An automatic inspection system for automobile plastic injection molded parts based on dual robot technology

Abstract

The invention relates to the technical field of automatic detection of automobile plastic injection molded parts, and specifically discloses an automatic detection system for automobile plastic injection molded parts based on dual robot technology, the system comprising: an injection molded part placement position assessment module, an injection molded part image quality assessment module, an injection molded part injection molding quality analysis module, a database and an injection molded part quality sorting module; the invention conducts a refined assessment on whether the placement position of the automobile plastic injection molded parts is accurate, and adjusts the placement position, and at the same time, flexibly collects images of various parts through a track-type detector, conducts an in-depth analysis of the quality of the collected images, and finally considers that the importance of the injection molding quality of each part in the comprehensive quality assessment is different, thereby reducing the possibility that the injection molded parts with inaccurate positions may cause the production line to stagnate, avoiding treating the defects of the secondary parts as the same as the defects of the main parts, thereby avoiding misleading the judgment of the overall quality.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic detection of automobile plastic injection molded parts, in particular to an automatic detection system for automobile plastic injection molded parts based on dual-robot technology. Background Art

[0002] With the rapid development of the automotive industry, quality control of automotive plastic injection molded parts, as key components widely used in vehicle interiors and exteriors, is crucial. Traditional manual inspection methods are inefficient, subjective, prone to fatigue and resulting in missed or incorrect inspections, and are difficult to adapt to the demands of large-scale, high-precision production. Therefore, to comprehensively, efficiently, and accurately complete inspection tasks, there is an urgent need to develop an efficient, highly accurate, and flexible automated inspection system.

[0003] For example, Chinese patent publication number CN118549436B discloses a defect detection system for automobile plastic injection molded parts based on machine vision technology, including: analyzing the crack defect degree, internal bubble defect degree, surface color defect degree and deformation degree of each injection molded part to be detected to obtain the overall defect degree of each injection molded part to be detected, analyzing the conformity of each through hole of each injection molded part to be detected and the cutting conformity of each injection molded part to be detected to obtain the local defect degree of each injection molded part to be detected. The present invention detects defects of automobile plastic injection molded parts through machine vision technology, and analyzes the overall defect degree of each injection molded part to be detected. It not only analyzes the surface defects of automobile plastic injection molded parts, but also analyzes the internal bubble defects of automobile plastic injection molded parts, thereby ensuring the reliability of product quality and improving customer satisfaction.

[0004] The above existing technologies still have the following problems: 1. Currently, the placement work is completed after the automotive plastic injection molded parts to be inspected are placed on the conveyor belt. Whether the placement position of the automotive plastic injection molded parts is accurate is not carefully evaluated, and the position of the automotive plastic injection molded parts with inaccurate placement is not adjusted. Position deviation may cause errors in the inspection results, thereby affecting the accurate judgment of the quality of the injection molded parts. On the automated production line, injection molded parts with inaccurate positioning may cause the production line to stagnate, further reducing production efficiency.

[0005] 2. Currently, when capturing images of automotive plastic injection molded parts, most of them are scanned using fixed detectors, rather than flexibly capturing images of various parts using track-type detectors placed on the inner wall of the testing equipment. Furthermore, the quality of the captured images is not deeply analyzed, which reduces the comprehensiveness of image capture of automotive plastic injection molded parts and the quality of image capture cannot be guaranteed, thereby reducing the accuracy of subsequent image information extraction, and further reducing the accuracy and reliability of subsequent injection quality analysis of automotive plastic injection molded parts.

[0006] 3. When comprehensively analyzing the injection molding quality of automotive plastic injection molding parts, the different importance of the injection molding quality of various parts in the comprehensive quality assessment is not taken into account. If this difference is not taken into account, defects in minor parts may be equated with defects in major parts, thereby misleading the judgment of overall quality. The injection molding quality of certain key parts directly affects the overall performance of the injection molding parts. If these key parts are not given sufficient attention, potential quality problems may be ignored, which in turn affects the reliability and safety of the product. Summary of the Invention

[0007] In view of this, in order to solve the problems raised in the above background technology, an automatic detection system for automobile plastic injection molded parts based on dual robot technology is proposed.

