Method and system for detecting surface defects of automotive parts

Abstract

The application discloses a kind of surface defect detection method and system of automobile parts, based on the three-dimensional model of automobile and the model definition of automobile corresponding detection mode, and along the detection mode to the three-dimensional model of automobile is carried out abnormal detection, in order to be directed at the three-dimensional model of automobile is targeted detection.Further, according to the component category of automobile parts and the abnormal category of abnormal component definition corresponding surface defect detection mode, based on the surface defect detection mode to corresponding automobile parts is carried out surface defect detection, according to surface defect image definition surface defect element, based on the multiple surface defect elements of different automobile parts definition overall defect grade of automobile, multiple dimensions control of multiple surface defect elements of different automobile parts is realized, based on the dimension of automobile parts and the dimension of surface defect element overall control is carried out, improve the surface defect detection precision of automobile parts, and then improve the accuracy of overall defect grade of automobile.

Description

Technical Field

[0001] This invention relates to the technical field of automotive parts, and more particularly to a method and system for detecting surface defects in automotive parts. Background Technology

[0002] With the development of technology, automobiles are widely used in people's lives. During the production process, automobiles need to undergo surface defect detection. In the existing technology, the model number of the automobile is collected, and the corresponding surface defect detection method is matched according to the model number. The overall inspection is carried out along the surface defect detection method. However, automobiles contain multiple automobile parts, which are distributed in different locations in the automobile. The existing surface defect detection method only controls the dimension of the model number, which affects the accuracy of surface defect detection of automobile parts, and thus affects the accuracy of the overall defect level of the automobile. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a method and system for detecting surface defects in automotive components. The method involves collecting the location and model of the vehicle, defining a traversal pattern based on the location and model, triggering the traversal of the vehicle along this pattern, and constructing a 3D model of the vehicle. Based on the 3D model and model, a corresponding detection method is defined, and anomaly detection is performed on the 3D model along this method. This facilitates targeted detection of the 3D model and fully considers the various automotive components within the 3D model.

[0004] Furthermore, based on the anomaly detection of the 3D model of the car, each abnormal part is defined; based on each abnormal part, its corresponding location, and the 3D model of the car, corresponding abnormal components are defined, and based on the abnormal components, the corresponding types of car components are defined; based on the component types of car components and the anomaly types of abnormal components, corresponding surface defect detection methods are defined, and surface defect detection is performed on the corresponding car components based on these surface defect detection methods; surface defect elements are defined based on the surface defect images; and based on multiple surface defect elements of different car components, the overall defect level of the car is defined. This achieves multi-dimensional control of multiple surface defect elements of different car components, and overall control is performed based on the dimensions of car components and the dimensions of surface defect elements, improving the surface defect detection accuracy of car components, and thus improving the accuracy of the overall defect level of the car.

[0005] This invention provides a method for detecting surface defects in automotive components, applicable to surface defect detection scenarios in automotive components;

[0006] The method for detecting surface defects in automotive components includes:

[0007] Collect the location and model of the car, and define the traversal mode based on the car's location and model.

[0008] Trigger the traversal of the car along this traversal pattern and build a 3D model of the car;

[0009] Based on the 3D model of the car and the corresponding detection method for the car model, anomaly detection is performed on the 3D model of the car along the detection method.

[0010] Each abnormal part is defined based on the anomaly detection of the 3D model of the car;

[0011] Based on each abnormal part, its corresponding location, and the 3D model of the car, define the corresponding abnormal components, and based on the abnormal components, define the corresponding types of car components.

[0012] The surface defect detection method is defined according to the type of automotive component and the type of abnormality of the abnormal component. The surface defect detection method is used to detect surface defects in the corresponding automotive component. The surface defect elements are defined according to the surface defect image. The overall defect level of the car is defined based on multiple surface defect elements of different automotive components.

[0013] Optionally, the step of collecting the location and model of the vehicle, and defining a traversal mode based on the vehicle's location and model, includes:

[0014] Locate the vehicle to be inspected and collect its location;

[0015] The vehicle's location is determined by detection, and its attitude is defined.

[0016] The first parameter is defined based on the vehicle's location and its attitude;

[0017] Collect the car model and associate it with the car's location. Define a second parameter based on the car model and the car's location.

[0018] Associate the first parameter and the second parameter, and define the corresponding traversal pattern based on the first parameter, the second parameter and the traversal matching table.

[0019] Optionally, the step of triggering the traversal of the car along the traversal pattern and constructing a 3D model of the car includes:

[0020] Freeze the traversal pattern, and trigger the traversal of the car along the traversal pattern;

[0021] During the traversal, dynamic shots were taken of the car from different positions.

[0022] Multiple car images from different directions are captured based on dynamic photography of the vehicle.

[0023] Multiple stereo features are defined based on multiple car images from different directions, and these stereo features are distributed in different directions of the car.

