A real-time monitoring system and method for automobile spraying defects at the user end of the Internet of Things

Abstract

The invention relates to the technical field of automobile spraying defect detection, and discloses a user-end automobile spraying defect real-time monitoring system and method of the Internet of Things. The system comprises a vehicle body detection point construction module, a vehicle body detection point paint surface defect analysis module, and a vehicle body detection point paint surface defect monitoring and display module; the vehicle body paint surface defect type is identified by scientifically presetting the vehicle body paint surface grid detection surface image positioning parameters combined with an intelligent recognition algorithm and the vehicle body paint surface defect type image parameters, and the defect recognition result parameters of each paint surface detection surface sample of the vehicle body are accurately analyzed to achieve comprehensive and reliable monitoring of the vehicle body paint surface defects; the number of vehicle body paint surface detection surface samples with defects is autonomously screened and counted, and the vehicle body paint surface detection surface sample defect rate parameters are scientifically measured and the vehicle body paint surface defect detection evaluation parameters are analyzed to achieve digital statistics of vehicle body paint surface spraying defects and accurate evaluation and analysis of spraying quality.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile spraying defect detection, and in particular to a user-side automobile spraying defect real-time monitoring system and method for the Internet of Things. Background Art

[0002] Car spraying refers to coating the surface of a car with a layer of paint to protect the car. Car spraying generally uses baking paint. In the car manufacturing plant, after the welding of the frame and the shell is completed, the next process is spraying. The paint used for spraying is divided into ordinary paint, metallic paint, and pearlescent paint according to the form. Car spraying process: scraping putty, grinding, spraying, etc.; after the car completes the spraying process, the car paint surface needs to be inspected for spraying defects to ensure that the car spraying meets the manufacturing process requirements. However, the existing car spraying paint surface defect quality inspection mostly uses simple image recognition technology and cooperates with quality inspectors to conduct sampling inspections on the paint surface. It is impossible to achieve scientific and accurate defect location detection of the entire car body paint surface, and it is also impossible to provide intuitive feedback on the car paint surface defect information.

[0003] The Chinese invention patent with announcement number CN115561249B discloses an intelligent monitoring method and system for spraying equipment, which relates to the field of automated spraying monitoring technology. The method obtains the position parameters of the target spraying device, performs image acquisition on the target spraying device to obtain the initial image acquisition result, and performs actual size accuracy evaluation on the target spraying device to generate a size deviation result; obtains the spraying image acquisition result, further performs spraying defect analysis and spraying accuracy analysis on the spraying equipment, generates spraying defect evaluation results and spraying accuracy evaluation results, and then generates spraying monitoring results for the spraying equipment. The method solves the technical problems in the prior art that the spraying monitoring method is not intelligent enough and the monitoring process is not rigorous enough, resulting in the final spraying monitoring result being not accurate enough and having a certain deviation from the actual spraying effect, and realizes intelligent and accurate monitoring of the target spraying device. The above technical scheme cannot realize comprehensive defect positioning monitoring of the automobile spray paint surface and visual feedback of defect detection results. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In order to solve the above-mentioned problems that the existing automobile spray paint defect quality inspection mostly uses simple image recognition technology and cooperates with quality inspectors to conduct sampling inspection of the paint surface, it is impossible to achieve scientific and accurate defect positioning detection of the entire automobile paint surface, and it is also impossible to provide intuitive feedback on the automobile paint surface defect information. The above purpose is to achieve complete and scientific collection of body paint defect detection surface sample objects, accurate analysis of body paint detection surface sample defect type parameters, accurate evaluation of automobile body spraying quality, and visual display of automobile spray paint surface defect detection location and results.

[0006] (II) Technical solution

[0007] The present invention is implemented by the following technical scheme: a real-time monitoring method for automobile spraying defects at the user end of the Internet of Things, the method comprising the following steps:

[0008] S1. Collecting vehicle body paint surface spatial model data;

[0009] S2, performing body paint surface mesh division and mesh coordinate processing based on the body paint surface spatial model data, and generating body paint surface mesh detection surface image data and body paint surface mesh detection surface coordinate data respectively;

[0010] S3, performing image positioning processing on the vehicle body paint surface grid detection surface according to the image and coordinate parameters of the vehicle body paint surface grid detection surface, and generating vehicle body paint surface grid detection surface image positioning data;

[0011] S4, identifying vehicle body paint surface defects according to the vehicle body paint surface grid detection surface image positioning data and the vehicle body paint surface defect type image data, constructing vehicle body paint surface grid detection surface defect identification data and screening and generating the number of vehicle body paint surface grid detection surface defect individuals;

[0012] S5, performing body paint surface detection defect statistics based on the body paint surface grid detection surface defect identification data and the body paint surface grid detection surface defect individual number, and calculating body paint surface detection surface defect rate data;

[0013] S6, performing body paint defect assessment analysis based on the body paint detection surface defect rate data and the body paint detection qualified defect rate data to generate body paint defect detection assessment data;

[0014] S7. Construct three-dimensional result data of vehicle body paint defect detection and output it through a display device; push the vehicle body paint defect detection evaluation data to the vehicle body quality inspection management center through the Internet of Things communication network.

