A visual system and related method for an automobile processing workstation

Abstract

The present application provides a visual system and related methods for an automobile processing workstation, the methods comprising: intercepting a corresponding gluing area image between two gluing positions adjacent to a target gluing position through a first image; obtaining first attribute information of the target gluing position; determining a first high-frequency feature extraction algorithm and a first algorithm control parameter corresponding to the first attribute information, and determining a preset condition corresponding to the first attribute information; extracting a high-frequency component image of the gluing area image; determining a first energy ratio between the high-frequency component image and the gluing area image; when the first energy ratio is within a preset range, extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set; when the high-frequency feature set does not meet the preset conditions, controlling the gluing gun through a mechanical arm to perform a gluing operation on the target gluing position. The present application can achieve precise gluing to ensure the gluing effect.

Description

Technical Field

[0001] The present application relates to the field of automotive technology, and in particular to a vision system and related methods for an automobile processing workstation. Background Art

[0002] Automotive glass is not only the main design element of the car, but also an important safety component of the car. It needs to have higher shockproof ability and strength. Among them, the gluing process directly affects the shockproof and waterproof capabilities of the car. Therefore, how to achieve accurate glue filling to ensure the gluing effect needs to be solved urgently. Summary of the invention

[0003] The embodiment of the present application provides a visual system and related methods for an automobile processing workstation, which can achieve precise glue filling to ensure the gluing effect.

[0004] In a first aspect, an embodiment of the present application provides a visual method for an automobile processing workstation, which is applied to a gluing robot, wherein the gluing robot comprises: a robot body, a gluing gun, and a visual device, wherein the robot body is connected to the gluing gun via a mechanical arm, and the method comprises:

[0005] When the glue gun completes the glue coating work between the automobile frame and the glass, the glue coating track between the automobile frame and the glass is identified by the visual device, and the glue coating position points constituting the glue coating track are determined according to the glue coating track to obtain a plurality of glue coating position points, and the distance between adjacent glue coating position points is within a preset distance range;

[0006] Acquire a first image with the target gluing position point as the focus, and intercept a corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image;

[0007] Acquire first attribute information of the target gluing position point;

[0008] Determining a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determining a preset condition corresponding to the first attribute information;

[0009] Extracting a high-frequency component image of the glue-coated area image;

[0010] determining a first energy ratio between the high frequency component image and the glued area image;

[0011] When the first energy ratio is within a preset range, extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set;

[0012] Detecting whether the high-frequency feature set meets a preset condition;

[0013] When the high-frequency feature set meets the preset condition, determining that the gluing area image corresponding to the target gluing position point meets the gluing requirement;

[0014] When the high-frequency feature set does not meet the preset condition, the gluing gun is controlled by the robotic arm to perform a gluing operation on the target gluing position point.

[0015] In a second aspect, an embodiment of the present application provides a visual system of an automobile processing workstation, which is applied to a gluing robot, wherein the gluing robot comprises: a robot body, a gluing gun, and a visual device, wherein the robot body is connected to the gluing gun via a mechanical arm, and the system comprises: a visual recognition unit, an acquisition unit, a determination unit, an extraction unit, a detection unit, and a control unit, wherein:

[0016] The visual recognition unit is used to identify the gluing track between the automobile frame and the glass through the visual device when the gluing gun completes the gluing work between the automobile frame and the glass, determine the gluing position points constituting the gluing track according to the gluing track, and obtain a plurality of gluing position points, wherein the distance between adjacent gluing position points is within a preset distance range; acquire a first image with the target gluing position point as the focus, and intercept the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image;

[0017] The acquisition unit is used to acquire the first attribute information of the target gluing position point;

[0018] The determining unit is used to determine a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine a preset condition corresponding to the first attribute information;

[0019] The extraction unit is used to extract the high-frequency component image of the glue-coated area image;

[0020] The determining unit is further used to determine a first energy ratio between the high-frequency component image and the glue-coated area image;

[0021] The extraction unit is further configured to extract features of the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set when the first energy ratio is within a preset range;

[0022] The detection unit is used to detect whether the high-frequency feature set meets a preset condition;

[0023] The determination unit is further configured to determine that the gluing area image corresponding to the target gluing position point meets the gluing requirement when the high-frequency feature set meets the preset condition;

[0024] The control unit is used to control the glue gun to perform glue filling operation on the target glue coating position point through the robotic arm when the high-frequency feature set does not meet the preset condition.

[0025] In the third aspect, an embodiment of the present application provides a glue-spreading robot, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0026] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in the first aspect of the embodiment of the present application.

[0027] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0028] The implementation of the embodiments of the present application has the following beneficial effects:

[0029] It can be seen that the vision system and related methods of the automobile processing workstation described in the embodiments of the present application are applied to a gluing robot, which includes: a robot body, a gluing gun, and a vision device. The robot body is connected to the gluing gun through a mechanical arm. When the gluing gun completes the gluing work between the automobile frame and the glass, the gluing trajectory between the automobile frame and the glass is identified through the vision device, and the gluing position points constituting the gluing trajectory are determined according to the gluing trajectory to obtain a plurality of gluing position points, wherein the distance between adjacent gluing position points is within a preset distance range; a first image is acquired with the target gluing position point as the focus, and a corresponding gluing area image between two adjacent gluing position points of the target gluing position point is intercepted through the first image. image; obtain first attribute information of the target gluing position point; determine the first high-frequency feature extraction algorithm corresponding to the first attribute information and the first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine the preset condition corresponding to the first attribute information; extract the high-frequency component image of the gluing area image; determine the first energy ratio between the high-frequency component image and the gluing area image; when the first energy ratio is within the preset range, perform feature extraction on the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set; detect whether the high-frequency feature set meets the preset condition; when the high-frequency feature set meets the preset condition, determine that the gluing area image corresponding to the target gluing position point meets the gluing requirements;When the high-frequency feature set does not meet the preset conditions, the gluing gun is controlled by the robotic arm to perform a gluing operation on the target gluing position point. First, when the gluing gun completes the gluing work between the car frame and the glass, the glued area and the un-glued area are obvious. Then, the gluing trajectory between the car frame and the glass can be identified by a visual device, and the gluing position points constituting the gluing trajectory are determined according to the gluing trajectory to obtain a plurality of gluing position points. The first image is acquired with the target gluing position point as the focus, thereby ensuring the image quality of the gluing area image. The corresponding gluing area image between two adjacent gluing position points of the target gluing position point is intercepted through the first image, and the image of a section of the gluing area image where there is colloid between the two gluing position points is captured. In this way, the gluing trajectory can be divided into multiple sections, and the gluing inspection can be performed in sections, thereby preventing the phenomenon of missing glue, wrong gluing, uneven gluing, etc. Secondly, due to different attribute information, the requirements corresponding to the gluing detail features that need to be checked are different. The preset conditions reflect the requirements corresponding to the gluing detail features that need to be checked to a certain extent, not only based on the gluing The gluing process, gluing method and corresponding gluing effect determine the corresponding requirements of the gluing detail features to be checked, as well as the corresponding high-frequency feature extraction algorithm and algorithm control parameters. Thirdly, when the first energy ratio is within the preset range, it means that the high-frequency details are appropriate. Then, the high-frequency component image can be feature extracted according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set. When the high-frequency feature set meets the preset conditions, it means that the gluing area image corresponding to the target gluing position point meets the gluing requirements. Thus, the image recognition technology can be used to deeply inspect the gluing effect, which helps to improve the gluing inspection efficiency. When the high-frequency feature set does not meet the preset conditions, that is, the gluing area image corresponding to the target gluing position point does not meet the gluing requirements, it means that the gluing is qualified. Then, the gluing gun can be controlled by the mechanical arm to perform a gluing operation on the target gluing position point. In this way, not only can the image recognition technology be used to deeply inspect the gluing effect, which helps to improve the gluing inspection efficiency, but also the gluing can be further performed, thereby deeply ensuring the gluing effect and achieving accurate gluing to ensure the gluing effect. ; BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 It is a flowchart of a visual method for an automobile processing workstation provided in an embodiment of the present application;