[0008] The objectives of the present invention can be achieved through the following technical solutions: The present invention provides an automatic detection system for automobile plastic injection molded parts based on dual robot technology, including: an injection molded part placement position evaluation module, which is used to construct a three-dimensional model of a conveyor belt in a target production line. When robot A in the target production line places the processed automobile plastic injection molded parts on the conveyor belt, the three-dimensional coordinate values ​​corresponding to each monitoring point in the automobile plastic injection molded parts are obtained to evaluate whether the position of the automobile plastic injection molded parts is accurately placed. If it is accurate, the injection molded part image quality evaluation module is executed; if it is inaccurate, the position of the automobile plastic injection molded parts is adjusted.

[0009] The injection molded parts image quality assessment module is used to collect images of various parts of the automobile plastic injection molded parts when the automobile plastic injection molded parts are transported to the testing equipment via a conveyor belt. The track-type detector installed on the inner wall of the testing equipment is used to evaluate whether the image quality of each part of the automobile plastic injection molded parts meets the requirements. If so, the injection molded parts injection quality analysis module is executed. If not, the image of the parts whose image quality does not meet the requirements is re-collected.

[0010] The injection molding quality analysis module is used to record the parts whose image quality meets the requirements as target parts, extract the injection molding quality information from the images corresponding to each target part in the automobile plastic injection molding part, and analyze the injection molding quality coefficient of the automobile plastic injection molding part.

[0011] The database is used to store the standard positions corresponding to the monitoring points when the automobile plastic injection molded parts are placed on the conveyor belt, and to store the weights of the target parts in the automobile plastic injection molded parts in the injection quality assessment of the automobile plastic injection molded parts.

[0012] The injection molded parts quality sorting module is used to perform quality sorting on automotive plastic injection molded parts by robot B in the target production line.

[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention makes a detailed assessment of whether the placement position of automobile plastic injection molded parts is accurate, thereby adjusting the position of automobile plastic injection molded parts that are not accurately placed, avoiding errors in the detection results that may be caused by position deviation, thereby avoiding affecting the accurate judgment of the quality of the injection molded parts, reducing the possibility that the production line may be stagnant due to injection molded parts with inaccurate placement, and further improving production efficiency.

[0014] (2) The present invention flexibly collects images of various parts of the detection equipment through a track-type detector placed on the inner wall of the detection equipment, and at the same time conducts in-depth analysis of the quality of the collected images, thereby improving the comprehensiveness of image collection of automobile plastic injection molded parts and ensuring the quality of image collection, thereby improving the accuracy of subsequent image information extraction, and further improving the accuracy and reliability of subsequent injection quality analysis of automobile plastic injection molded parts.

[0015] (3) The present invention considers the different importance of the injection molding quality of various parts in the comprehensive quality assessment when comprehensively analyzing the injection molding quality of automobile plastic injection molding parts. Taking this difference into consideration, it avoids treating defects in minor parts as the same as defects in major parts, thereby avoiding misleading judgment on the overall quality. The injection molding quality of certain key parts directly affects the overall performance of the injection molding parts. Sufficient attention is paid to these key parts to avoid ignoring potential quality problems, thereby improving the reliability and safety of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is a schematic diagram of the system module structure connection of the present invention.

[0018] Figure 2 This is a flow chart for evaluating the placement position of automotive plastic injection molded parts according to the present invention.

[0019] Figure 3 This is a flow chart for image quality assessment of automotive plastic injection molded parts according to the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1 As shown, the present invention provides an automatic detection system for automobile plastic injection molded parts based on dual robot technology, including: an injection molded part placement position assessment module, an injection molded part image quality assessment module, an injection molded part injection quality analysis module, a database and an injection molded part quality sorting module.

[0022] The injection molded part placement position assessment module is connected to the injection molded part image quality assessment module, the injection molded part image quality assessment module is connected to the injection molded part injection quality analysis module, the injection molded part injection quality analysis module is connected to the injection molded part quality sorting module, and the injection molded part placement position assessment module, the injection molded part image quality assessment module and the injection molded part injection quality analysis module are all connected to the database.