[0024] A three-dimensional model of a car is constructed based on multiple three-dimensional features and the car model. At this point, a preliminary three-dimensional model is defined based on the synthesis of multiple three-dimensional features, and a three-dimensional model of the car is constructed based on the preliminary three-dimensional model and the reference model corresponding to the car model.

[0025] Optionally, the detection method based on the 3D model of the car and the car model definition, and the anomaly detection of the 3D model of the car along the detection method, includes:

[0026] A freeze-frame 3D model of a car;

[0027] Detection coefficients based on the 3D model of the car and the model definition of the car;

[0028] Define the corresponding detection method based on the detection coefficient and the corresponding detection method matching table;

[0029] This detection method is now fixed.

[0030] Associate the 3D model of the car with the detection method, and define the corresponding matching coefficients based on the 3D model of the car and the detection method;

[0031] The matching coefficient is compared with the preset matching coefficient threshold. At this time, the preset matching coefficient threshold corresponds to the corresponding car model.

[0032] If the matching coefficient is greater than the preset matching coefficient threshold, then anomaly detection is performed on the 3D model of the car using this detection method.

[0033] Optionally, the definition of each abnormal part based on the anomaly detection of the vehicle's 3D model includes:

[0034] A freeze-frame 3D model of a car;

[0035] Multiple sub-models are defined based on the division of the car's three-dimensional model, and these sub-models are distributed in different positions of the car.

[0036] Associate multiple sub-models and their corresponding detection methods;

[0037] Multiple sub-anomaly components are defined based on anomaly detection using multiple sub-models and corresponding detection methods;

[0038] In each sub-model, each sub-abnormal part constructs the corresponding abnormal part. At this point, there are multiple abnormal parts in the 3D model of the car.

[0039] Optionally, the definition of corresponding abnormal components based on each abnormal part, its corresponding location, and the three-dimensional model of the vehicle, and the definition of corresponding vehicle component types based on the abnormal components, include:

[0040] Freeze the frame on each abnormal part and mark the corresponding location;

[0041] Anomaly profiles are defined based on the identification of each anomaly component;

[0042] Define the corresponding abnormal region based on the abnormal outline and its location.

[0043] Optionally, the step of defining corresponding abnormal components based on each abnormal part, its corresponding location, and the three-dimensional model of the vehicle, and defining the types of corresponding vehicle components based on the abnormal components, further includes:

[0044] Based on the correlation between various abnormal areas and the 3D model of the car;

[0045] Define the corresponding abnormal components based on each abnormal area and the 3D model of the car;

[0046] The classification table of abnormal components and automobile structure defines the corresponding types of automobile components.

[0047] Optionally, the step of defining a corresponding surface defect detection method based on the type of automotive component and the type of abnormality of the abnormal component, performing surface defect detection on the corresponding automotive component based on the surface defect detection method, defining surface defect elements based on the surface defect image, and defining the overall defect level of the vehicle based on multiple surface defect elements of different automotive components includes:

[0048] The types of related automotive components and the types of abnormalities in abnormal components;

[0049] The surface defect detection method is defined according to the type of automotive component and the type of abnormality of the abnormal component. At this time, the corresponding surface defect detection method is matched based on different automotive components.

[0050] Optionally, the step of defining a corresponding surface defect detection method based on the type of automotive component and the type of abnormality of the abnormal component, performing surface defect detection on the corresponding automotive component based on the surface defect detection method, defining surface defect elements based on the surface defect image, and defining the overall defect level of the vehicle based on multiple surface defect elements of different automotive components, further includes:

[0051] This surface defect detection method is used to detect surface defects in corresponding automotive parts.

[0052] Collect corresponding surface defect images based on the surface defect detection of automotive parts;

[0053] Define the corresponding surface defect elements based on each surface defect image;

[0054] Multiple surface defect elements are distributed in the three-dimensional model of the car, and different defect levels of the car parts are defined based on the multiple surface defect elements of different car parts. A defect level set is constructed according to the defect level of each car part, and the overall defect level of the car is defined based on the defect level set.

[0055] In addition, embodiments of the present invention also provide a surface defect detection system for automotive components, the surface defect detection system for automotive components comprising:

[0056] The data acquisition module is used to collect the location and model of the car, and defines the traversal mode based on the car's location and model.

[0057] The 3D model module is used to trigger the traversal of the car along this traversal pattern and build a 3D model of the car.

[0058] The anomaly detection module is used to perform anomaly detection on the 3D model of the car and the corresponding detection method based on the car model, and to perform anomaly detection on the 3D model of the car along the detection method.

[0059] The anomaly module is used to define various anomaly parts based on anomaly detection of the 3D model of the car.

[0060] The automotive parts module is used to define corresponding abnormal parts based on each abnormal part, its corresponding location, and the 3D model of the car, and to define the types of automotive parts based on the abnormal parts.

[0061] The defect module is used to define corresponding surface defect detection methods based on the type of automotive parts and the type of abnormality of abnormal parts. Based on the surface defect detection method, surface defect detection is performed on the corresponding automotive parts. Surface defect elements are defined based on the surface defect images. The overall defect level of the vehicle is defined based on multiple surface defect elements of different automotive parts.