[0032] Figure 2 It is a structural schematic diagram of a glue coating robot provided in an embodiment of the present application;

[0033] Figure 3 This is a block diagram of the functional units of a visual system of an automobile processing workstation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but in a possible example also includes steps or units that are not listed, or in a possible example also includes other steps or units inherent to these processes, methods, products or devices.

[0035] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0036] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] In the embodiment of the present application, the gluing robot can be understood as an industrial robot or an automobile processing robot or a humanoid robot with a gluing function.

[0038] See also Figure 1 , Figure 1 The flowchart of a visual method of an automobile processing workstation provided in an embodiment of the present application is applied to a gluing robot. The gluing robot includes: a robot body, a gluing gun, and a visual device. The robot body is connected to the gluing gun through a mechanical arm. The visual method of the automobile processing workstation includes:

[0039] 101. When the gluing gun completes the gluing work between the automobile frame and the glass, the gluing track between the automobile frame and the glass is identified by the visual device, and the gluing position points constituting the gluing track are determined according to the gluing track to obtain a plurality of gluing position points, and the distance between adjacent gluing position points is within a preset distance range.

[0040] Among them, the gluing robot can be applied to automobile processing workstations.

[0041] In the embodiments of the present application, the visual device may include at least one of the following: an infrared imager, a laser sensor, a radar sensor, a camera, an ultrasonic sensor, etc., which are not limited here.

[0042] In a specific implementation, the gluing robot may include: a robot body, a gluing gun, and a visual device. The robot body is connected to the gluing gun via a robotic arm to control the gluing gun to perform gluing operations.

[0043] Among them, the preset distance range can be pre-set or system default, the distance between adjacent gluing position points is within the preset distance range, the preset distance range can include a maximum value and a minimum value, the distance between adjacent gluing position points is between the maximum value and the minimum value, the distance between adjacent gluing position points is related to the curvature between the adjacent gluing position points, the greater the curvature, the shorter the distance between adjacent gluing position points, conversely, the smaller the curvature, the greater the distance between adjacent gluing position points.

[0044] The longer the gluing track is, the more gluing position points there are.

[0045] In a specific implementation, when the gluing gun completes the gluing work between the car frame and the glass, since the glued areas and the un-glueed areas are obvious, the gluing track between the car frame and the glass can be identified through a visual device, and the gluing position points constituting the gluing track can be determined according to the gluing track to obtain multiple gluing position points, and the distance between adjacent gluing position points is within a preset distance range.

[0046] In a specific implementation, each gluing position point may also be numbered.

[0047] In a specific implementation, a gluing effect image of a certain area (e.g., sunroof, windshield, front seat glass, rear seat glass, etc.) can be obtained through a visual device, the gluing area can be extracted, and then the contour of the gluing area can be extracted, and sampling is performed on the contour to obtain multiple sampling points, each sampling point corresponds to a gluing position point, and a gluing trajectory is generated based on the contour. The distances between adjacent gluing position points can be the same or different.

[0048] 102. Acquire a first image with the target gluing position point as the focus, and intercept a corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image.

[0049] In a specific implementation, a first image can be acquired with the target gluing position point as the focus, thereby ensuring the image quality of the gluing area image. The first image can be used to capture the corresponding gluing area image between two adjacent gluing positions of the target gluing position point, and the image of the area where the colloid exists between the two gluing positions of the gluing area image. In this way, the gluing trajectory can be divided into multiple segments, and the gluing inspection can be performed in segments, thereby preventing phenomena such as missing glue, incorrect gluing, and uneven gluing.

[0050] 103. Obtain first attribute information of the target gluing position point.

[0051] Among them, the first attribute information of the target gluing position point may include at least one of the following: the gluing position coordinates of the gluing position point, the gluing process of the gluing position point, the glass material of the gluing position point, the cleanliness of the glass at the gluing position point, the car frame material at the gluing position point, the glass thickness at the gluing position point, the glass curvature at the gluing position point, etc., which are not limited here.

[0052] 104. Determine a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine a preset condition corresponding to the first attribute information.

[0053] In the embodiment of the present application, the first attribute information reflects the gluing process, gluing method and corresponding gluing effect to a certain extent, and different first attribute information has different requirements corresponding to the gluing detail features that need to be checked.

[0054] In a specific implementation, the mapping relationship between the attribute information of the preset glue coating position point and the high-frequency feature extraction algorithm can be pre-stored, and then the first high-frequency feature extraction algorithm corresponding to the first attribute information can be determined based on the mapping relationship. The mapping relationship between the attribute information of the preset glue coating position point and the algorithm control parameters of the high-frequency feature extraction algorithm can also be pre-stored, and then the first algorithm control parameters corresponding to the first high-frequency feature extraction algorithm corresponding to the first attribute information can be determined based on the mapping relationship. The first algorithm control parameters are used to implement at least one of the following functions: feature extraction position, feature extraction speed, feature type, etc., which are not limited here.

[0055] The high-frequency feature extraction algorithm may include at least one of the following: a feature point extraction algorithm, a feature vector extraction algorithm, a feature texture extraction algorithm, etc., which are not limited here.

[0056] Among them, due to different attribute information, the requirements corresponding to the gluing detail features that need to be checked are different. The preset conditions reflect the requirements corresponding to the gluing detail features that need to be checked to a certain extent. Therefore, the mapping relationship between the attribute information of the preset gluing position point and the preset conditions can be pre-stored, and then the preset conditions corresponding to the first attribute information can be determined based on the mapping relationship.

[0057] In this way, not only can the corresponding requirements for the glue coating detail features that need to be checked, as well as the corresponding high-frequency feature extraction algorithm and algorithm control parameters be determined based on the glue coating process, glue coating method and the corresponding glue coating effect.

[0058] 105. Extract a high-frequency component image of the glue-coated area image.

[0059] In an embodiment of the present application, a multi-scale decomposition algorithm can be used to extract the high-frequency component image of the glued area image. At the same time, a low-frequency component image can also be obtained. The low-frequency component image reflects the main body of the image, and the high-frequency component image reflects the deep detail characteristics of the image.