[0023] See also Figure 2 As shown, the injection molded part placement position assessment module is used to construct a three-dimensional model of the conveyor belt in the target production line. When the robot A in the target production line places the processed automobile plastic injection molded parts on the conveyor belt, the three-dimensional coordinate values ​​corresponding to each monitoring point in the automobile plastic injection molded parts are obtained to assess whether the position of the automobile plastic injection molded parts is accurately placed. If accurate, the injection molded part image quality assessment module is executed. If inaccurate, the position of the automobile plastic injection molded parts is adjusted.

[0024] In a specific embodiment of the present invention, the specific method of constructing the three-dimensional model of the conveyor belt in the target production line is: taking the plane of the conveyor belt in the target production line as The plane is perpendicular to the upward direction of the conveyor belt. Build a three-dimensional model of the conveyor belt in the target production line in the axial direction.

[0025] In a specific embodiment of the present invention, the specific method for evaluating whether the position of the automobile plastic injection molded part is accurately placed is as follows: A1, the three-dimensional coordinate values ​​corresponding to each monitoring point in the automobile plastic injection molded part are recorded as ,in, Indicates the number of the monitoring point. .

[0026] A2. Extract the standard positions of each monitoring point when the automobile plastic injection molded parts are placed on the conveyor belt from the database and insert them into the 3D model of the conveyor belt in the target production line to obtain the standard 3D coordinate values ​​of each monitoring point when the automobile plastic injection molded parts are placed on the conveyor belt, which are recorded as .

[0027] A3. Calculate the position deviation index of each monitoring point in the automotive plastic injection molded parts ,

[0028] ,in, and Respectively indicate the setting permission axis, Axis and The position offset value corresponding to the axis.

[0029] A4. Calculate the position offset index of automotive plastic injection molded parts , ,in, Indicates the number of monitoring points.

[0030] A5. Compare the position offset index corresponding to the automobile plastic injection molded part with the set allowable position offset index. If the position offset index corresponding to the automobile plastic injection molded part is less than the set allowable position offset index, it indicates that the position of the automobile plastic injection molded part is accurately placed. Otherwise, it indicates that the position of the automobile plastic injection molded part is inaccurate.

[0031] In a specific embodiment of the present invention, the method for adjusting the position of the automobile plastic injection molding part is: when the position of the automobile plastic injection molding part is not accurately placed, the automobile plastic injection molding part position adjustment instruction is sent to robot A; when robot A receives the automobile plastic injection molding part position adjustment instruction, it clamps the automobile plastic injection molding part from the conveyor belt and repositions it; and evaluates whether the position of the automobile plastic injection molding part after position adjustment is accurately placed in the same way as the evaluation method for whether the position of the automobile plastic injection molding part is accurately placed; if the placement is accurate, the position adjustment of the automobile plastic injection molding part is terminated; if the placement is inaccurate, the automobile plastic injection molding part is adjusted again in the above method, and so on, until the position of the automobile plastic injection molding part is accurately placed.

[0032] The embodiment of the present invention performs a detailed assessment of whether the placement position of automobile plastic injection molded parts is accurate, thereby adjusting the position of automobile plastic injection molded parts that are not accurately placed, avoiding errors in detection results that may be caused by position deviations, thereby avoiding affecting the accurate judgment of the quality of the injection molded parts, reducing the possibility that inaccurately positioned injection molded parts may cause the production line to stagnate, and further improving production efficiency.

[0033] See also Figure 3As shown, the injection molded part image quality assessment module is used to collect images corresponding to various parts of the automobile plastic injection molded parts when the automobile plastic injection molded parts are transported to the testing equipment via a conveyor belt. The track-type detector installed on the inner wall of the testing equipment is used to evaluate whether the image quality corresponding to various parts of the automobile plastic injection molded parts meets the requirements. If so, the injection molded part injection quality analysis module is executed. If not, the image of the parts whose image quality does not meet the requirements is re-collected.