[0062] In this embodiment of the invention, the location and model of the car are collected using the method described in this embodiment. A traversal pattern is defined based on the location and model of the car. The traversal of the car is triggered along the traversal pattern, and a three-dimensional model of the car is constructed. A corresponding detection method is defined based on the three-dimensional model of the car and the model of the car. Anomaly detection is performed on the three-dimensional model of the car along the detection method, so as to facilitate targeted detection of the three-dimensional model of the car and fully consider the various different car parts in the three-dimensional model of the car.

[0063] Furthermore, based on the anomaly detection of the 3D model of the car, each abnormal part is defined; based on each abnormal part, its corresponding location, and the 3D model of the car, corresponding abnormal components are defined, and based on the abnormal components, the corresponding types of car components are defined; based on the component types of car components and the anomaly types of abnormal components, corresponding surface defect detection methods are defined, and surface defect detection is performed on the corresponding car components based on these surface defect detection methods; surface defect elements are defined based on the surface defect images; and based on multiple surface defect elements of different car components, the overall defect level of the car is defined. This achieves multi-dimensional control of multiple surface defect elements of different car components, and overall control is performed based on the dimensions of car components and the dimensions of surface defect elements, improving the surface defect detection accuracy of car components, and thus improving the accuracy of the overall defect level of the car. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a schematic flowchart of the surface defect detection method for automotive parts in an embodiment of the present invention;

[0066] Figure 2 This is a schematic flowchart of step S11 in the surface defect detection method for automotive parts according to an embodiment of the present invention.

[0067] Figure 3 This is a schematic flowchart of step S12 in the surface defect detection method for automotive parts in an embodiment of the present invention;

[0068] Figure 4 This is a flowchart illustrating step S13 of the surface defect detection method for automotive parts in an embodiment of the present invention.

[0069] Figure 5 This is a flowchart illustrating step S14 of the surface defect detection method for automotive parts in an embodiment of the present invention.

[0070] Figure 6 This is a flowchart illustrating step S15 of the surface defect detection method for automotive parts in an embodiment of the present invention.

[0071] Figure 7 This is a flowchart illustrating step S16 of the surface defect detection method for automotive parts in an embodiment of the present invention.

[0072] Figure 8This is a schematic diagram of the structural composition of the surface defect detection system for automotive components in an embodiment of the present invention;

[0073] Figure 9 This is a hardware diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0074] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] Please see Figures 1 to 9 A method for detecting surface defects in automotive components, applied to surface defect detection scenarios for automotive components; the method for detecting surface defects in automotive components includes:

[0076] Step S11: Collect the location and model of the car, and define the traversal mode based on the location and model of the car;

[0077] Step S12: Trigger the traversal of the car along this traversal pattern and construct the 3D model of the car;

[0078] Step S13: Based on the 3D model of the car and the model definition of the car, perform anomaly detection on the 3D model of the car according to the detection method;

[0079] Step S14: Define each abnormal part based on the anomaly detection of the 3D model of the car;

[0080] Step S15: Define the corresponding abnormal parts based on each abnormal part, its corresponding location, and the 3D model of the car; define the corresponding types of car parts based on the abnormal parts.

[0081] Step S16: Define the corresponding surface defect detection method according to the component type of the automobile component and the abnormality type of the abnormal component. Perform surface defect detection on the corresponding automobile component based on the surface defect detection method. Define surface defect elements according to the surface defect image. Define the overall defect level of the automobile based on multiple surface defect elements of different automobile components.

[0082] In this embodiment of the invention, the location and model of the car are collected using the method described in this embodiment. A traversal pattern is defined based on the location and model of the car. The traversal of the car is triggered along the traversal pattern, and a three-dimensional model of the car is constructed. A corresponding detection method is defined based on the three-dimensional model of the car and the model of the car. Anomaly detection is performed on the three-dimensional model of the car along the detection method, so as to facilitate targeted detection of the three-dimensional model of the car and fully consider the various different car parts in the three-dimensional model of the car.

[0083] Furthermore, based on the anomaly detection of the 3D model of the car, each abnormal part is defined; based on each abnormal part, its corresponding location, and the 3D model of the car, corresponding abnormal components are defined, and based on the abnormal components, the corresponding types of car components are defined; based on the component types of car components and the anomaly types of abnormal components, corresponding surface defect detection methods are defined, and surface defect detection is performed on the corresponding car components based on these surface defect detection methods; surface defect elements are defined based on the surface defect images; and based on multiple surface defect elements of different car components, the overall defect level of the car is defined. This achieves multi-dimensional control of multiple surface defect elements of different car components, and overall control is performed based on the dimensions of car components and the dimensions of surface defect elements, improving the surface defect detection accuracy of car components, and thus improving the accuracy of the overall defect level of the car.