[0060] The multi-scale decomposition algorithm may include at least one of the following: wavelet transform algorithm, contourlet transform algorithm, non-subsampled contourlet transform algorithm, pyramid transform algorithm, barren wave transform algorithm, etc., which are not limited here.

[0061] 106. Determine a first energy ratio between the high-frequency component image and the glue-coated area image.

[0062] In a specific implementation, the first energy ratio between the high-frequency component image and the glue-coated area image can be determined. Specifically, the first energy value of the high-frequency component image and the second energy value of the glue-coated area image can be obtained, and then the first energy ratio is calculated based on the two. Specifically, the first energy ratio = the first energy value of the high-frequency component image / the second energy value of the glue-coated area image.

[0063] 107. When the first energy ratio is within a preset range, extract features of the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set.

[0064] The preset range may be preset or set by system default.

[0065] In a specific implementation, when the first energy ratio is within a preset range, it indicates that the high-frequency details are appropriate, and the high-frequency component image can be feature extracted according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set.

[0066] Among them, the high-frequency feature set may include at least one of the following features: at least one feature point, at least one feature contour, at least one feature vector, at least one feature density distribution parameter, the number of feature points, the number of feature points of at least one layer, the number of feature contours, at least one layer of feature contours, at least one layer of feature vectors, at least one layer of feature density distribution parameters, etc., without limitation here.

[0067] Since the first attribute information reflects the gluing process, gluing method and the corresponding gluing effect to a certain extent, different first attribute information requires different corresponding requirements for the gluing detail features to be checked. Therefore, the corresponding features can be extracted based on the high-frequency feature extraction algorithm and algorithm control parameters related to the gluing process, gluing method and the corresponding gluing effect.

[0068] 108. Detect whether the high-frequency feature set meets a preset condition.

[0069] The preset conditions may be preset or set by system default. For example, the preset conditions may include at least one of the following: the number of feature points meets a certain condition, the feature vector meets a certain condition, the feature point distribution density meets a certain condition, the feature profile meets a certain condition, etc., which are not limited here.

[0070] 109. When the high-frequency feature set meets the preset condition, determine whether the gluing area image corresponding to the target gluing position point meets the gluing requirement.

[0071] In the embodiment of the present application, when the high-frequency feature set meets the preset conditions, it means that the image of the glued area corresponding to the target glued position point meets the glued requirements. Therefore, image recognition technology can be used to deeply inspect the glue effect, which helps to improve the efficiency of glue inspection.

[0072] Next, the next gluing position point after the target gluing position point can be used as the focus, and step 102 can be repeated, that is, the next gluing position point after the target gluing position point is used as the new target gluing position point. In this way, the gluing verification can be implemented in sections, thereby preventing gluing omissions, uneven gluing, incorrect gluing, etc.

[0073] 110. When the high-frequency feature set does not satisfy the preset condition, the gluing gun is controlled by the robotic arm to perform a gluing operation on the target gluing position point.

[0074] In an embodiment of the present application, when the high-frequency feature set does not meet the preset conditions, that is, the glue coating area image corresponding to the target glue coating position point does not meet the glue coating requirements, it means that the glue coating is unqualified. Then, the gluing gun can be controlled by a robotic arm to perform a glue filling operation on the target glue coating position point. Specifically, the glue filling operation can be performed on the area corresponding to the glue coating area image. In this way, not only can the image recognition technology be used to deeply inspect the gluing effect, which helps to improve the efficiency of glue coating inspection, but it can also further perform glue filling, thereby deeply ensuring the gluing effect.

[0075] Optionally, the high-frequency component image includes multiple layers of first high-frequency component images; the preset condition includes multiple first preset conditions, and each layer of high-frequency component image corresponds to one first preset condition; the above step 108, detecting whether the high-frequency feature set meets the preset condition, may include the following steps:

[0076] Determine a first high-frequency feature set of each first high-frequency component image in the multiple layers of first high-frequency component images according to the high-frequency feature set, to obtain a plurality of first high-frequency feature sets;

[0077] Detect whether each of the multiple first high-frequency feature sets meets the corresponding first preset condition among the multiple first preset conditions. If so, confirm that the high-frequency feature set meets the preset condition. Otherwise, confirm that the high-frequency feature set does not meet the preset condition.

[0078] In a specific implementation, the high-frequency component image may include multiple layers of first high-frequency component images; the preset conditions include multiple first preset conditions, each layer of high-frequency component image corresponds to a first preset condition, and different high-frequency component images correspond to different image depths (feature detail depths).

[0079] In an embodiment of the present application, the first high-frequency feature set of each first high-frequency component image in the multi-layer first high-frequency component image can be determined according to the high-frequency feature set to obtain multiple first high-frequency feature sets. Then, each first high-frequency feature set in the multiple first high-frequency feature sets is detected to see whether it satisfies the corresponding first preset condition in the multiple first preset conditions. If so, it is confirmed that the high-frequency feature set satisfies the preset condition. Otherwise, it is confirmed that the high-frequency feature set does not satisfy the preset condition. In this way, different first preset conditions reflect different gluing process requirements, gluing method requirements and corresponding gluing effect requirements. Furthermore, corresponding depth feature verification requirements can be configured based on the gluing process, gluing method and corresponding gluing effect, thereby achieving depth inspection of the gluing effect, which helps to improve the efficiency of gluing inspection.

[0080] Optionally, the preset condition includes a reference high-frequency feature set, and the above step 110, controlling the glue gun to perform a glue filling operation on the target glue coating position point by the robotic arm, may include the following steps:

[0081] Determining a first degree of deviation between the high-frequency feature set and the reference high-frequency feature set;

[0082] Obtaining reference gluing parameters corresponding to the target gluing position point;

[0083] determining a first adjustment parameter corresponding to the first deviation;

[0084] Determine a first gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0085] Determining a second gluing parameter according to the gluing area image;

[0086] Determine a third gluing parameter according to the first gluing parameter and the second gluing parameter;

[0087] Determining a first control parameter of the robotic arm according to the third gluing parameter;

[0088] The robot arm operates according to the first control parameter, and the glue gun is controlled to operate according to the third glue coating parameter, so as to perform glue filling operation on the target glue coating position point.

[0089] Among them, the reference high-frequency feature set reflects the requirements of preset conditions, and the reference high-frequency feature set may include at least one of the following features: at least one feature point, at least one feature contour, at least one feature vector, at least one feature density distribution parameter, the number of feature points, the number of feature points of at least one layer, the number of feature contours, at least one layer of feature contours, at least one layer of feature vectors, at least one layer of feature density distribution parameters, etc., which are not limited here.

[0090] Among them, the reference gluing parameters may include at least one of the following: gluing nozzle opening size, gluing pressure, gluing position, gluing speed, gluing angle, gluing height, gluing gun nozzle opening posture, gluing color, gluing temperature, glue amount, etc., which are not limited here.

[0091] Among them, the second gluing parameter may include at least one of the following: gluing nozzle opening size, gluing pressure, gluing position, gluing speed, gluing angle, gluing height, gluing gun nozzle opening posture, gluing color, gluing temperature, glue amount, etc., which are not limited here.