[0015] Preferably, the operation steps of collecting the vehicle body paint surface spatial model data are as follows:

[0016] S11. Use a 3D laser scanner to scan the surface contour of the car body paint after the car spraying process to generate the body paint surface spatial model data .

[0017] Preferably, the steps of performing body paint surface meshing and mesh coordinate processing based on the body paint surface space model data to respectively generate body paint surface mesh detection surface image data and body paint surface mesh detection surface coordinate data are as follows:

[0018] S21, the vehicle body paint surface space model data Use equal area square Perform paint surface meshing and obtain meshed paint surface images to construct a body paint surface mesh detection surface image data set , ;in Indicates Car body paint surface mesh detection surface image data, Indicates the maximum number of mesh detection surfaces on the car body paint surface;

[0019] S22, establishing a spatial rectangular coordinate system and calculating the vehicle body paint surface grid detection surface image data set All the body paint surface mesh detection surface image data The spatial coordinates of the center point of the image are used to generate the coordinate data set of the body paint surface grid detection surface ,in Indicates Vehicle body paint surface mesh detection surface coordinate data corresponding to the vehicle body paint surface mesh detection surface image; ,in Indicates The horizontal coordinate corresponding to the body paint mesh detection surface image, Indicates The vertical coordinate corresponding to the body paint mesh detection surface image, Indicates The vertical coordinate corresponding to the body paint mesh detection surface image.

[0020] Preferably, the steps of performing the image positioning processing of the vehicle body paint surface grid detection surface according to the image and coordinate parameters of the vehicle body paint surface grid detection surface to generate the image positioning data of the vehicle body paint surface grid detection surface are as follows:

[0021] S31, calling the vehicle body paint surface grid detection surface image data set and the body paint surface grid detection surface coordinate data set ;

[0022] S32, the body paint surface grid detection surface coordinate data set Coordinate data of the mesh detection surface of the vehicle body paint surface Taking the space rectangular coordinate system as the reference system and the vehicle body paint surface grid detection surface image data set Car body paint surface mesh detection surface image data Match and fuse the grid detection surface coordinates with the grid detection surface image data to generate a body paint surface grid detection surface image positioning data set ,in Represents the body paint surface grid detection surface coordinate data and body paint mesh detection surface image data The data is matched and fused with the corresponding body paint surface grid detection surface image positioning data.

[0023] Preferably, the steps of performing body paint defect recognition according to the body paint grid detection surface image positioning data and the body paint defect type image data, constructing body paint grid detection surface defect recognition data and screening and generating the number of body paint grid detection surface defect individuals are as follows:

[0024] S41. Establishing a vehicle body paint defect type image data set , ;in Indicates Types of car body paint defect image data, Indicates the maximum value of the number of paint defect types of the vehicle body, wherein the paint defect types of the vehicle body include paint wrinkling and shrinkage, paint sandpaper marks, paint tears, paint cracks and paint lumps;

[0025] S42, using a data recognition algorithm to locate the vehicle body paint surface grid detection surface image data set Car body paint surface grid detection surface image positioning data The image data set of the vehicle body paint surface defect type is ordered according to the vehicle body paint surface grid detection surface number. Car body paint defect type image data Perform image matching to search for the vehicle body paint surface grid detection surface image positioning data The type of vehicle body paint defect is searched for the vehicle body paint grid detection surface image positioning data The specific steps for the types of body paint defects are as follows:

[0026] S421, initializing algorithm parameters, population size N, maximum number of iterations T;

[0027] S422, initializing the population, calculating fitness, and determining defect recognition pathfinders and defect recognition followers;

[0028] S423, according to the position formula Car body paint defect type image data collection Update the defect identification pathfinder position in the search space, where t represents the current iteration number of the algorithm; Denotes the defect identification pathfinder after the tth iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification pathfinder after the t-1th iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification pathfinder after the t+1th iteration Car body paint defect type image data collection The position in the search space, The step size factor for the defect identification pathfinder movement, It follows a uniform distribution in [0,1];