[0084] refer to Figure 2 In step S11, the location of the car and the model of the car are collected, and the traversal mode is defined according to the location of the car and the model of the car.

[0085] In the specific implementation of this invention, the specific steps can be as follows:

[0086] S111: Locate the vehicle to be detected and collect its location;

[0087] S112: Perform localization detection based on the vehicle's location and define the vehicle's attitude;

[0088] S113: Define the first parameter based on the vehicle's location and attitude;

[0089] S114: Collect the car model and associate it with the car model and the car's location. Define the second parameter based on the car model and the car's location.

[0090] S115: Associate the first parameter and the second parameter, and define the corresponding traversal pattern based on the first parameter, the second parameter and the traversal matching table.

[0091] In the embodiments of this application, the vehicle to be detected is located and its location is collected. Further steps are taken based on the location of the vehicle to perform localization detection and define the vehicle's attitude, thus introducing the vehicle's location and corresponding attitude.

[0092] Furthermore, the location of the car and its corresponding attitude are associated, and a first parameter is defined based on the location and attitude of the car, which takes into account the overall location and attitude of the car and ensures the accuracy of the first parameter.

[0093] At the same time, the model of the car is collected and associated with the car model and its location. A second parameter is defined based on the car model and its location, which takes into account the overall consideration of the car model and its location, and ensures the accuracy of the second parameter.

[0094] Therefore, by associating the first parameter and the second parameter, and defining the corresponding traversal mode based on the first parameter, the second parameter, and the traversal matching table, the first parameter, the second parameter, and the traversal matching table are introduced, realizing multi-dimensional control of the first parameter, the second parameter, and the traversal matching table, ensuring the accuracy of the traversal mode. Optionally, the traversal mode includes global traversal or partial traversal.

[0095] refer to Figure 3 In step S12, the traversal of the car is triggered along the traversal pattern, and a three-dimensional model of the car is constructed.

[0096] In the specific implementation of this invention, the specific steps can be as follows:

[0097] S121: Freeze the traversal pattern and trigger the traversal of the car along the traversal pattern;

[0098] S122: During the traversal, dynamic shots are taken of the car from different positions.

[0099] S123: Capture multiple car images from different directions based on dynamic car photography;

[0100] S124: Define multiple stereo features based on multiple car images from different directions, with the multiple stereo features distributed in different directions of the car;

[0101] S125: Construct a three-dimensional model of a car based on multiple three-dimensional features and the car model. At this point, a preliminary three-dimensional model is defined based on the synthesis of multiple three-dimensional features, and a three-dimensional model of the car is constructed based on the preliminary three-dimensional model and the reference model corresponding to the car model.

[0102] In the embodiments of this application, the traversal pattern is fixed, and the traversal of the car is triggered along the traversal pattern to realize the traversal of the car. At this time, during the traversal process, corresponding dynamic shooting is performed on different positions of the car, thereby acquiring multiple car images in different directions based on the dynamic shooting of the car, realizing the dynamic shooting of the car, ensuring the correspondence of multiple car images in different directions, so as to facilitate further processing of multiple car images.

[0103] Therefore, multiple stereo features are defined based on multiple car images from different directions, and these features are distributed across different directions of the car. A stereo model of the car is constructed based on these features and the car model. At this point, a preliminary stereo model is defined based on the synthesis of these features, and a stereo model of the car is constructed based on the preliminary stereo model and the reference model corresponding to the car model. Further control is applied to the preliminary stereo model, taking into account both the preliminary stereo model and the reference model corresponding to the car model, thus ensuring the accuracy of the stereo model of the car.

[0104] refer to Figure 4 In step S13, based on the 3D model of the car and the model of the car, the corresponding detection method is defined, and anomaly detection is performed on the 3D model of the car along the detection method.

[0105] In the specific implementation of this invention, the specific steps can be as follows:

[0106] S131: A three-dimensional model of a car in still life;

[0107] S132: Detection coefficients based on the 3D model of the vehicle and the vehicle model definition;

[0108] S133: Define the corresponding detection method based on the detection coefficient and the corresponding detection method matching table;

[0109] S134: Freeze the detection method; associate the 3D model of the car with the detection method, and define the corresponding matching coefficients based on the 3D model of the car and the detection method;

[0110] S135: Compare the matching coefficient with the preset matching coefficient threshold. At this time, the preset matching coefficient threshold corresponds to the corresponding car model.

[0111] S136: If the matching coefficient is greater than the preset matching coefficient threshold, then perform anomaly detection on the 3D model of the car along this detection method.

[0112] In the embodiments of this application, the location and model of the car are collected, and a traversal pattern is defined based on the location and model of the car. The traversal of the car is triggered along the traversal pattern, and a three-dimensional model of the car is constructed. Based on the three-dimensional model of the car and the model of the car, a corresponding detection method is defined, and anomaly detection is performed on the three-dimensional model of the car along the detection method, so as to facilitate targeted detection of the three-dimensional model of the car and fully consider the various different car parts in the three-dimensional model of the car.