[0034] In a specific embodiment of the present invention, the specific process of evaluating whether the image quality corresponding to each part of the automobile plastic injection molded part meets the requirements is as follows: B1, locating the pixel value of each pixel point from the image corresponding to each part of the automobile plastic injection molded part, and recording the pixel value of each pixel point corresponding to each part of the automobile plastic injection molded part as ,in, Indicates the number of the part, , Indicates the pixel number. .

[0035] It should be noted that the images corresponding to various parts of the automobile plastic injection molded part are imported into the image processing software, and the pixel value of each pixel point is obtained from the image processing software.

[0036] B2. Calculate the variance of pixel values ​​corresponding to each part of the automobile plastic injection molded part ,

[0037] ,in, Indicates the number of pixels.

[0038] B3. Calculate the image quality coefficients for each part of the automotive plastic injection molded part , ,in, Indicates the pixel value variance of the setting reference.

[0039] It should be noted that the larger the variance, the higher the image clarity. Conversely, the smaller the variance, the lower the image clarity.

[0040] B4. Compare the image quality coefficient corresponding to each part of the automobile plastic injection molded part with the set reference image quality coefficient. If the image quality coefficient corresponding to a part of the automobile plastic injection molded part is smaller than the set reference image quality coefficient, it indicates that the image quality corresponding to the part of the automobile plastic injection molded part does not meet the requirements. Otherwise, it indicates that the image quality corresponding to the part of the automobile plastic injection molded part meets the requirements. In this way, it is determined whether the image quality corresponding to each part of the automobile plastic injection molded part meets the requirements.

[0041] In a specific embodiment of the present invention, the specific method of re-capturing images of parts whose image quality does not meet the requirements is: marking the parts whose image quality does not meet the requirements as abnormal parts, sending the abnormal parts in the automobile plastic injection molded parts to the track-type detector on the inner wall of the detection equipment, and after receiving the corresponding instructions, the track-type detector re-captures images of the abnormal parts in the automobile plastic injection molded parts, and evaluates whether the image quality corresponding to each abnormal part in the automobile plastic injection molded parts meets the requirements in the same way as the evaluation method of whether the image quality corresponding to each part in the automobile plastic injection molded parts meets the requirements. If yes, the image acquisition is terminated. If not, the image acquisition is re-captured for the abnormal parts whose image quality does not meet the requirements by analogy.

[0042] The embodiment of the present invention uses a track-type detector placed on the inner wall of the detection equipment to flexibly collect images of various parts of the equipment, and at the same time conducts in-depth analysis of the quality of the collected images, thereby improving the comprehensiveness of image acquisition of automobile plastic injection molded parts and ensuring the quality of image acquisition, thereby improving the accuracy of subsequent image information extraction, and further improving the accuracy and reliability of subsequent injection quality analysis of automobile plastic injection molded parts.

[0043] The injection molding quality analysis module is used to record the parts whose image quality meets the requirements as target parts, extract the injection molding quality information from the images corresponding to the target parts in the automobile plastic injection molding parts, and analyze the injection molding quality coefficient of the automobile plastic injection molding parts.

[0044] In a specific embodiment of the present invention, the injection molded part quality information includes the number of internal pores, the volume of each internal pore, the number of internal cracks, the length of each internal crack, the number of surface scratches, and the length of each surface scratch.

[0045] It should be noted that since the number of internal pores, the volume of each internal pore, the number of internal cracks and the length of each internal crack of automobile plastic injection molding parts cannot be captured by the high-definition camera in the track-type detector installed on the inner wall of the detection equipment, the image acquisition instrument here is an X-ray detector installed in the track-type detector on the inner wall of the detection equipment. The X-ray detector can capture both the internal conditions and the surface conditions of automobile plastic injection molding parts.