[0029] S424, according to the position formula Update defect recognition follower in the body paint defect type image data set The position in the search space where Denotes the defect identification follower after the tth iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification follower after the t+1th iteration Car body paint defect type image data collection Position in the search space; Indicates other defect identification followers after the tth iteration Car body paint defect type image data collection Position in the search space, where the defect identification follows Mobile not only with defect identification pathfinder position Related, and subject to other defects to identify the follower position The impact of Represents the defect recognition follower in the body paint defect type image data set The position distance parameter in the search space, Represents the image data set of the defect type between the defect recognition pathfinder and the defect recognition follower in the body paint defect type The position distance parameter in the search space, , ; represents the interaction coefficient between defect recognition followers, represents the attraction coefficient of the defect recognition pathfinder to the defect recognition follower, , All are uniformly distributed in [1,2]; is the step size factor for the defect recognition follower to move with other defect recognition followers, The step size factor for the Defect Recognition Follower and Defect Recognition Pathfinder to move, , All are random numbers in the range [0,1];

[0030] S425, calculate the fitness value and update the global optimal value, that is, in the body paint defect type image data set Search the search space to find the image positioning data of the vehicle body paint surface grid detection surface Matching the vehicle body paint defect type image data Corresponding body paint defect type;

[0031] S426: When the maximum number of iterations is met, output the image positioning data of the vehicle body paint surface grid detection surface searched out in step S425 Matching body paint defect types and constructing a body paint grid detection surface defect recognition data set ,in Represents the body paint surface grid detection surface image positioning data Corresponding vehicle body paint surface grid detection surface defect identification data; the vehicle body paint surface grid detection surface defect identification data includes the presence of specific defect types on the vehicle body paint surface and the absence of specific defect types on the vehicle body paint surface;

[0032] S43, using a breadth-first search algorithm to search for the body paint surface grid detection surface defect recognition data set according to the body paint surface defect type keyword Defect recognition data of the body paint surface grid detection surface in which the body paint surface has a specific defect type And perform counting processing to generate the number of individual defects on the body paint mesh detection surface , .

[0033] Preferably, the steps of performing body paint surface detection defect statistics based on the body paint surface grid detection surface defect recognition data and the body paint surface grid detection surface defect individual number to measure the body paint surface detection surface defect rate data are as follows:

[0034] S51, based on the vehicle body paint surface grid detection surface defect recognition data set Get the total number of mesh detection surfaces on the car body paint surface ;

[0035] S52, the number of individual defects on the vehicle body paint surface grid detection surface The total number of mesh detection surfaces with the vehicle body paint surface Perform numerical quotient processing to calculate the defect rate data of the body paint surface inspection surface .

[0036] Preferably, the vehicle body paint surface defect evaluation and analysis is performed based on the vehicle body paint surface detection surface defect rate data and the vehicle body paint surface detection qualified defect rate data, and the operating steps for generating the vehicle body paint surface defect detection and evaluation data are as follows:

[0037] S61. Establish qualified defect rate data for vehicle body paint inspection ,in It indicates the minimum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification. It indicates the maximum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification;

[0038] S62, the vehicle body paint surface detection surface defect rate data The body paint inspection qualified defect rate data Compare defect rate values ​​and build body paint defect detection evaluation data based on defect rate value comparison results ;

[0039] when ∈ , indicating that the body paint surface detection defect rate meets the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output To be qualified;

[0040] when ∉ , indicating that the body paint surface defect rate does not meet the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output Is unqualified.

[0041] Preferably, the steps of constructing the three-dimensional result data of the vehicle body paint defect detection and displaying and outputting it through a display device; and pushing the vehicle body paint defect detection evaluation data to the vehicle body quality inspection management center through the Internet of Things communication network are as follows:

[0042] S71, the vehicle body paint surface grid detection surface defect recognition data set Car body paint surface mesh detection surface defect recognition data The vehicle body paint surface grid detection surface image positioning data set Car body paint surface grid detection surface image positioning data According to the body paint surface grid detection surface number, data matching and fusion processing are performed to generate the body paint surface grid detection surface image positioning and defect recognition data set ,in Represents the body paint surface mesh detection surface defect recognition data and body paint mesh detection surface image positioning data The data is matched and integrated with the corresponding vehicle body paint surface grid detection surface image positioning and defect recognition data;

[0043] S72: locating the vehicle body paint grid detection surface image and collecting defect recognition data Car body paint surface grid detection surface image positioning and defect recognition data The spatial model data of the vehicle body paint surface is projected onto the vehicle body paint surface using the spatial rectangular coordinate system as the reference system. Corresponding to the three-dimensional model surface of the car body paint, construct the three-dimensional result data of the car body paint defect detection and displaying the output through a display device, wherein the display device includes any one of a display screen and a projection device;

[0044] S73, the vehicle body paint surface defect detection evaluation data Pushed to the vehicle body quality inspection management center through the Internet of Things communication network.