[0113] At this point, the 3D model of the car is frozen, and further control is applied to the 3D model of the car. The 3D model of the car and the car model are associated, and detection coefficients are defined based on the 3D model of the car and the car model. Multi-dimensional control is applied to the 3D model of the car and the car model to ensure the accuracy of the detection coefficients, and the detection coefficients are further processed.

[0114] Furthermore, based on the detection coefficient and the corresponding detection method matching table, the corresponding detection method is defined, and the detection coefficient and the corresponding detection method matching table are matched to ensure the accuracy of the detection method.

[0115] Therefore, this detection method is defined; the 3D model of the car is associated with this detection method, and a corresponding matching coefficient is defined based on the 3D model of the car and the detection method. The matching coefficient is further controlled, and at the same time, the matching coefficient is compared with a preset matching coefficient threshold. At this time, the preset matching coefficient threshold corresponds to the corresponding car model. If the matching coefficient is greater than the preset matching coefficient threshold, anomaly detection is performed on the 3D model of the car according to this detection method, which ensures the targeted detection of the 3D model and the accuracy of anomaly detection of the 3D model of the car.

[0116] refer to Figure 5 S14: Define each abnormal part based on the anomaly detection of the 3D model of the car;

[0117] In the specific implementation of this invention, the specific steps can be as follows:

[0118] S141: A three-dimensional model of a car in still life;

[0119] S142: Define multiple sub-models based on the division of the three-dimensional model of the car, with the multiple sub-models distributed in different positions of the car;

[0120] S143: Associate multiple sub-models and their corresponding detection methods;

[0121] S144: Define multiple sub-anomaly parts based on anomaly detection using multiple sub-models and corresponding detection methods;

[0122] S145: In each sub-model, each sub-abnormal part constructs a corresponding abnormal part. At this time, there are multiple abnormal parts in the 3D model of the car.

[0123] In the embodiments of this application, a three-dimensional model of a car is fixed, and the three-dimensional model of the car is further processed. At this time, multiple sub-models are defined according to the division of the three-dimensional model of the car. The multiple sub-models are distributed in different positions of the car. Multiple sub-models and corresponding detection methods are introduced, thereby associating multiple sub-models and corresponding detection methods to ensure targeted control of multiple sub-models and corresponding detection methods.

[0124] Therefore, multiple sub-anomaly parts are defined based on anomaly detection of multiple sub-models and corresponding detection methods. Anomaly detection of multiple sub-models and corresponding detection methods is controlled in multiple dimensions, thereby introducing multiple sub-anomaly parts and realizing targeted control of multiple sub-anomaly parts. At the same time, in each sub-model, each sub-anomaly part constructs a corresponding anomaly part. In the 3D model of the car, there are multiple anomaly parts, and multiple anomaly parts are controlled in multiple dimensions.

[0125] refer to Figure 6 S15: Define the corresponding abnormal parts based on each abnormal part, its corresponding location, and the three-dimensional model of the car; define the types of corresponding car parts based on the abnormal parts.

[0126] In the specific implementation of this invention, the specific steps can be as follows:

[0127] S151: Freeze each abnormal part and mark the corresponding position;

[0128] S152: Define the anomaly profile based on the identification of each anomaly component;

[0129] S153: Define the corresponding abnormal region based on the abnormal contour and its corresponding location;

[0130] S154: Correlate the various abnormal areas and the 3D model of the car;

[0131] S155: Define the corresponding abnormal components based on each abnormal area and the 3D model of the vehicle;

[0132] S156: Define the types of automotive parts based on the abnormal components and the automotive structure classification table.

[0133] In the embodiments of this application, each abnormal part is fixed and its corresponding position is marked, thereby introducing each abnormal part and its corresponding position, and thus defining an abnormal contour based on the identification of each abnormal part, and introducing the corresponding abnormal contour.

[0134] Furthermore, based on the abnormal contours and their corresponding locations, corresponding abnormal regions are defined, and overall control is exercised over the abnormal contours and their corresponding locations. At the same time, each abnormal region is associated with the 3D model of the vehicle; corresponding abnormal components are defined based on each abnormal region and the 3D model of the vehicle; and the types of corresponding vehicle components are defined based on the abnormal components and the vehicle's structural classification table, ensuring the accuracy of the types of vehicle components and realizing multi-dimensional control over each abnormal region, the 3D model of the vehicle, and the vehicle's structural classification table.

[0135] refer to Figure 7 S16: Define the corresponding surface defect detection method according to the component type of the automobile component and the abnormality type of the abnormal component, perform surface defect detection on the corresponding automobile component based on the surface defect detection method, define surface defect elements according to the surface defect image, and define the overall defect level of the automobile based on multiple surface defect elements of different automobile components.