[0046] It should be noted that the positioning method for the number of internal pores, the volume of each internal pore, the number of internal cracks, the length of each internal crack, the number of surface scratches and the length of each surface scratch corresponding to each target part in the automotive plastic injection molded part is as follows: 1) Image enhancement: The collected image is enhanced, such as contrast adjustment, sharpening, etc., to improve the image quality and remove noise and interference factors in the image, such as background, reflection, etc.; 2) Image segmentation: The target part in the injection molded part image is separated from the background and other non-target parts using an image segmentation algorithm. Image segmentation can be performed using methods such as threshold segmentation, edge detection, and region growing; 3) Feature extraction: Feature extraction is performed on the segmented target part image, including features such as internal pores, internal cracks, and surface scratches. Shape segmentation can be used to extract features. 4) Pore detection: Use image processing algorithms, such as connected domain analysis and morphological processing, to detect pores inside the molded part, count the number of pores, and measure the volume of each pore (which can be estimated through 3D reconstruction or assuming the pores are spheres). 5) Crack detection: Use image processing algorithms, such as edge detection and morphological processing, to detect cracks inside the molded part, count the number of cracks, and measure the length of each crack (which can be estimated by tracking the crack edge). 6) Scratch detection: Use image processing algorithms, such as edge detection and texture analysis, to detect scratches on the surface of the molded part, count the number of scratches, and measure the length of each scratch (which can be estimated by tracking the scratch path).

[0047] In a specific embodiment of the present invention, the specific process of analyzing the injection molding quality coefficient of automobile plastic injection molding parts is as follows: C1, extracting the number of internal pores, the volume of each internal pore, the number of internal cracks, the length of each internal crack, the number of surface scratches and the length of each surface scratch from the injection molding quality information corresponding to each target part of the automobile plastic injection molding part, and calculating the injection molding quality coefficient corresponding to each target part of the automobile plastic injection molding part ,in, Indicates the number of the target site,

[0048] .

[0049] In a specific embodiment of the present invention, the specific process of calculating the injection molding quality coefficient corresponding to each target part in the automobile plastic injection molded part is as follows: D1, the number of internal pores corresponding to each target part in the automobile plastic injection molded part is recorded as .

[0050] D2. Accumulate the internal pore volumes corresponding to the target locations in the automotive plastic injection molded parts to obtain the total internal pore volume corresponding to the target locations in the automotive plastic injection molded parts, and record it as .

[0051] D3. Calculate the internal porosity anomaly index corresponding to each target location in automotive plastic injection molded parts ,

[0052] ,in, and Respectively represent the internal pore number and internal pore volume of the set reference, Represents a natural constant.

[0053] D4. Calculate the internal crack anomaly index corresponding to each target location in the automotive plastic injection molded parts based on the number of internal cracks, the length of each internal crack, the number of surface scratches, and the length of each surface scratch. and surface scratch anomaly index .

[0054] It should be noted that the specific process of calculating the internal crack abnormality index corresponding to each target position in the automobile plastic injection molded part is: the number of internal cracks corresponding to each target position in the automobile plastic injection molded part is recorded as .

[0055] The maximum internal crack length is extracted from the internal crack lengths corresponding to each target part in the automobile plastic injection molded part and recorded as .

[0056] Calculate the internal crack anomaly index corresponding to each target location in automotive plastic injection molded parts ,

[0057] ,in, and Respectively represent the number of internal cracks and the length of internal cracks for setting reference.

[0058] It should be noted that the specific process of calculating the surface scratch abnormality index corresponding to each target position in the automobile plastic injection molded part is as follows: the number of surface scratches corresponding to each target position in the automobile plastic injection molded part is recorded as .

[0059] The average length of each surface scratch corresponding to each target part in the automobile plastic injection molded part is calculated to obtain the average scratch length corresponding to each target part in the automobile plastic injection molded part and recorded as .

[0060] Calculate the surface scratch anomaly index corresponding to each target area in automotive plastic injection molded parts ,

[0061] ,in, and They represent the number of scratches and the length of scratches on the reference surface.

[0062] It should be noted that the aforementioned setting references of internal pore number, internal pore volume, internal crack number, internal crack length, surface scratch number and scratch length are all obtained from the industry production specifications for automotive plastic injection molding parts.

[0063] D5. Calculate the injection molding quality coefficient corresponding to each target part in the automotive plastic injection molding part ,

[0064] ,in, 、 and They represent the weights of the set internal pore abnormality index, internal crack abnormality index and surface scratch abnormality index corresponding to the injection quality assessment. .