[0045] A user-side automobile spraying defect real-time monitoring system of the Internet of Things, used to implement the user-side automobile spraying defect real-time monitoring method of the Internet of Things, the system includes a body detection point construction module, a body detection point paint surface defect analysis module, and a body detection point paint surface defect monitoring and display module;

[0046] The vehicle body detection point construction module includes a vehicle body paint surface space model acquisition unit, a vehicle body paint surface detection surface division unit, a vehicle body paint surface detection coordinate generation unit, and a vehicle body paint surface detection surface and coordinate fusion unit;

[0047] The vehicle body paint surface space model acquisition unit acquires vehicle body paint surface space model data through a three-dimensional laser scanner; the vehicle body paint surface detection surface division unit performs vehicle body paint surface mesh division processing based on the vehicle body paint surface space model data to generate vehicle body paint surface mesh detection surface image data; the vehicle body paint surface detection coordinate generation unit performs vehicle body paint surface mesh coordinate processing based on the vehicle body paint surface space model data to generate vehicle body paint surface mesh detection surface coordinate data; the vehicle body paint surface detection surface and coordinate fusion unit performs vehicle body paint surface mesh detection surface image positioning processing according to the image and coordinate parameters of the vehicle body paint surface mesh detection surface to generate vehicle body paint surface mesh detection surface image positioning data;

[0048] The body inspection point paint surface defect analysis module includes a body paint surface defect type image storage unit, a body paint surface inspection surface defect recognition unit, a body paint surface inspection surface defect object screening unit, a body paint surface inspection surface defect rate measurement unit, a body paint surface inspection qualified defect rate storage unit, and a body paint surface defect detection result evaluation unit;

[0049] The vehicle body paint defect type image storage unit is used to store vehicle body paint defect type image data; the vehicle body paint detection surface defect recognition unit performs vehicle body paint defect recognition according to the vehicle body paint grid detection surface image positioning data and the vehicle body paint defect type image data, and constructs vehicle body paint grid detection surface defect recognition data; the vehicle body paint detection surface defect object screening unit is used to screen vehicle body paint detection surface defect objects with defects and measure and generate the number of vehicle body paint grid detection surface defect individuals; the vehicle body paint detection surface defect rate measurement unit performs vehicle body paint detection defect statistics according to the vehicle body paint grid detection surface defect recognition data and the number of vehicle body paint grid detection surface defect individuals, and measures the vehicle body paint detection surface defect rate data; the vehicle body paint detection qualified defect rate storage unit is used to store vehicle body paint detection qualified defect rate data; the vehicle body paint detection result evaluation unit performs vehicle body paint defect evaluation and analysis according to the vehicle body paint detection surface defect rate data and the vehicle body paint detection qualified defect rate data, and generates vehicle body paint defect detection evaluation data;

[0050] The vehicle body inspection point paint surface defect monitoring and display module includes a vehicle body paint surface defect detection result and vehicle body fusion unit, a vehicle body inspection point paint surface defect monitoring result display unit, and a vehicle body paint surface defect detection evaluation result push unit;

[0051] The vehicle body paint defect detection result and vehicle body fusion unit constructs three-dimensional result data of vehicle body paint defect detection; the vehicle body inspection point paint defect monitoring result display unit displays and outputs the three-dimensional result data of vehicle body paint defect detection through a display device; the vehicle body paint defect detection evaluation result push unit pushes the vehicle body paint defect detection evaluation data to the vehicle body quality inspection management center through the Internet of Things communication network.

[0052] (III) Beneficial effects

[0053] The present invention provides a real-time monitoring system and method for automobile spraying defects at the user end of the Internet of Things. It has the following beneficial effects:

[0054] 1. Through the body paint surface space model acquisition unit, a three-dimensional laser scanner is used to efficiently construct the body paint surface space model parameters, providing data support for the subsequent scientific detection of body paint surface defect results; the body paint surface detection surface division unit and the body paint surface detection coordinate generation unit cooperate with each other to evenly grid the automobile body paint surface space model and grid position it, so as to achieve complete and scientific collection of the image and position parameters of the automobile body paint surface defect detection surface sample; the body paint surface detection surface and coordinate fusion unit fuses the image and position data of the collected automobile body paint surface detection surface samples, accurately constructs the automobile body paint surface detection surface sample image positioning combination parameters, and realizes the accurate collection and positioning of the automobile body paint surface detection surface samples.

[0055] 2. Through the body paint detection surface defect recognition unit, the body paint grid detection surface image positioning parameters are scientifically preset, combined with the intelligent recognition algorithm and the body paint defect type image parameters to identify the body paint defect type, accurately analyze the defect recognition result parameters of each paint detection surface sample of the body, and realize comprehensive and reliable monitoring of body paint defects; the body paint detection surface defect rate measurement unit and the body paint defect detection result evaluation unit cooperate with each other to independently screen and count the number of defective automobile body paint detection surface samples, scientifically measure the automobile body paint detection surface sample defect rate parameters and analyze the body paint defect detection evaluation parameters, and realize the digital statistics of automobile body paint spraying defects and accurate evaluation and analysis of spraying quality.