[0136] In the specific implementation of this invention, the specific steps can be as follows:

[0137] S161: Types of related automotive components and types of abnormalities in abnormal components;

[0138] S162: Define the corresponding surface defect detection method according to the type of automotive component and the type of abnormality of the abnormal component. At this time, the corresponding surface defect detection method is matched based on different automotive components.

[0139] S163: Surface defect detection is performed on the corresponding automotive parts based on this surface defect detection method;

[0140] S164: Acquire corresponding surface defect images based on the surface defect detection of automotive parts;

[0141] S165: Define the corresponding surface defect elements based on each surface defect image;

[0142] S166: Distribute multiple surface defect elements in the three-dimensional model of the car, and define different defect levels of different car parts based on multiple surface defect elements of different car parts. Construct a defect level set according to the defect level of each car part, and define the overall defect level of the car based on the defect level set.

[0143] In the specific implementation of this invention, various abnormal parts are defined based on the anomaly detection of the three-dimensional model of the car; corresponding abnormal components are defined based on each abnormal part, its corresponding position, and the three-dimensional model of the car; the types of corresponding car components are defined based on the abnormal components; a corresponding surface defect detection method is defined based on the component type of the car component and the anomaly type of the abnormal component; surface defect detection is performed on the corresponding car component based on the surface defect detection method; surface defect elements are defined based on the surface defect image; and the overall defect level of the car is defined based on multiple surface defect elements of different car components. This achieves multi-dimensional control of multiple surface defect elements of different car components, and overall control is performed based on the dimensions of the car component and the dimensions of the surface defect elements, thereby improving the surface defect detection accuracy of the car component and thus improving the accuracy of the overall defect level of the car.

[0144] At this point, the types of related automotive components and the types of abnormal components are comprehensively controlled. At the same time, corresponding surface defect detection methods are defined according to the types of automotive components and the types of abnormal components. Then, the corresponding surface defect detection methods are matched based on different automotive components to facilitate targeted detection for different automotive components and ensure the matching of surface defect detection methods.

[0145] Furthermore, based on this surface defect detection method, surface defects are detected on the corresponding automotive parts, thereby acquiring corresponding surface defect images based on the surface defect detection of the automotive parts, introducing various surface defect images, and further identifying the surface defect images.

[0146] Therefore, based on each surface defect image, corresponding surface defect elements are defined; multiple surface defect elements are distributed in the three-dimensional model of the car, and different defect levels of different car parts are defined based on multiple surface defect elements of different car parts. A defect level set is constructed according to the defect level of each car part, and the overall defect level of the car is defined based on the defect level set. This achieves multi-dimensional control of multiple surface defect elements of different car parts. Overall control is carried out based on the dimensions of car parts and surface defect elements, which improves the surface defect detection accuracy of car parts, and thus improves the accuracy of the overall defect level of the car.

[0147] In this embodiment of the invention, the location and model of the car are collected using the method described in this embodiment. A traversal pattern is defined based on the location and model of the car. The traversal of the car is triggered along the traversal pattern, and a three-dimensional model of the car is constructed. A corresponding detection method is defined based on the three-dimensional model of the car and the model of the car. Anomaly detection is performed on the three-dimensional model of the car along the detection method, so as to facilitate targeted detection of the three-dimensional model of the car and fully consider the various different car parts in the three-dimensional model of the car.

[0148] Furthermore, based on the anomaly detection of the 3D model of the car, each abnormal part is defined; based on each abnormal part, its corresponding location, and the 3D model of the car, corresponding abnormal components are defined, and based on the abnormal components, the corresponding types of car components are defined; based on the component types of car components and the anomaly types of abnormal components, corresponding surface defect detection methods are defined, and surface defect detection is performed on the corresponding car components based on these surface defect detection methods; surface defect elements are defined based on the surface defect images; and based on multiple surface defect elements of different car components, the overall defect level of the car is defined. This achieves multi-dimensional control of multiple surface defect elements of different car components, and overall control is performed based on the dimensions of car components and the dimensions of surface defect elements, improving the surface defect detection accuracy of car components, and thus improving the accuracy of the overall defect level of the car.

[0149] Please see Figure 8 , Figure 8 This is a schematic diagram of the structural composition of the surface defect detection system for automotive components in an embodiment of the present invention.

[0150] like Figure 8 As shown, a surface defect detection system for automotive components includes:

[0151] The acquisition module 21 is used to acquire the location of the car and the model of the car, and to define the traversal mode based on the location of the car and the model of the car.

[0152] 3D model module 22 is used to trigger the traversal of the car along the traversal pattern and construct a 3D model of the car.

[0153] Anomaly detection module 23 is used to perform anomaly detection on the three-dimensional model of the car and the detection method corresponding to the model of the car, and to perform anomaly detection on the three-dimensional model of the car along the detection method.

[0154] Anomaly module 24 is used to define various anomaly parts based on anomaly detection of the 3D model of the car;

[0155] The vehicle component module 25 is used to define the corresponding abnormal components based on each abnormal part, its corresponding location, and the three-dimensional model of the vehicle, and to define the types of vehicle components based on the abnormal components.