[0092] Among them, the first control parameter of the robotic arm may include at least one of the following: the movement of the robotic arm, the angle of the robotic arm, the strength of the robotic arm, the posture of the robotic arm, the moving speed of the robotic arm, etc., which are not limited here.

[0093] In a specific implementation, a first deviation between the high-frequency feature set and the reference high-frequency feature set may be determined, where the first deviation=(high-frequency feature set−reference high-frequency feature set) / reference high-frequency feature set.

[0094] In the specific implementation, during the gluing process, the glue gun can bind the glue coating position and the glue coating parameters, and then, the reference glue coating parameters corresponding to the target glue coating position point can be obtained. The mapping relationship between the preset deviation and the adjustment parameter can also be pre-stored, and then, the first adjustment parameter corresponding to the first deviation can be determined based on the mapping relationship. The value range of the first adjustment parameter is 0~0.2. Different deviations reflect different degrees of glue deficiency. Therefore, the corresponding adjustment parameters can be adapted based on the actual degree of glue deficiency, and then the first glue coating parameters can be determined based on the first adjustment parameters and the reference glue coating parameters. The first glue coating parameter = first adjustment parameter * reference glue coating parameter. The reference glue coating parameter can continue the previous glue coating effect to a certain extent. Then, the corresponding adjustment parameters can be adapted based on the actual degree of glue deficiency to weaken the previous glue coating effect, so that the glue repair effect is natural and integrated with the surrounding glue coating track to achieve a smooth glue repair effect.

[0095] The glue coating area image reflects the glue coating area range to a certain extent, and then the second glue coating parameter can be determined according to the glue coating area image. The second glue coating parameter can mainly determine the glue coating area range.

[0096] Next, the third gluing parameter may be determined according to the first gluing parameter and the second gluing parameter. Specifically, the combination of the first gluing parameter and the second gluing parameter may be determined as the third gluing parameter.

[0097] In the specific implementation, since the third gluing parameter determines the entire process of gluing to a certain extent, and the gluing process also determines the movement of the robotic arm, the mapping relationship between the preset gluing parameters and the control parameters of the robotic arm can be pre-stored. Then, the first control parameter of the robotic arm corresponding to the third gluing parameter can be determined based on the mapping relationship, and then the robotic arm can work with the first control parameter, and the gluing gun can be controlled to work with the third gluing parameter to perform glue repair operations on the target gluing position points. Thereby, the probability of glue defects can be reduced, targeted gluing can be achieved to ensure glue repair and gluing effects, and the intelligence of the gluing robot can be improved.

[0098] Optionally, the above step of determining the first gluing parameter according to the first adjustment parameter and the reference gluing parameter may include the following steps:

[0099] Determine a fourth gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0100] Dividing the glued area image into a plurality of areas, each area being equal in area;

[0101] Determining a first signal-to-noise ratio of the high-frequency component image;

[0102] Determine a regional high-frequency component image corresponding to each of the multiple regions according to the high-frequency component image, to obtain multiple regional high-frequency component images;

[0103] Determine the number of feature points and the average energy value of the feature points in each region of the plurality of regional high-frequency component images to obtain a plurality of feature point numbers and a plurality of average energy values;

[0104] Determining a first standard deviation corresponding to the number of the plurality of feature points;

[0105] Determining a second standard deviation corresponding to the plurality of average energy values;

[0106] determining a weight pair corresponding to the first signal-to-noise ratio;

[0107] Determine a target standard deviation according to the first standard deviation, the second standard deviation and the weight pair;

[0108] determining a first fine-tuning parameter corresponding to the target standard deviation;

[0109] The fourth gluing parameter is fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter.

[0110] In the specific implementation, the fourth gluing parameter is determined according to the first adjustment parameter and the reference gluing parameter, and the fourth gluing parameter = the first adjustment parameter * the reference gluing parameter. The reference gluing parameter can continue the previous gluing effect to a certain extent. Then, based on the actual degree of glue deficiency, the corresponding adjustment parameter can be adapted to weaken the previous gluing effect, so that the glue repairing effect is natural and integrated with the surrounding glue track to achieve a smooth glue repairing effect.

[0111] Furthermore, the coated area image can be divided into multiple areas, each with an equal area, and the first signal-to-noise ratio of the high-frequency component image can be determined. The regional high-frequency component image corresponding to each of the multiple areas can also be determined based on the high-frequency component image to obtain multiple regional high-frequency component images. Then, the number of feature points and the average energy value of the feature points in each area of ​​the multiple regional high-frequency component images can be determined to obtain multiple numbers of feature points and multiple average energy values. The standard deviation operation is then performed on the multiple numbers of feature points to obtain a first standard deviation, which reflects the stability of the high-frequency component image in terms of the number of feature points. The standard deviation operation is performed on the multiple average energy values ​​to obtain a second standard deviation, which reflects the stability of the high-frequency component image in terms of the average energy.

[0112] In the specific implementation, due to the influence of the signal-to-noise ratio, the stability reflected by the two is different. Therefore, the mapping relationship between the preset signal-to-noise ratio and the weight pair can be pre-stored. The weight pair includes a first weight and a second weight, and the first weight + the second weight = 1, wherein the first weight is the weight corresponding to the first standard deviation, and the second weight is the weight corresponding to the second standard deviation. Then, based on the mapping relationship, the weight pair corresponding to the first signal-to-noise ratio can be determined, and then a weighted operation is performed according to the first standard deviation, the second standard deviation and the weight pair to obtain the target standard deviation. Then, the stability of the high-frequency image component can be deeply analyzed based on the actual signal-to-noise ratio.

[0113] Next, the mapping relationship between the preset standard deviation and the fine-tuning parameter can be pre-stored, and then the first fine-tuning parameter corresponding to the target standard deviation can be determined based on the mapping relationship, and then the fourth gluing parameter can be fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter, the first gluing parameter = (1 + first fine-tuning parameter) * fourth gluing parameter. Since stability determines the smoothness of the gluing effect to a certain extent, the stability of the high-frequency image component can be deeply analyzed based on the actual signal-to-noise ratio, and the gluing parameter can be deeply adjusted based on the stability, so that the final gluing effect can achieve a smooth effect, thereby achieving higher shockproof ability and strength, as well as helping to improve shockproof and waterproof capabilities, and can also improve the intelligence of the gluing robot.

[0114] Optionally, the preset range includes an upper threshold and a lower threshold, and may further include the following steps:

[0115] When the first energy ratio is less than the lower limit threshold, the step of acquiring the first image with the target gluing position point as the focus and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image is executed again.

[0116] In a specific implementation, when the first energy ratio is less than the lower limit threshold, it means that the details are not prominent, or the details are blocked by the main body of the image. Then, the step of acquiring the first image with the target glue coating position point as the focus is re-executed, and the step of capturing the corresponding glue coating area image between two glue coating position points adjacent to the target glue coating position point through the first image is performed. In this way, the accuracy of the depth inspection of the glue coating effect can be ensured to avoid misidentification of glue coating defects.