[0056] 3. Through the cooperation between the body paint defect detection results and the body fusion unit and the body inspection point paint defect monitoring result display unit, the automobile body paint defect detection results are intelligently mapped to the sample position of the automobile body paint inspection surface, and the automobile body spray defect detection results are intuitively visualized in three dimensions; the body paint defect detection evaluation result push unit, combined with the Internet of Things, flexibly pushes the body paint defect detection evaluation parameters to the body quality inspection management center, so as to realize the online monitoring and feedback of the automobile body spray paint quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of a module of a real-time monitoring system and method for automobile spraying defects at a user end of the Internet of Things provided by the present invention;

[0058] Figure 2 A flow chart of a method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things provided by the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0060] The embodiments of the real-time monitoring system and method for automobile spraying defects at the user end of the Internet of Things are as follows:

[0061] Embodiment 1:

[0062] See also Figure 1 - Figure 2 , a real-time monitoring method for automobile spraying defects at the user end of the Internet of Things, the method comprises the following steps:

[0063] S1. Collecting vehicle body paint surface spatial model data;

[0064] S2, based on the vehicle body paint surface space model data, perform vehicle body paint surface mesh division and mesh coordinate processing, and generate vehicle body paint surface mesh detection surface image data and vehicle body paint surface mesh detection surface coordinate data respectively;

[0065] S3, performing image positioning processing on the vehicle body paint surface grid detection surface according to the image and coordinate parameters of the vehicle body paint surface grid detection surface, and generating vehicle body paint surface grid detection surface image positioning data;

[0066] S4, identifying vehicle body paint surface defects according to the vehicle body paint surface grid detection surface image positioning data and the vehicle body paint surface defect type image data, constructing vehicle body paint surface grid detection surface defect identification data and screening and generating the number of vehicle body paint surface grid detection surface defect individuals;

[0067] S5. Count the defects of the vehicle body paint surface inspection according to the vehicle body paint surface grid inspection surface defect recognition data and the number of individual defects of the vehicle body paint surface grid inspection surface, and calculate the vehicle body paint surface inspection surface defect rate data;

[0068] S6. Performing body paint defect assessment and analysis based on body paint inspection surface defect rate data and body paint inspection qualified defect rate data to generate body paint defect inspection and assessment data;

[0069] S7. Construct the three-dimensional result data of the vehicle body paint defect detection and display it through the display device; push the vehicle body paint defect detection evaluation data to the vehicle body quality inspection management center through the Internet of Things communication network.

[0070] For further information, see Figure 1 - Figure 2 ,The steps for collecting the body paint surface space model data are as follows:

[0071] S11. Use a 3D laser scanner to scan the surface contour of the car body paint after the car spraying process to generate the body paint surface spatial model data .

[0072] The steps for meshing the paint surface of the vehicle body and processing the mesh coordinates based on the paint surface space model data of the vehicle body to generate the paint surface mesh detection surface image data and the paint surface mesh detection surface coordinate data of the vehicle body are as follows:

[0073] S21, body paint surface space model data Use equal area square Perform paint surface meshing and obtain meshed paint surface images to construct a body paint surface mesh detection surface image data set , ;in Indicates Car body paint surface mesh detection surface image data, Indicates the maximum number of mesh detection surfaces on the car body paint surface;

[0074] S22. Establish a spatial rectangular coordinate system and calculate the body paint surface grid detection surface image data set All the body paint mesh detection surface image data in The spatial coordinates of the center point of the image are used to generate the coordinate data set of the body paint surface grid detection surface ,in Indicates Vehicle body paint surface mesh detection surface coordinate data corresponding to the vehicle body paint surface mesh detection surface image; ,in Indicates The horizontal coordinate corresponding to the body paint mesh detection surface image, Indicates The vertical coordinate corresponding to the body paint mesh detection surface image, Indicates The vertical coordinate corresponding to the body paint mesh detection surface image.

[0075] The steps of performing image positioning processing on the body paint surface grid detection surface according to the image and coordinate parameters of the body paint surface grid detection surface and generating the image positioning data of the body paint surface grid detection surface are as follows:

[0076] S31. Calling the vehicle body paint surface grid detection surface image data set and the body paint surface mesh detection surface coordinate data set ;

[0077] S32, the body paint surface grid detection surface coordinate data set Coordinate data of the mesh detection surface of the vehicle body paint surface The image data set of the vehicle body paint surface grid detection surface with the spatial rectangular coordinate system as the reference system Car body paint surface mesh detection surface image data Match and fuse the grid detection surface coordinates with the grid detection surface image data to generate a body paint surface grid detection surface image positioning data set ,in Represents the body paint surface grid detection surface coordinate data and body paint mesh detection surface image data The data is matched and fused with the corresponding body paint surface grid detection surface image positioning data.