[0065] In a specific embodiment of the present invention, The setting value is 0.4, The setting value is 0.3, The setting value is 0.3. When calculating the injection quality coefficient corresponding to each target part in automobile plastic injection molding, the internal pore abnormality index, internal crack abnormality index and surface scratch abnormality index are all important evaluation indicators. They each reflect the quality problems of injection molding parts in different aspects. Porosity is one of the key factors affecting the quality of injection molding parts. The existence of pores not only affects the strength and rigidity of injection molding parts, but may also lead to poor sealing, decreased corrosion resistance and wear resistance. Therefore, the degree of internal pore abnormality is more important for calculating the injection quality coefficient corresponding to each target part in automobile plastic injection molding.

[0066] C2. Extract the weight of each target part in the injection quality assessment of automobile plastic injection molded parts from the database and record it as .

[0067] It should be noted that in the injection molding quality assessment of automotive plastic injection molded parts, the weighting of various components does vary, primarily depending on the component's function, location, visibility, and its impact on overall safety and performance. In one specific embodiment of the present invention, the weighting of key safety components such as bumpers and doors in the injection molding quality assessment of automotive plastic injection molded parts is 1, the weighting of exterior and functional components such as instrument panels and headlights is 0.8, the weighting of interior and decorative components such as seats and carpets is 0.5, and the weighting of other components in the injection molding quality assessment of automotive plastic injection molded parts is 0.2.

[0068] C3. Calculate the injection quality coefficient of automotive plastic injection molded parts , ,in, Indicates the number of target sites.

[0069] The embodiment of the present invention considers the different importance of the injection molding quality of various parts in the comprehensive quality assessment when comprehensively analyzing the injection molding quality of automobile plastic injection molding parts. Taking this difference into consideration, defects in minor parts are avoided from being treated the same as defects in major parts, thereby avoiding misleading judgments on the overall quality. The injection molding quality of certain key parts directly affects the overall performance of the injection molding parts. Sufficient attention is paid to these key parts to avoid ignoring potential quality problems, thereby improving the reliability and safety of the product.

[0070] The database is used to store the standard positions corresponding to each monitoring point when the automotive plastic injection molded parts are placed on the conveyor belt, and to store the weight of each target location in the automotive plastic injection molded parts in the injection quality assessment of the automotive plastic injection molded parts. The data sources in the database of this embodiment are shown in Table 1.

[0071] Table 1 Data sources in the database

[0072]

[0073] The injection molded parts quality sorting module is used to perform quality sorting on automobile plastic injection molded parts by robot B in a target production line.

[0074] In a specific embodiment of the present invention, the method of quality sorting of automobile plastic injection molded parts by the robot B in the target production line is: comparing the injection quality coefficient of the automobile plastic injection molded parts with the set reference injection quality coefficient; if the injection quality coefficient of the automobile plastic injection molded parts is greater than or equal to the set reference injection quality coefficient, the automobile plastic injection molded parts are recorded as qualified injection molded parts; otherwise, the automobile plastic injection molded parts are recorded as unqualified injection molded parts; the robot B in the target production line picks up the qualified injection molded parts and places them on the conveyor belt corresponding to the qualified injection molded parts, and picks up the unqualified injection molded parts and places them on the conveyor belt corresponding to the unqualified injection molded parts.