[0117] Optionally, the following steps may also be included:

[0118] When the first energy ratio is greater than the upper threshold, determining a second deviation between the first energy ratio and the upper threshold;

[0119] When the second deviation is greater than the preset deviation, re-performing the step of acquiring the first image with the target gluing position point as the focus, and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image;

[0120] When the second deviation is less than or equal to the preset deviation, determining a first image enhancement algorithm corresponding to the preset deviation, and determining a first control parameter of the first image enhancement algorithm corresponding to the second deviation;

[0121] Acquire a low-frequency component image corresponding to the high-frequency component image in the glue-coated area image;

[0122] Performing image enhancement processing on the low-frequency component image according to the first image enhancement algorithm and the first control parameter to obtain a first low-frequency component image;

[0123] Reconstructing the first low-frequency component image and the high-frequency component image to obtain an enhanced glue-coated area image;

[0124] Extract the high-frequency component image of the enhanced glued area image, and perform the step of extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set.

[0125] In a specific implementation, when the first energy ratio is greater than the upper threshold, it means that the image details are full, but the main body of the image is insufficient, which may affect the stability of the high-frequency part. Further, the second deviation between the first energy ratio and the upper threshold can be determined, and the second deviation = (first energy ratio-upper threshold) / upper threshold.

[0126] The preset deviation degree may be preset or set by system default.

[0127] Furthermore, when the second deviation is greater than the preset deviation, it means that the image details are somewhat too full, while the main body of the image is seriously insufficient, resulting in poor stability of the high-frequency part and cannot be compensated by image enhancement of the low-frequency component image. Therefore, the step of acquiring the first image with the target glue position point as the focus and intercepting the corresponding glue area image between the two glue position points adjacent to the target glue position point through the first image can be re-executed. In this way, the accuracy of the depth inspection of the glue effect can be guaranteed to avoid misidentification of glue defects.

[0128] In addition, when the second deviation is less than or equal to the preset deviation, it means that although the image details are somewhat full, the problem of insufficient image body can be compensated by image enhancement of the low-frequency component image. Therefore, the mapping relationship between the preset deviation and the image enhancement algorithm can be pre-stored, and then, the first image enhancement algorithm corresponding to the preset deviation can be determined based on the mapping relationship. The mapping relationship between the preset deviation and the control parameter of the first image enhancement algorithm can also be pre-stored, and then, the first control parameter of the first image enhancement algorithm corresponding to the second deviation can be determined based on the mapping relationship. In this way, the corresponding first image enhancement algorithm can be matched based on the preset deviation (that is, the controllable deviation), and the first control parameter corresponding to the first image enhancement algorithm can be matched based on the second deviation (that is, the actual deviation), thereby achieving accurate image enhancement of the low-frequency component image.

[0129] Next, a low-frequency component image corresponding to the high-frequency component image in the glued area image can be obtained, and then the low-frequency component image is enhanced according to the first image enhancement algorithm and the first control parameter to obtain a first low-frequency component image, that is, the corresponding first image enhancement algorithm is matched based on a preset deviation (i.e., a controllable deviation), and the first control parameter corresponding to the first image enhancement algorithm is matched based on the second deviation (i.e., the actual deviation), thereby achieving accurate image enhancement of the low-frequency component image. Next, the image is reconstructed according to the first low-frequency component image and the high-frequency component image to obtain an enhanced glued area image. The reconstruction method can correspond to the corresponding multi-scale decomposition algorithm of 105 to extract the high-frequency component image of the enhanced glued area image. The enhanced glued area image is equivalent to ensuring The stability of the high-frequency part, the high-frequency component image of the enhanced glued area image is extracted and the image quality of the high-frequency component image is correspondingly changed. Since the image body of the glued area image is enhanced, the first energy ratio between the high-frequency component image and the glued area image is correspondingly reduced, so that the high-frequency details are moderate, and then the step of extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set can ensure the accuracy of the deep inspection of the glue effect and avoid the misidentification of glue defects. Furthermore, the image recognition technology can be used to deeply inspect the glue effect, which helps to improve the efficiency of glue inspection and can further perform glue filling, thereby deeply ensuring the glue effect and achieving accurate glue filling to ensure the glue effect.

[0130] It can be seen that the visual method of the automobile processing workstation described in the embodiment of the present application, firstly, when the gluing gun completes the gluing work between the automobile frame and the glass, since the glued area and the un-glueed area are obvious, the gluing track between the automobile frame and the glass can be identified by a visual device, and the gluing position points constituting the gluing track are determined according to the gluing track to obtain a plurality of gluing position points, and a first image is acquired with the target gluing position point as the focus, thereby ensuring the image quality of the gluing area image, and the corresponding gluing area image between two gluing position points adjacent to the target gluing position point is intercepted by the first image, and an image of a region where colloid exists between the two gluing position points of the gluing area image is obtained. In this way, the gluing track can be divided into multiple segments, and the gluing inspection can be performed in segments, thereby preventing the phenomenon of missing glue, wrong gluing, uneven gluing, etc. Secondly, due to different attribute information, the requirements corresponding to the gluing detail features that need to be checked are different. The preset conditions reflect the requirements corresponding to the gluing detail features that need to be checked to a certain extent, which can not only be based on the gluing process, coating The gluing method and the corresponding gluing effect determine the corresponding requirements of the corresponding gluing detail features that need to be checked, as well as the corresponding high-frequency feature extraction algorithm and algorithm control parameters. Thirdly, when the first energy ratio is within the preset range, it means that the high-frequency details are appropriate, and the high-frequency component image can be feature extracted according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set. When the high-frequency feature set meets the preset conditions, it means that the image of the glued area corresponding to the target gluing position point meets the gluing requirements. Therefore, the image recognition technology can be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection. When the high-frequency feature set does not meet the preset conditions, that is, the image of the glued area corresponding to the target gluing position point does not meet the gluing requirements, it means that the gluing is qualified. Then, the gluing gun can be controlled by the robotic arm to perform gluing operations on the target gluing position point. In this way, not only can the image recognition technology be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection, but also glue can be further repaired, thereby deeply ensuring the gluing effect and achieving precise glue repair to ensure the gluing effect.

[0131] In accordance with the above embodiment, please refer to Figure 2 , Figure 2 : is a structural schematic diagram of a gluing robot provided in an embodiment of the present application, the gluing robot includes a processor, a memory, a communication interface and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor. In an embodiment of the present application, the gluing robot is applied to the gluing robot, the gluing robot includes: a robot body, a gluing gun, and a visual device, the robot body is connected to the gluing gun through a mechanical arm, and the program includes instructions for executing the following steps:

[0132] When the glue gun completes the glue coating work between the automobile frame and the glass, the glue coating track between the automobile frame and the glass is identified by the visual device, and the glue coating position points constituting the glue coating track are determined according to the glue coating track to obtain a plurality of glue coating position points, and the distance between adjacent glue coating position points is within a preset distance range;

[0133] Acquire a first image with the target gluing position point as the focus, and intercept a corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image;

[0134] Acquire first attribute information of the target gluing position point;

[0135] Determining a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determining a preset condition corresponding to the first attribute information;

[0136] Extracting a high-frequency component image of the glue-coated area image;

[0137] determining a first energy ratio between the high-frequency component image and the glue-coated area image;

[0138] When the first energy ratio is within a preset range, extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set;

[0139] Detecting whether the high-frequency feature set meets a preset condition;

[0140] When the high-frequency feature set meets the preset condition, determining that the gluing area image corresponding to the target gluing position point meets the gluing requirement;

[0141] When the high-frequency feature set does not meet the preset condition, the gluing gun is controlled by the robotic arm to perform a gluing operation on the target gluing position point.