[0078] Through the body paint space model acquisition unit, a three-dimensional laser scanner is used to efficiently construct the body paint space model parameters, providing data support for the subsequent scientific detection of body paint defect results; the body paint detection surface division unit and the body paint detection coordinate generation unit cooperate with each other to evenly grid the automobile body paint space model and grid position it, so as to achieve complete and scientific collection of the image and position parameters of the automobile body paint defect detection surface samples; the body paint detection surface and coordinate fusion unit fuses the image and position data of the collected automobile body paint detection surface samples, accurately constructs the automobile body paint detection surface sample image positioning combination parameters, and realizes the accurate collection and positioning of the automobile body paint detection surface samples.

[0079] For further information, see Figure 1 - Figure 2 The steps of identifying body paint surface defects according to the body paint surface grid detection surface image positioning data and body paint surface defect type image data, constructing body paint surface grid detection surface defect identification data and screening and generating the number of body paint surface grid detection surface defect individuals are as follows:

[0080] S41. Establishing a vehicle body paint defect type image data set , ;in Indicates Types of car body paint defect image data, Indicates the maximum number of types of paint defects on the vehicle body. The types of paint defects on the vehicle body include paint wrinkling and shrinkage, paint sandpaper marks, paint tears, paint cracks and paint bumps;

[0081] S42, using data recognition algorithm to locate the body paint surface grid detection surface image data set Car body paint surface grid detection surface image positioning data Image data set of vehicle body paint surface defect types ordered by vehicle body paint surface grid detection surface number Car body paint defect type image data Perform image matching to search for the image positioning data of the body paint grid detection surface The type of body paint defect is searched to obtain the body paint grid detection surface image positioning data The specific steps for the types of body paint defects are as follows:

[0082] S421, initializing algorithm parameters, population size N, maximum number of iterations T;

[0083] S422, initializing the population, calculating fitness, and determining defect recognition pathfinders and defect recognition followers;

[0084] S423, according to the position formula Car body paint defect type image data collection Update the defect identification pathfinder position in the search space, where t represents the current iteration number of the algorithm; Denotes the defect identification pathfinder after the tth iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification pathfinder after the t-1th iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification pathfinder after the t+1th iteration Car body paint defect type image data collection The position in the search space, The step size factor for the defect identification pathfinder movement, It follows a uniform distribution in [0,1];

[0085] S424, according to the position formula Update defect recognition follower in the body paint defect type image data set The position in the search space where Denotes the defect identification follower after the tth iteration Car body paint defect type image data collection The position in the search space, Denotes the defect identification follower after the t+1th iteration Car body paint defect type image data collection Position in the search space; Indicates other defect identification followers after the tth iteration Car body paint defect type image data collection Position in the search space, where the defect identification follows Mobile not only with defect identification pathfinder position Related, and subject to other defects to identify the follower position The impact of Represents the defect recognition follower in the body paint defect type image data set The position distance parameter in the search space, Represents the image data set of the defect type between the defect recognition pathfinder and the defect recognition follower in the body paint defect type The position distance parameter in the search space, , ; represents the interaction coefficient between defect recognition followers, represents the attraction coefficient of the defect recognition pathfinder to the defect recognition follower, , All are uniformly distributed in [1,2]; is the step size factor for the defect recognition follower to move with other defect recognition followers, The step size factor for the Defect Recognition Follower and Defect Recognition Pathfinder to move, , All are random numbers in the range [0,1];

[0086] S425, calculate the fitness value and update the global optimal value, that is, in the body paint defect type image data set Search the search space to find the image positioning data of the body paint surface grid detection surface Matching car body paint defect type image data Corresponding body paint defect type;

[0087] S426: When the maximum number of iterations is met, output the image positioning data of the vehicle body paint surface grid detection surface searched in step S425 Matching body paint defect types and constructing a body paint grid detection surface defect recognition data set ,in Represents the body paint surface grid detection surface image positioning data Corresponding vehicle body paint surface grid detection surface defect recognition data; the vehicle body paint surface grid detection surface defect recognition data includes the presence of specific defect types on the vehicle body paint surface and the absence of specific defect types on the vehicle body paint surface;

[0088] S43, using the breadth-first search algorithm to search for the body paint surface grid detection surface defect recognition data set according to the body paint surface defect type keywords Defect recognition data of the body paint mesh detection surface with specific defect types in the body paint And perform counting processing to generate the number of individual defects on the body paint mesh detection surface , .

[0089] The steps for counting the defects of the paint surface inspection according to the defect recognition data of the paint surface grid inspection and the number of defects of the paint surface grid inspection are as follows:

[0090] S51, based on the body paint surface grid detection surface defect recognition data set Get the total number of mesh detection surfaces on the car body paint surface ;

[0091] S52, the number of individual defects on the body paint surface grid detection surface Total number of mesh detection surfaces with body paint Perform numerical quotient processing to calculate the defect rate data of the body paint surface inspection surface .