[0156] The defect module 26 is used to define the corresponding surface defect detection method according to the component type of the automobile component and the abnormality type of the abnormal component, perform surface defect detection on the corresponding automobile component based on the surface defect detection method, define surface defect elements according to the surface defect image, and define the overall defect level of the automobile based on multiple surface defect elements of different automobile components.

[0157] Please see Figure 9 See below for reference. Figure 9 To describe an electronic intelligent device 40 according to this embodiment of the present invention. Figure 9 The electronic smart device 40 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0158] like Figure 9 As shown, the electronic intelligent device 40 is manifested in the form of a general-purpose computing intelligent device. The components of the electronic intelligent device 40 may include, but are not limited to: at least one processing unit 41, at least one storage unit 42, and a bus 43 connecting different system components (including storage unit 42 and processing unit 41).

[0159] The storage unit stores program code, which can be executed by the processing unit 41 to perform the steps described in the "Embodiment Methods" section of this specification according to various exemplary embodiments of the present invention.

[0160] Storage unit 42 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 421 and / or cache memory 422, and may further include a read-only memory (ROM) 423.

[0161] Storage unit 42 may also include a program / utility 424 having a set (at least one) of program modules 425, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0162] Bus 43 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0163] The electronic intelligent device 40 can also communicate with one or more external intelligent devices (e.g., keyboards, pointing intelligent devices, Bluetooth intelligent devices, etc.), and with one or more intelligent devices that enable users to interact with the electronic intelligent device 40, and / or with any intelligent device (e.g., routers, modems, etc.) that enables the electronic intelligent device 40 to communicate with one or more other computing intelligent devices. This communication can be performed through the input / output (I / O) interface 44. Furthermore, the electronic intelligent device 40 can also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through the network adapter 45. Figure 9 As shown, network adapter 45 communicates with other modules of electronic intelligent device 40 via bus 43. It should be understood that, although... Figure 9 As not shown, other hardware and / or software modules can be used in conjunction with the electronic intelligent device 40, including but not limited to: microcode, intelligent device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup planning systems.

[0164] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing intelligent device (such as a personal computer, server, terminal device, or network intelligent device, etc.) to execute the method according to the embodiments of this disclosure.

[0165] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc. Furthermore, it stores computer program instructions, which, when executed by a computer, cause the computer to perform the methods described above.

[0166] Furthermore, the surface defect detection method and system for automotive parts provided by the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for detecting surface defects in automotive parts, characterized in that, Applications include surface defect detection for automotive parts. The method for detecting surface defects in automotive components includes: Collect the location and model of the car, and define the traversal mode based on the car's location and model. The process involves triggering a traversal of the car along the traversal pattern and constructing a 3D model of the car, including: freezing the traversal pattern and triggering a traversal of the car along the traversal pattern; taking dynamic photos of the car from different positions during the traversal; acquiring multiple car images from different directions based on the dynamic photos of the car; defining multiple 3D features based on the multiple car images from different directions, with the multiple 3D features distributed in different directions of the car; constructing a 3D model of the car based on the multiple 3D features and the car model; at this point, defining a preliminary 3D model based on the synthesis of multiple 3D features, and constructing a 3D model of the car based on the preliminary 3D model and the reference model corresponding to the car model. Based on the 3D model of the car and the corresponding detection method for the car model, anomaly detection is performed on the 3D model of the car along the detection method. Each abnormal part is defined based on the anomaly detection of the 3D model of the car; Based on each abnormal part, its corresponding location, and the 3D model of the car, define the corresponding abnormal components, and based on the abnormal components, define the corresponding types of car components. The surface defect detection method is defined according to the type of automotive component and the type of abnormality of the abnormal component. The surface defect detection method is used to detect surface defects in the corresponding automotive component. The surface defect elements are defined according to the surface defect image. The overall defect level of the car is defined based on multiple surface defect elements of different automotive components.

2. The method for detecting surface defects in automotive parts according to claim 1, characterized in that, The process involves collecting the location and model of the vehicle, and defining a traversal mode based on the vehicle's location and model, including: Locate the vehicle to be inspected and collect its location; The vehicle's location is determined by detection, and its attitude is defined. The first parameter is defined based on the vehicle's location and its attitude; Collect the car model and associate it with the car's location. Define a second parameter based on the car model and the car's location. Associate the first parameter and the second parameter, and define the corresponding traversal pattern based on the first parameter, the second parameter and the traversal matching table.

3. The method for detecting surface defects in automotive parts according to claim 2, characterized in that, The detection method based on the 3D model of the car and the car model definition, and the anomaly detection performed on the 3D model of the car along this detection method, includes: A freeze-frame 3D model of a car; Detection coefficients based on the 3D model of the car and the model definition of the car; Define the corresponding detection method based on the detection coefficient and the corresponding detection method matching table; The detection method is fixed; the 3D model of the car and the detection method are associated, and the corresponding matching coefficients are defined based on the 3D model of the car and the detection method; The matching coefficient is compared with the preset matching coefficient threshold. At this time, the preset matching coefficient threshold corresponds to the corresponding car model. If the matching coefficient is greater than the preset matching coefficient threshold, then anomaly detection is performed on the 3D model of the car using this detection method.