[0142] Optionally, the high-frequency component image includes multiple layers of first high-frequency component images; the preset condition includes multiple first preset conditions, and each layer of high-frequency component image corresponds to one first preset condition;

[0143] In terms of detecting whether the high-frequency feature set meets a preset condition, the program includes instructions for executing the following steps:

[0144] Determine a first high-frequency feature set of each first high-frequency component image in the multiple layers of first high-frequency component images according to the high-frequency feature set, to obtain a plurality of first high-frequency feature sets;

[0145] Detect whether each of the multiple first high-frequency feature sets meets the corresponding first preset condition among the multiple first preset conditions. If so, confirm that the high-frequency feature set meets the preset condition. Otherwise, confirm that the high-frequency feature set does not meet the preset condition.

[0146] Optionally, the preset condition includes a reference high-frequency feature set, and in the aspect of controlling the gluing gun by the robotic arm to perform a gluing operation on the target gluing position point, the program includes instructions for executing the following steps:

[0147] Determining a first degree of deviation between the high-frequency feature set and the reference high-frequency feature set;

[0148] Obtaining reference gluing parameters corresponding to the target gluing position point;

[0149] determining a first adjustment parameter corresponding to the first deviation;

[0150] Determine a first gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0151] Determining a second gluing parameter according to the gluing area image;

[0152] Determine a third gluing parameter according to the first gluing parameter and the second gluing parameter;

[0153] Determining a first control parameter of the robotic arm according to the third gluing parameter;

[0154] The robot arm operates according to the first control parameter, and the glue gun is controlled to operate according to the third glue coating parameter, so as to perform glue filling operation on the target glue coating position point.

[0155] Optionally, in the aspect of determining the first gluing parameter according to the first adjustment parameter and the reference gluing parameter, the program includes instructions for executing the following steps:

[0156] Determine a fourth gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0157] Dividing the glued area image into a plurality of areas, each area being equal in area;

[0158] Determining a first signal-to-noise ratio of the high-frequency component image;

[0159] Determine a regional high-frequency component image corresponding to each of the multiple regions according to the high-frequency component image, to obtain multiple regional high-frequency component images;

[0160] Determine the number of feature points and the average energy value of the feature points in each region of the plurality of regional high-frequency component images to obtain a plurality of feature point numbers and a plurality of average energy values;

[0161] Determining a first standard deviation corresponding to the number of the plurality of feature points;

[0162] Determining a second standard deviation corresponding to the plurality of average energy values;

[0163] determining a weight pair corresponding to the first signal-to-noise ratio;

[0164] Determine a target standard deviation according to the first standard deviation, the second standard deviation and the weight pair;

[0165] determining a first fine-tuning parameter corresponding to the target standard deviation;

[0166] The fourth gluing parameter is fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter.

[0167] Optionally, the preset range includes an upper threshold and a lower threshold, and the program further includes instructions for executing the following steps:

[0168] When the first energy ratio is less than the lower limit threshold, the step of acquiring the first image with the target gluing position point as the focus and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image is executed again.

[0169] It can be seen that the gluing robot described in the embodiment of the present application, firstly, when the gluing gun completes the gluing work between the automobile frame and the glass, since the glued area and the un-glueed area are obvious, the gluing trajectory between the automobile frame and the glass can be identified by a visual device, and the gluing position points constituting the gluing trajectory are determined according to the gluing trajectory to obtain a plurality of gluing position points, and a first image is acquired with the target gluing position point as the focus, thereby ensuring the image quality of the gluing area image, and the corresponding gluing area image between two gluing position points adjacent to the target gluing position point is intercepted by the first image, and an image of a region where colloid exists between the two gluing position points of the gluing area image is obtained. In this way, the gluing trajectory can be divided into multiple segments, and the gluing inspection can be performed in segments, thereby preventing the phenomenon of missing glue, wrong glue, uneven glue, etc. Secondly, due to different attribute information, the requirements corresponding to the gluing detail features that need to be checked are different. The preset conditions reflect the requirements corresponding to the gluing detail features that need to be checked to a certain extent, which can not only be based on the gluing process, the gluing method and the like. And the corresponding gluing effect determines the corresponding requirements of the corresponding gluing detail features that need to be checked, as well as the corresponding high-frequency feature extraction algorithm and algorithm control parameters. Thirdly, when the first energy ratio is within the preset range, it means that the high-frequency details are appropriate, then the high-frequency component image can be feature extracted according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set. When the high-frequency feature set meets the preset conditions, it means that the image of the glued area corresponding to the target gluing position point meets the gluing requirements. Therefore, the image recognition technology can be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection. When the high-frequency feature set does not meet the preset conditions, that is, the image of the glued area corresponding to the target gluing position point does not meet the gluing requirements, it means that the gluing is qualified. Then, the gluing gun can be controlled by the robotic arm to perform glue filling operations on the target gluing position point. In this way, not only can the image recognition technology be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection, but also glue filling can be further performed, thereby deeply ensuring the gluing effect and achieving precise glue filling to ensure the gluing effect.

[0170] Figure 3 300 is a functional unit composition block diagram of a visual system 300 of an automobile processing workstation involved in an embodiment of the present application. The visual system 300 of the automobile processing workstation is applied to a gluing robot, the gluing robot comprises: a robot body, a gluing gun, and a visual device, the robot body is connected to the gluing gun through a mechanical arm, the visual system 300 of the automobile processing workstation comprises: a visual recognition unit 301, an acquisition unit 302, a determination unit 303, an extraction unit 304, a detection unit 305 and a control unit 306, wherein,

[0171] The visual recognition unit 301 is used to identify the gluing track between the automobile frame and the glass through the visual device when the gluing gun completes the gluing work between the automobile frame and the glass, determine the gluing position points constituting the gluing track according to the gluing track, and obtain multiple gluing position points, and the distance between adjacent gluing position points is within a preset distance range; obtain a first image with the target gluing position point as the focus, and intercept the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image;

[0172] The acquisition unit 302 is used to acquire the first attribute information of the target glue coating position point;

[0173] The determining unit 303 is used to determine a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine a preset condition corresponding to the first attribute information;

[0174] The extraction unit 304 is used to extract the high-frequency component image of the glue-coated area image;

[0175] The determining unit 303 is further configured to determine a first energy ratio between the high-frequency component image and the glue-coated area image;

[0176] The extraction unit 304 is further configured to extract features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set when the first energy ratio is within a preset range;

[0177] The detection unit 305 is used to detect whether the high-frequency feature set meets a preset condition;

[0178] The determination unit 303 is further configured to determine that the gluing area image corresponding to the target gluing position point meets the gluing requirement when the high-frequency feature set meets the preset condition;

[0179] The control unit 306 is used to control the glue gun to perform glue filling operation on the target glue coating position point through the robotic arm when the high-frequency feature set does not meet the preset condition.

