切点，具体包括：
    将所述第三直线与所述第一直线之间的交点作为所述第一切点；并且
    将所述第四直线与所述第一直线的之间交点作为所述第二切点。
  8. 一种机器视觉检测装置，其特征在于，包括：
    接收模块，用于接收来自线激光器的三维图像，所述三维图像中包含：至少一部分的第一部件的边界，至少一部分的第二部件的边界以及至少一个位于所述第一部件的边界和第二部件的边界上的焊点；
    转换模块，用于将所述三维图像转换为二维灰度图像；
    拟合模块，用于在所述二维灰度图像中获取所述第一部件的边界以及所述第二部件的边界；
    间隙计算模块，用于确定N条位于所述第一部件的边界和第二部件的边界之间的垂线；并且计算N条所述垂线长度的平均值作为所述第一部件与所述第二部件之间的间隙，N为正整数。
  9. 一种电子设备，其特征在于，包括处理器以及与所述处理器通信连接的处理器；所述存储器存储有计算机程序指令，所述计算机程序指令在被所述处理器调用时，以使所述处理器执行如权利要求1-7任一项所述的视觉检测方法。
  10. 一种非易失性计算机存储介质，其特征在于，所述非易失性计算机存储介质存储有计算机程序指令，以使所述计算机程序指令被处理器调用时，执行如权利要求1-7任一项所述的视觉检测方法。
  11. 一种视觉检测系统，其特征在于，包括：
    图像采集设备，所述图像采集设备包括若干线激光器，用于采集三维图像；
    驱动机构，用于使所述图像采集设备与待测部件之间发生相对移动；
    与所述图像采集设备通信连接的第一控制器，所述第一控制器用于处理所述三维图像，以使所述三维图像的处理结果用于所述待测部件的检测。
  12. 根据权利要求11所述的视觉检测系统，其特征在于，所述图像采集设备包括：两个线激光器、传感器支架以及遮光罩；
    两个所述线激光器分别设置在所述传感器支架的两侧；所述遮光罩固定在所述传感器支架上，罩套在所述线激光器外；
    所述传感器支架包括：高度调节模组和间距调节模组；
    所述高度调节模组用于调节所述线激光器所在的高度；所述间距调节模组用于调节两个所述线激光器之间的间距。
  13. 根据权利要求11所述的视觉检测系统，其特征在于，还包括：
    第二控制器，所述第二控制器用于控制所述高度调节模组和所述间距调节模组，以使两个所述线激光器达到目标间距和/或目标高度；
    所述第二控制器存储有若干个记录所述目标间距和目标高度的配置信息；每个配置信息与至少一种待测部件对应。
PCT/CN2021/142250 2021-12-29 2021-12-29 机器视觉检测方法、其检测装置及其检测系统 WO2023123003A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
EP21969381.9A EP4286789A1 (en) 2021-12-29 2021-12-29 Machine vision detection method, detection apparatus and detection system thereof
CN202180097091.3A CN117203486A (zh) 2021-12-29 2021-12-29 机器视觉检测方法、其检测装置及其检测系统
JP2023553313A JP2024508331A (ja) 2021-12-29 2021-12-29 マシンビジョンによる検出方法、その検出装置及びその検出システム
PCT/CN2021/142250 WO2023123003A1 (zh) 2021-12-29 2021-12-29 机器视觉检测方法、其检测装置及其检测系统
KR1020237029433A KR20230134597A (ko) 2021-12-29 2021-12-29 머신 비전 검출 방법, 이의 검출 장치 및 검출 시스템
US18/524,990 US20240095949A1 (en) 2021-12-29 2023-11-30 Machine vision detection method, detection apparatus and detection system thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/142250 WO2023123003A1 (zh) 2021-12-29 2021-12-29 机器视觉检测方法、其检测装置及其检测系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/524,990 Continuation US20240095949A1 (en) 2021-12-29 2023-11-30 Machine vision detection method, detection apparatus and detection system thereof

Publications (1)

Publication Number Publication Date
WO2023123003A1 true WO2023123003A1 (zh) 2023-07-06

Family

ID=86996914

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/142250 WO2023123003A1 (zh) 2021-12-29 2021-12-29 机器视觉检测方法、其检测装置及其检测系统

Country Status (6)