[0092] Car body paint surface defect evaluation and analysis is performed based on the car body paint surface defect rate data and the car body paint surface qualified defect rate data. The operating steps for generating car body paint surface defect detection and evaluation data are as follows:

[0093] S61. Establish qualified defect rate data for vehicle body paint inspection ,in It indicates the minimum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification. It indicates the maximum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification;

[0094] S62, the body paint surface defect rate data Data on qualified defect rate of vehicle body paint inspection Compare defect rate values ​​and build body paint defect detection evaluation data based on defect rate value comparison results ;

[0095] when ∈ , indicating that the body paint surface detection defect rate meets the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output To be qualified;

[0096] when ∉ , indicating that the body paint surface defect rate does not meet the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output Is unqualified.

[0097] Through the body paint detection surface defect recognition unit, the body paint grid detection surface image positioning parameters are scientifically preset and combined with the intelligent recognition algorithm and the body paint defect type image parameters to identify the body paint defect type, accurately analyze the defect recognition result parameters of each paint detection surface sample of the body, and realize comprehensive and reliable monitoring of body paint defects; the body paint detection surface defect rate measurement unit and the body paint defect detection result evaluation unit cooperate with each other to independently screen and count the number of automobile body paint detection surface samples with defects, scientifically measure the automobile body paint detection surface sample defect rate parameters and analyze the body paint defect detection evaluation parameters, and realize the digital statistics of automobile body paint spraying defects and accurate evaluation and analysis of spraying quality.

[0098] For further information, see Figure 1 - Figure 2, construct the three-dimensional result data of the body paint defect detection and display it through the display device; the operation steps of pushing the body paint defect detection evaluation data to the body quality inspection management center through the Internet of Things communication network are as follows:

[0099] S71, collect the body paint surface grid detection surface defect recognition data Car body paint surface mesh detection surface defect recognition data Car body paint surface mesh detection surface image positioning data set Car body paint surface grid detection surface image positioning data According to the body paint surface grid detection surface number, data matching and fusion processing are performed to generate the body paint surface grid detection surface image positioning and defect recognition data set ,in Represents the body paint surface mesh detection surface defect recognition data and body paint mesh detection surface image positioning data The data is matched and integrated with the corresponding vehicle body paint surface grid detection surface image positioning and defect recognition data;

[0100] S72, the body paint surface grid detection surface image positioning and defect recognition data collection Car body paint surface grid detection surface image positioning and defect recognition data The spatial model data projected onto the body paint surface is taken as the reference system using the spatial rectangular coordinate system. Corresponding to the three-dimensional model surface of the car body paint, construct the three-dimensional result data of the car body paint defect detection and displaying the output through a display device, the display device including any one of a display screen and a projection device;

[0101] S73, the body paint defect detection evaluation data Pushed to the vehicle body quality inspection management center through the Internet of Things communication network.

[0102] Through the cooperation between the body paint defect detection results and the body fusion unit and the body inspection point paint defect monitoring result display unit, the automobile body paint defect detection results are intelligently mapped to the automobile body paint inspection surface sample position, and the automobile body spray defect detection results are intuitively visualized in three dimensions; the body paint defect detection evaluation result push unit, combined with the Internet of Things, flexibly pushes the body paint defect detection evaluation parameters to the body quality inspection management center, to realize online monitoring and feedback of the automobile body spray paint quality.

[0103] Embodiment 2:

[0104] See also Figure 1 - Figure 2, a user-end automobile spraying defect real-time monitoring system of the Internet of Things, used to implement a user-end automobile spraying defect real-time monitoring method of the Internet of Things, the system includes a body detection point construction module, a body detection point paint surface defect analysis module, and a body detection point paint surface defect monitoring display module;

[0105] The vehicle body inspection point construction module includes a vehicle body paint surface space model acquisition unit, a vehicle body paint surface inspection surface division unit, a vehicle body paint surface inspection coordinate generation unit, and a vehicle body paint surface inspection surface and coordinate fusion unit;

[0106] The vehicle body paint surface space model acquisition unit acquires the vehicle body paint surface space model data through a three-dimensional laser scanner; the vehicle body paint surface detection surface division unit performs vehicle body paint surface mesh division processing based on the vehicle body paint surface space model data, and generates vehicle body paint surface mesh detection surface image data; the vehicle body paint surface detection coordinate generation unit performs vehicle body paint surface mesh coordinate processing based on the vehicle body paint surface space model data, and generates vehicle body paint surface mesh detection surface coordinate data; the vehicle body paint surface detection surface and coordinate fusion unit performs vehicle body paint surface mesh detection surface image positioning processing based on the image and coordinate parameters of the vehicle body paint surface mesh detection surface, and generates vehicle body paint surface mesh detection surface image positioning data;