4. The method for detecting surface defects in automotive parts according to claim 3, characterized in that, The definition of various abnormal parts based on anomaly detection of the vehicle's 3D model includes: A freeze-frame 3D model of a car; Multiple sub-models are defined based on the division of the car's three-dimensional model, and these sub-models are distributed in different positions of the car. Associate multiple sub-models and their corresponding detection methods; Multiple sub-anomaly components are defined based on anomaly detection using multiple sub-models and corresponding detection methods; In each sub-model, each sub-abnormal part constructs the corresponding abnormal part. At this point, there are multiple abnormal parts in the 3D model of the car.

5. The method for detecting surface defects in automotive parts according to claim 4, characterized in that, The abnormal components are defined based on each abnormal part, its corresponding location, and the 3D model of the car. The types of car components are defined based on these abnormal components, including: Freeze the frame on each abnormal part and mark the corresponding location; Anomaly profiles are defined based on the identification of each anomaly component; Define the corresponding abnormal region based on the abnormal outline and its location.

6. The method for detecting surface defects in automotive parts according to claim 5, characterized in that, The definition of corresponding abnormal components based on each abnormal part, its corresponding location, and the three-dimensional model of the car, and the definition of the types of corresponding car components based on the abnormal components, also includes: Based on the correlation between various abnormal areas and the 3D model of the car; Define the corresponding abnormal components based on each abnormal area and the 3D model of the car; The classification table of abnormal components and automobile structure defines the corresponding types of automobile components.

7. The method for detecting surface defects in automotive parts according to claim 6, characterized in that, The method defines a corresponding surface defect detection method based on the type of automotive component and the type of defect in the abnormal component. Surface defect detection is then performed on the corresponding automotive component using this method. Surface defect elements are defined based on the surface defect image. The overall defect level of the vehicle is defined based on multiple surface defect elements of different automotive components, including: The types of related automotive components and the types of abnormalities in abnormal components; The surface defect detection method is defined according to the type of automotive component and the type of abnormality of the abnormal component. At this time, the corresponding surface defect detection method is matched based on different automotive components.

8. The method for detecting surface defects in automotive parts according to claim 7, characterized in that, The method of defining corresponding surface defect detection methods based on the type of automotive component and the type of abnormality of the abnormal component, performing surface defect detection on the corresponding automotive component based on the surface defect detection method, defining surface defect elements based on the surface defect image, and defining the overall defect level of the vehicle based on multiple surface defect elements of different automotive components, further includes: This surface defect detection method is used to detect surface defects in corresponding automotive parts. Collect corresponding surface defect images based on the surface defect detection of automotive parts; Define the corresponding surface defect elements based on each surface defect image; Multiple surface defect elements are distributed in the three-dimensional model of the car, and different defect levels of the car parts are defined based on the multiple surface defect elements of different car parts. A defect level set is constructed according to the defect level of each car part, and the overall defect level of the car is defined based on the defect level set.

9. A surface defect detection system for automotive parts, characterized in that, The surface defect detection system for automotive components is applied to the surface defect detection method for automotive components as described in any one of claims 1-8, and the surface defect detection system for automotive components comprises: The data acquisition module is used to collect the location and model of the car, and defines the traversal mode based on the car's location and model. The 3D model module is used to trigger the traversal of the car along the traversal pattern and construct a 3D model of the car. This includes: freezing the traversal pattern; triggering the traversal of the car along the traversal pattern; taking dynamic photos of the car from different positions during the traversal; acquiring multiple car images from different directions based on the dynamic photos; defining multiple 3D features based on the multiple car images from different directions, with these features distributed across different directions of the car; constructing a 3D model of the car based on the multiple 3D features and the car's model number. At this point, a preliminary 3D model is defined based on the synthesis of the multiple 3D features, and a 3D model of the car is constructed based on the preliminary 3D model and the reference model corresponding to the car's model number. The anomaly detection module is used to perform anomaly detection on the 3D model of the car and the corresponding detection method based on the car model, and to perform anomaly detection on the 3D model of the car along the detection method. The anomaly module is used to define various anomaly parts based on anomaly detection of the 3D model of the car. The automotive parts module is used to define corresponding abnormal parts based on each abnormal part, its corresponding location, and the 3D model of the car, and to define the types of automotive parts based on the abnormal parts. The defect module is used to define corresponding surface defect detection methods based on the type of automotive parts and the type of abnormality of abnormal parts. Based on the surface defect detection method, surface defect detection is performed on the corresponding automotive parts. Surface defect elements are defined based on the surface defect images. The overall defect level of the vehicle is defined based on multiple surface defect elements of different automotive parts.

Images