Country Link
US (1) US20240095949A1 (zh)
EP (1) EP4286789A1 (zh)
JP (1) JP2024508331A (zh)
KR (1) KR20230134597A (zh)
CN (1) CN117203486A (zh)
WO (1) WO2023123003A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117094964A (zh) * 2023-08-17 2023-11-21 正泰集团研发中心(上海)有限公司 电池片间距检测方法、装置、计算机设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2495525A1 (de) * 2011-03-01 2012-09-05 SmartRay GmbH Optisches Prüfverfahren mittels Intensitätsverlauf
DE102014016087A1 (de) * 2014-11-03 2014-12-24 In-Situ Gmbh Dreidimensionale optische Erfassung von Objektoberflächen
CN110530278A (zh) * 2019-10-09 2019-12-03 易思维（杭州）科技有限公司 利用多线结构光测量间隙面差的方法
CN111630342A (zh) * 2018-08-29 2020-09-04 深圳配天智能技术研究院有限公司 视觉焊接系统的缝隙检测方法以及系统
CN111901499A (zh) * 2020-07-17 2020-11-06 青岛聚好联科技有限公司 一种计算视频图像中像素实际距离的方法及设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2495525A1 (de) * 2011-03-01 2012-09-05 SmartRay GmbH Optisches Prüfverfahren mittels Intensitätsverlauf
DE102014016087A1 (de) * 2014-11-03 2014-12-24 In-Situ Gmbh Dreidimensionale optische Erfassung von Objektoberflächen
CN111630342A (zh) * 2018-08-29 2020-09-04 深圳配天智能技术研究院有限公司 视觉焊接系统的缝隙检测方法以及系统
CN110530278A (zh) * 2019-10-09 2019-12-03 易思维（杭州）科技有限公司 利用多线结构光测量间隙面差的方法
CN111901499A (zh) * 2020-07-17 2020-11-06 青岛聚好联科技有限公司 一种计算视频图像中像素实际距离的方法及设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117094964A (zh) * 2023-08-17 2023-11-21 正泰集团研发中心(上海)有限公司 电池片间距检测方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
US20240095949A1 (en) 2024-03-21
KR20230134597A (ko) 2023-09-21
CN117203486A (zh) 2023-12-08
JP2024508331A (ja) 2024-02-26
EP4286789A1 (en) 2023-12-06

Similar Documents

Publication Publication Date Title
CN108240793A (zh) 物体尺寸测量方法、装置和系统
CN107121093A (zh) 一种基于主动视觉的齿轮测量装置及测量方法
CN107907063B (zh) 一种基于视觉测量的钢带冲孔加工检测系统及方法
KR102056076B1 (ko) 용접비드 비전 검사 장치 및 용접 불량 검사방법
CN102735138B (zh) 测量结构以及用于至少确定压接触头的导线压接体的压接高度的方法
CN107869954B (zh) 一种双目视觉体积重量测量系统及其实现方法
CN107703513B (zh) 一种基于图像处理的非接触式接触网相对位置检测方法
US20240095949A1 (en) Machine vision detection method, detection apparatus and detection system thereof
JP5971028B2 (ja) 回転角度計測装置及び方法
CN102889864A (zh) 带卷边部物体塔形检测系统及其检测方法
CN110570412A (zh) 一种零件误差视觉判断系统
CN110470247B (zh) 一种零件内外圆面同轴度的检测装置及检测方法
CN111397529A (zh) 一种基于双目视觉结构光的复杂表面形状检测方法
CN109829897B (zh) 一种齿轮毛刺检测方法及齿轮高精度视觉测量系统
CN209640238U (zh) 用于弧面外观检测的机器视觉系统
CN107462187B (zh) 陶瓷插芯同轴度检测时光斑圆心确定方法及装置
JP4549931B2 (ja) ミキシングベーン検査方法と検査装置
WO2023097491A1 (zh) 机器视觉检测方法、其检测装置及其检测系统
CN111435075A (zh) 一种计算机视觉测量系统
CN211697564U (zh) 滤光片检测设备
CN114322780A (zh) 在线监测贴装重复精度方法
CN211042118U (zh) 一种三维检测系统
CN116907360B (zh) 基于机器视觉的ro膜胶线宽度检测系统及检测方法
JP5912666B2 (ja) 計測装置及びその処理方法
CN110907470A (zh) 滤光片检测设备及滤光片检测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21969381

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20237029433

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 21969381.9

Country of ref document: EP

Ref document number: 1020237029433

Country of ref document: KR

Ref document number: 2021969381

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2023553313

Country of ref document: JP

ENP Entry into the national phase

Ref document number: 2021969381

Country of ref document: EP

Effective date: 20230829

WWE Wipo information: entry into national phase

Ref document number: 202180097091.3

Country of ref document: CN