[0180] Optionally, the high-frequency component image includes multiple layers of first high-frequency component images; the preset condition includes multiple first preset conditions, and each layer of high-frequency component image corresponds to one first preset condition;

[0181] In terms of detecting whether the high-frequency feature set meets a preset condition, the detection unit 305 is specifically used to:

[0182] Determine a first high-frequency feature set of each first high-frequency component image in the multiple layers of first high-frequency component images according to the high-frequency feature set, to obtain a plurality of first high-frequency feature sets;

[0183] Detect whether each of the multiple first high-frequency feature sets meets the corresponding first preset condition among the multiple first preset conditions. If so, confirm that the high-frequency feature set meets the preset condition. Otherwise, confirm that the high-frequency feature set does not meet the preset condition.

[0184] Optionally, the preset condition includes a reference high-frequency feature set, and in controlling the gluing gun by the robotic arm to perform a gluing operation on the target gluing position point, the control unit 306 is specifically used to:

[0185] Determining a first degree of deviation between the high-frequency feature set and the reference high-frequency feature set;

[0186] Obtaining reference gluing parameters corresponding to the target gluing position point;

[0187] determining a first adjustment parameter corresponding to the first deviation;

[0188] Determine a first gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0189] Determining a second gluing parameter according to the gluing area image;

[0190] Determine a third gluing parameter according to the first gluing parameter and the second gluing parameter;

[0191] Determining a first control parameter of the robotic arm according to the third gluing parameter;

[0192] The robot arm operates according to the first control parameter, and the glue gun is controlled to operate according to the third glue coating parameter, so as to perform glue filling operation on the target glue coating position point.

[0193] Optionally, in determining the first gluing parameter according to the first adjustment parameter and the reference gluing parameter, the control unit 306 is specifically configured to:

[0194] Determine a fourth gluing parameter according to the first adjustment parameter and the reference gluing parameter;

[0195] Dividing the glued area image into a plurality of areas, each area being equal in area;

[0196] Determining a first signal-to-noise ratio of the high-frequency component image;

[0197] Determine a regional high-frequency component image corresponding to each of the multiple regions according to the high-frequency component image, to obtain multiple regional high-frequency component images;

[0198] Determine the number of feature points and the average energy value of the feature points in each region of the plurality of regional high-frequency component images to obtain a plurality of feature point numbers and a plurality of average energy values;

[0199] Determining a first standard deviation corresponding to the number of the plurality of feature points;

[0200] Determining a second standard deviation corresponding to the plurality of average energy values;

[0201] determining a weight pair corresponding to the first signal-to-noise ratio;

[0202] Determine a target standard deviation according to the first standard deviation, the second standard deviation and the weight pair;

[0203] determining a first fine-tuning parameter corresponding to the target standard deviation;

[0204] The fourth gluing parameter is fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter.

[0205] Optionally, the preset range includes an upper threshold and a lower threshold, and the visual system 300 of the automobile processing workstation is further specifically used for:

[0206] When the first energy ratio is less than the lower limit threshold, the step of acquiring the first image with the target gluing position point as the focus and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image is executed again.

[0207] It can be seen that the visual system of the automobile processing workstation described in the embodiment of the present application, firstly, when the gluing gun completes the gluing work between the automobile frame and the glass, since the glued area and the un-glueed area are obvious, the gluing track between the automobile frame and the glass can be identified by the visual device, and the gluing position points constituting the gluing track are determined according to the gluing track to obtain a plurality of gluing position points, and the first image is acquired with the target gluing position point as the focus, thereby ensuring the image quality of the gluing area image, and the corresponding gluing area image between two gluing position points adjacent to the target gluing position point is intercepted by the first image, and the image of the area where the colloid exists between the two gluing position points of the gluing area image is obtained. In this way, the gluing track can be divided into multiple sections, and the gluing inspection can be performed in sections, thereby preventing the phenomenon of missing glue, wrong gluing, uneven gluing, etc. Secondly, due to different attribute information, the requirements corresponding to the gluing detail features that need to be checked are different. The preset conditions reflect the requirements corresponding to the gluing detail features that need to be checked to a certain extent, which can not only be based on the gluing process, coating The gluing method and the corresponding gluing effect determine the corresponding requirements of the corresponding gluing detail features that need to be checked, as well as the corresponding high-frequency feature extraction algorithm and algorithm control parameters. Thirdly, when the first energy ratio is within the preset range, it means that the high-frequency details are appropriate, and the high-frequency component image can be feature extracted according to the first high-frequency feature extraction algorithm and the first algorithm control parameters to obtain a high-frequency feature set. When the high-frequency feature set meets the preset conditions, it means that the image of the glued area corresponding to the target gluing position point meets the gluing requirements. Therefore, the image recognition technology can be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection. When the high-frequency feature set does not meet the preset conditions, that is, the image of the glued area corresponding to the target gluing position point does not meet the gluing requirements, it means that the gluing is qualified. Then, the gluing gun can be controlled by the robotic arm to perform gluing operations on the target gluing position point. In this way, not only can the image recognition technology be used to deeply inspect the gluing effect, which helps to improve the efficiency of gluing inspection, but also glue can be further repaired, thereby deeply ensuring the gluing effect and achieving precise glue repair to ensure the gluing effect.

[0208] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method described in the above method embodiment, and the above computer includes a gluing robot.

[0209] The present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any method described in the method embodiment. The computer program product may be a software installation package, and the computer includes a glue coating robot.

[0210] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0211] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0212] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0213] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0214] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0215] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the above-mentioned methods in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0216] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable memory, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0217] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A visual method for an automobile processing workstation, characterized in that: Applied to a gluing robot, the gluing robot comprises: a robot body, a gluing gun, and a visual device, the robot body is connected to the gluing gun via a mechanical arm, and the method comprises: When the glue gun completes the glue coating work between the automobile frame and the glass, the glue coating track between the automobile frame and the glass is identified by the visual device, and the glue coating position points constituting the glue coating track are determined according to the glue coating track to obtain a plurality of glue coating position points, and the distance between adjacent glue coating position points is within a preset distance range; Acquire a first image with the target gluing position point as the focus, and intercept a corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image; Acquire first attribute information of the target gluing position point; determine a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine a preset condition corresponding to the first attribute information; Extracting a high-frequency component image of the glue-coated area image; determining a first energy ratio between the high frequency component image and the glued area image; When the first energy ratio is within a preset range, extracting features from the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set; Detecting whether the high-frequency feature set meets a preset condition; When the high-frequency feature set meets the preset condition, determining that the gluing area image corresponding to the target gluing position point meets the gluing requirement; When the high-frequency feature set does not meet the preset condition, the gluing gun is controlled by the robotic arm to perform a gluing operation on the target gluing position point; The preset condition includes referring to a high-frequency feature set, and the step of controlling the gluing gun by the robotic arm to perform a gluing operation on the target gluing position point includes: Determine a first deviation between the high-frequency feature set and the reference high-frequency feature set; obtain a reference gluing parameter corresponding to the target gluing position point; determine a first adjustment parameter corresponding to the first deviation; determine a first gluing parameter based on the first adjustment parameter and the reference gluing parameter; determine a second gluing parameter based on the gluing area image; determine a third gluing parameter based on the first gluing parameter and the second gluing parameter; determine a first control parameter of the robotic arm based on the third gluing parameter; operate the robotic arm with the first control parameter, and control the gluing gun to operate with the third gluing parameter, so as to perform a gluing operation on the target gluing position point.