[0075] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. An automatic inspection system for automobile plastic injection molded parts based on dual robot technology, characterized in that: include: The injection molded parts placement position assessment module is used to construct a three-dimensional model of the conveyor belt in the target production line. When the robot A in the target production line places the processed automobile plastic injection molded parts on the conveyor belt, the three-dimensional coordinate values ​​corresponding to each monitoring point in the automobile plastic injection molded parts are obtained to assess whether the position of the automobile plastic injection molded parts is accurately placed. If accurate, the injection molded parts image quality assessment module is executed. If inaccurate, the position of the automobile plastic injection molded parts is adjusted; The injection molded parts image quality assessment module is used to collect images corresponding to various parts of the automobile plastic injection molded parts through a track-type detector placed on the inner wall of the testing equipment when the automobile plastic injection molded parts are transported to the testing equipment through a conveyor belt, and evaluate whether the image quality corresponding to various parts of the automobile plastic injection molded parts meets the requirements. If yes, the injection molded parts injection quality analysis module is executed. If not, the image collection is performed again for the parts whose image quality does not meet the requirements; The specific process of evaluating whether the image quality corresponding to each part of the automobile plastic injection molded part meets the requirements is as follows: B1. Locate the pixel value of each pixel point from the image corresponding to each part of the automobile plastic injection molded part, and record the pixel value of each pixel point corresponding to each part of the automobile plastic injection molded part as ,in, Indicates the number of the part. , Indicates the number of the pixel. ; B2. Calculate the variance of pixel values ​​corresponding to each part of the automobile plastic injection molded part , ,in, Indicates the number of pixels; B3. Calculate the image quality coefficients of various parts of automobile plastic injection molded parts , ,in, Indicates the pixel value variance of the set reference; B4. Compare the image quality coefficient corresponding to each part of the automobile plastic injection molded part with the set reference image quality coefficient. If the image quality coefficient corresponding to a part of the automobile plastic injection molded part is less than the set reference image quality coefficient, it indicates that the image quality corresponding to the part of the automobile plastic injection molded part does not meet the requirements. Otherwise, it indicates that the image quality corresponding to the part of the automobile plastic injection molded part meets the requirements. Therefore, it is determined whether the image quality corresponding to each part of the automobile plastic injection molded part meets the requirements. The injection molding quality analysis module is used to record the parts whose image quality meets the requirements as target parts, extract the injection molding quality information from the images corresponding to the target parts in the automobile plastic injection molding parts, and analyze the injection molding quality coefficient of the automobile plastic injection molding parts; A database for storing the standard positions of each monitoring point when the automobile plastic injection molded parts are placed on the conveyor belt, and storing the weight of each target part in the automobile plastic injection molded parts in the injection molding quality assessment of the automobile plastic injection molded parts; The injection molded parts quality sorting module is used to perform quality sorting on automobile plastic injection molded parts by robot B in the target production line.

2. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 1 is characterized in that: The specific method of constructing the three-dimensional model of the conveyor belt in the target production line is: taking the plane of the conveyor belt in the target production line as The plane is perpendicular to the upward direction of the conveyor belt. A three-dimensional model of the conveyor belt in the target production line is constructed in the axial direction.

3. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 2 is characterized in that: The specific method for evaluating whether the position of the automobile plastic injection molding part is accurately placed is: A1. Record the three-dimensional coordinate values ​​corresponding to each monitoring point in the automobile plastic injection molded part as ,in, Indicates the number of the monitoring point. ; A2. Extract the standard positions of each monitoring point when the automobile plastic injection molded parts are placed on the conveyor belt from the database, and bring them into the three-dimensional model of the conveyor belt in the target production line to obtain the standard three-dimensional coordinate values ​​of each monitoring point when the automobile plastic injection molded parts are placed on the conveyor belt, which are recorded as ; A3. Calculate the position deviation index of each monitoring point in the automobile plastic injection molding part , ,in, and Respectively indicate the setting permission axis, Axis and The position offset value corresponding to the axis; A4. Calculate the position offset index of automotive plastic injection molded parts , ,in, Indicates the number of monitoring points; A5. Compare the position offset index corresponding to the automobile plastic injection molding part with the set permissible position offset index. If the position offset index corresponding to the automobile plastic injection molding part is less than the set permissible position offset index, it indicates that the position of the automobile plastic injection molding part is accurately placed. Otherwise, it indicates that the position of the automobile plastic injection molding part is inaccurate.

4. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 3 is characterized in that: The method for adjusting the position of the automobile plastic injection molded parts is as follows: when the position of the automobile plastic injection molded parts is not accurately placed, a position adjustment instruction for the automobile plastic injection molded parts is sent to robot A; when robot A receives the position adjustment instruction for the automobile plastic injection molded parts, the automobile plastic injection molded parts are clamped from the conveyor belt and repositioned; and the position of the automobile plastic injection molded parts after position adjustment is evaluated in the same way as that for evaluating whether the position of the automobile plastic injection molded parts is accurately placed; if the position is accurately placed, the position adjustment of the automobile plastic injection molded parts is terminated; if the position is inaccurate, the automobile plastic injection molded parts are adjusted again in the above manner, and so on, until the position of the automobile plastic injection molded parts is accurately placed.

5. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 1 is characterized in that: The specific method of re-capturing the image of the part whose image quality does not meet the requirements is: recording the part whose image quality does not meet the requirements as an abnormal part, sending each abnormal part in the automobile plastic injection molded part to the track-type detector on the inner wall of the detection equipment, and after receiving the corresponding instruction, the track-type detector re-captures the image of each abnormal part in the automobile plastic injection molded part, and evaluates whether the image quality corresponding to each abnormal part in the automobile plastic injection molded part meets the requirements in the same way as the evaluation method of whether the image quality corresponding to each part in the automobile plastic injection molded part meets the requirements. If it meets the requirements, the image acquisition is terminated, and if it does not meet the requirements, and so on, the image acquisition is re-captured for the abnormal part whose image quality does not meet the requirements.

6. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 1 is characterized in that: The injection molded part quality information includes the number of internal pores, the volume of each internal pore, the number of internal cracks, the length of each internal crack, the number of surface scratches, and the length of each surface scratch.

7. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 6 is characterized in that: The specific process of analyzing the injection molding quality coefficient of automobile plastic injection molding parts is as follows: C1. Extract the number of internal pores, the volume of each internal pore, the number of internal cracks, the length of each internal crack, the number of surface scratches and the length of each surface scratch from the quality information of the injection molded parts corresponding to each target part of the automobile plastic injection molded parts, and calculate the injection molding quality coefficient corresponding to each target part of the automobile plastic injection molded parts. ,in, Indicates the number of the target part, ; C2. Extract the weight of each target part in the injection quality assessment of automobile plastic injection molded parts from the database and record it as ; C3. Calculation of injection molding quality coefficient of automotive plastic injection molding parts , ,in, Indicates the number of target sites.

8. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 7 is characterized in that: The specific process of calculating the injection molding quality coefficient corresponding to each target part in the automobile plastic injection molding part is as follows: D1. The number of internal pores corresponding to each target part in the automobile plastic injection molded part is recorded as ; D2. Accumulate the internal pore volumes corresponding to each target part in the automobile plastic injection molded part to obtain the total internal pore volume corresponding to each target part in the automobile plastic injection molded part, and record it as ; D3. Calculate the internal pore abnormality index corresponding to each target part in the automotive plastic injection molded parts , ,in, and Respectively represent the number of internal pores and the volume of internal pores for setting reference, represents a natural constant; D4. Based on the number of internal cracks, the length of each internal crack, the number of surface scratches and the length of each surface scratch corresponding to each target part in the automobile plastic injection molding, calculate the internal crack abnormality index corresponding to each target part in the automobile plastic injection molding. and surface scratch anomaly index ; D5. Calculate the injection molding quality coefficient corresponding to each target part in the automotive plastic injection molding part , ,in, , and They represent the weights of the set internal pore abnormality index, internal crack abnormality index and surface scratch abnormality index corresponding to the injection quality assessment. .

9. The automatic detection system for automobile plastic injection molded parts based on dual robot technology according to claim 7 is characterized in that: The method of quality sorting of automobile plastic injection molded parts by robot B in the target production line is as follows: comparing the injection molding quality coefficient of the automobile plastic injection molded parts with the injection molding quality coefficient of the set reference; if the injection molding quality coefficient of the automobile plastic injection molded parts is greater than or equal to the injection molding quality coefficient of the set reference, the automobile plastic injection molded parts are recorded as qualified injection molded parts; otherwise, the automobile plastic injection molded parts are recorded as unqualified injection molded parts; the robot B in the target production line clamps the qualified injection molded parts and places them on the conveyor belt corresponding to the qualified injection molded parts, and clamps the unqualified injection molded parts and places them on the conveyor belt corresponding to the unqualified injection molded parts.

Images