2. The method according to claim 1, characterized in that The high-frequency component image includes multiple layers of first high-frequency component images; the preset condition includes multiple first preset conditions, and each layer of high-frequency component image corresponds to one first preset condition; The detecting whether the high-frequency feature set meets a preset condition includes: Determine a first high-frequency feature set of each first high-frequency component image in the multiple layers of first high-frequency component images according to the high-frequency feature set, to obtain a plurality of first high-frequency feature sets; Detect whether each of the multiple first high-frequency feature sets meets the corresponding first preset condition among the multiple first preset conditions. If so, confirm that the high-frequency feature set meets the preset condition. Otherwise, confirm that the high-frequency feature set does not meet the preset condition.

3. The method according to claim 1 or 2, characterized in that: The determining the first gluing parameter according to the first adjustment parameter and the reference gluing parameter includes: Determine a fourth gluing parameter according to the first adjustment parameter and the reference gluing parameter; Dividing the glued area image into a plurality of areas, each area being equal in area; Determining a first signal-to-noise ratio of the high-frequency component image; Determine a regional high-frequency component image corresponding to each of the multiple regions according to the high-frequency component image, to obtain multiple regional high-frequency component images; Determine the number of feature points and the average energy value of the feature points in each region of the plurality of regional high-frequency component images to obtain a plurality of feature point numbers and a plurality of average energy values; Determining a first standard deviation corresponding to the number of the plurality of feature points; Determining a second standard deviation corresponding to the plurality of average energy values; determining a weight pair corresponding to the first signal-to-noise ratio; Determine a target standard deviation according to the first standard deviation, the second standard deviation and the weight pair; determining a first fine-tuning parameter corresponding to the target standard deviation; The fourth gluing parameter is fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter.

4. The method according to claim 1 or 2, characterized in that: The preset range includes an upper threshold and a lower threshold, and the method further includes: When the first energy ratio is less than the lower limit threshold, the step of acquiring the first image with the target gluing position point as the focus and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image is executed again.

5. A visual system for an automobile processing workstation, characterized in that: Applied to a gluing robot, the gluing robot comprises: a robot body, a gluing gun, and a visual device, the robot body is connected to the gluing gun via a mechanical arm, the system comprises: a visual recognition unit, an acquisition unit, a determination unit, an extraction unit, a detection unit, and a control unit, wherein: The visual recognition unit is used to identify the gluing track between the automobile frame and the glass through the visual device when the gluing gun completes the gluing work between the automobile frame and the glass, determine the gluing position points constituting the gluing track according to the gluing track, and obtain a plurality of gluing position points, wherein the distance between adjacent gluing position points is within a preset distance range; acquire a first image with the target gluing position point as the focus, and intercept the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image; The acquisition unit is used to acquire the first attribute information of the target gluing position point; The determining unit is used to determine a first high-frequency feature extraction algorithm corresponding to the first attribute information and a first algorithm control parameter corresponding to the first high-frequency feature extraction algorithm, and determine a preset condition corresponding to the first attribute information; The extraction unit is used to extract the high-frequency component image of the glue-coated area image; The determining unit is further used to determine a first energy ratio between the high-frequency component image and the glue-coated area image; The extraction unit is further configured to extract features of the high-frequency component image according to the first high-frequency feature extraction algorithm and the first algorithm control parameter to obtain a high-frequency feature set when the first energy ratio is within a preset range; The detection unit is used to detect whether the high-frequency feature set meets a preset condition; The determination unit is further configured to determine that the gluing area image corresponding to the target gluing position point meets the gluing requirement when the high-frequency feature set meets the preset condition; The control unit is used to control the glue gun to perform glue filling operation on the target glue coating position point through the mechanical arm when the high-frequency feature set does not meet the preset condition; in, The preset condition includes a reference high-frequency feature set. In the aspect of controlling the gluing gun by the robotic arm to perform a gluing operation on the target gluing position point, the control unit is specifically used to: Determining a first degree of deviation between the high-frequency feature set and the reference high-frequency feature set; Acquire a reference gluing parameter corresponding to the target gluing position point; determine a first adjustment parameter corresponding to the first deviation; Determine a first gluing parameter according to the first adjustment parameter and the reference gluing parameter; Determining a second gluing parameter according to the gluing area image; Determine a third gluing parameter according to the first gluing parameter and the second gluing parameter; Determining a first control parameter of the robotic arm according to the third gluing parameter; The robot arm operates according to the first control parameter, and the glue gun is controlled to operate according to the third glue coating parameter, so as to perform glue filling operation on the target glue coating position point.

6. The system according to claim 5, characterized in that The high-frequency component image includes multiple layers of first high-frequency component images; the preset condition includes multiple first preset conditions, and each layer of high-frequency component image corresponds to one first preset condition; In terms of detecting whether the high-frequency feature set meets a preset condition, the detection unit is specifically used to: Determine a first high-frequency feature set of each first high-frequency component image in the multiple layers of first high-frequency component images according to the high-frequency feature set, to obtain a plurality of first high-frequency feature sets; Detect whether each of the multiple first high-frequency feature sets meets the corresponding first preset condition among the multiple first preset conditions. If so, confirm that the high-frequency feature set meets the preset condition. Otherwise, confirm that the high-frequency feature set does not meet the preset condition.

7. The system according to claim 5 or 6, characterized in that: In the aspect of determining the first gluing parameter according to the first adjustment parameter and the reference gluing parameter, the control unit is specifically used for: Determine a fourth gluing parameter according to the first adjustment parameter and the reference gluing parameter; Dividing the glued area image into a plurality of areas, each area being equal in area; Determining a first signal-to-noise ratio of the high-frequency component image; Determine a regional high-frequency component image corresponding to each of the multiple regions according to the high-frequency component image, to obtain multiple regional high-frequency component images; Determine the number of feature points and the average energy value of the feature points in each region of the plurality of regional high-frequency component images to obtain a plurality of feature point numbers and a plurality of average energy values; Determining a first standard deviation corresponding to the number of the plurality of feature points; Determining a second standard deviation corresponding to the plurality of average energy values; determining a weight pair corresponding to the first signal-to-noise ratio; Determine a target standard deviation according to the first standard deviation, the second standard deviation and the weight pair; determining a first fine-tuning parameter corresponding to the target standard deviation; The fourth gluing parameter is fine-tuned according to the first fine-tuning parameter to obtain the first gluing parameter.

8. The system according to claim 5 or 6, characterized in that: The preset range includes an upper threshold and a lower threshold, and the system is further specifically used for: When the first energy ratio is less than the lower limit threshold, the step of acquiring the first image with the target gluing position point as the focus and intercepting the corresponding gluing area image between two gluing position points adjacent to the target gluing position point through the first image is executed again.

Images