[0107] The paint defect analysis module of the vehicle body inspection point includes a vehicle body paint defect type image storage unit, a vehicle body paint inspection surface defect recognition unit, a vehicle body paint inspection surface defect object screening unit, a vehicle body paint inspection surface defect rate measurement unit, a vehicle body paint inspection qualified defect rate storage unit, and a vehicle body paint inspection result evaluation unit;

[0108] A body paint defect type image storage unit is used to store body paint defect type image data; a body paint detection surface defect recognition unit is used to recognize body paint defects according to body paint grid detection surface image positioning data and body paint defect type image data, and construct body paint grid detection surface defect recognition data; a body paint detection surface defect object screening unit is used to screen body paint detection surface defect objects with defects and measure and generate the number of body paint grid detection surface defect individuals; a body paint detection surface defect rate measurement unit is used to perform body paint detection defect statistics according to body paint grid detection surface defect recognition data and the number of body paint grid detection surface defect individuals, and measure the body paint detection surface defect rate data; a body paint detection qualified defect rate storage unit is used to store body paint detection qualified defect rate data; a body paint defect detection result evaluation unit is used to perform body paint defect evaluation and analysis according to body paint detection surface defect rate data and body paint detection qualified defect rate data, and generate body paint defect detection evaluation data;

[0109] The paint defect monitoring and display module of the vehicle body inspection point includes a vehicle body paint defect detection result and vehicle body fusion unit, a vehicle body inspection point paint defect monitoring result display unit, and a vehicle body paint defect detection evaluation result push unit;

[0110] The body paint defect detection result and the body fusion unit construct the three-dimensional result data of the body paint defect detection; the body inspection point paint defect monitoring result display unit displays and outputs the three-dimensional result data of the body paint defect detection through the display device; the body paint defect detection evaluation result push unit pushes the body paint defect detection evaluation data to the body quality inspection management center through the Internet of Things communication network.

[0111] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring method for automobile spraying defects at the user end of the Internet of Things, characterized in that: The method comprises the following steps: S1. Collecting vehicle body paint surface spatial model data; S2, performing body paint surface mesh division and mesh coordinate processing based on the body paint surface spatial model data, and generating body paint surface mesh detection surface image data and body paint surface mesh detection surface coordinate data respectively; S3, performing image positioning processing on the vehicle body paint surface grid detection surface according to the image and coordinate parameters of the vehicle body paint surface grid detection surface, and generating vehicle body paint surface grid detection surface image positioning data; S4, identifying vehicle body paint surface defects according to the vehicle body paint surface grid detection surface image positioning data and the vehicle body paint surface defect type image data, constructing vehicle body paint surface grid detection surface defect identification data and screening and generating the number of vehicle body paint surface grid detection surface defect individuals; S5, performing body paint surface detection defect statistics based on the body paint surface grid detection surface defect identification data and the body paint surface grid detection surface defect individual number, and calculating body paint surface detection surface defect rate data; S6, performing body paint defect assessment analysis based on the body paint detection surface defect rate data and the body paint detection qualified defect rate data to generate body paint defect detection assessment data; S7. Construct three-dimensional result data of vehicle body paint defect detection and output it through a display device; push the vehicle body paint defect detection evaluation data to the vehicle body quality inspection management center through the Internet of Things communication network.

2. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 1 is characterized in that: The S1 comprises the following steps: S11. Use a 3D laser scanner to scan the surface contour of the car body paint after the car spraying process to generate the body paint surface spatial model data .

3. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 2 is characterized in that: The S2 comprises the following steps: S21, the vehicle body paint surface space model data Use equal area square Perform paint surface meshing and obtain meshed paint surface images to construct a body paint surface mesh detection surface image data set , ;in Indicates Car body paint surface mesh detection surface image data, Indicates the maximum number of mesh detection surfaces on the car body paint surface; S22, establishing a spatial rectangular coordinate system and calculating the vehicle body paint surface grid detection surface image data set All the body paint surface mesh detection surface image data The spatial coordinates of the center point of the image are used to generate the coordinate data set of the body paint surface grid detection surface ,in Indicates The body paint surface mesh detection surface coordinate data corresponding to the body paint surface mesh detection surface image.

4. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 3 is characterized in that: The S3 comprises the following steps: S31, calling the vehicle body paint surface grid detection surface image data set and the body paint surface grid detection surface coordinate data set ; S32, the body paint surface grid detection surface coordinate data set Coordinate data of the mesh detection surface of the vehicle body paint surface Taking the space rectangular coordinate system as the reference system and the vehicle body paint surface grid detection surface image data set Car body paint surface mesh detection surface image data Match and fuse the grid detection surface coordinates with the grid detection surface image data to generate a body paint surface grid detection surface image positioning data set ,in Represents the body paint surface grid detection surface coordinate data and body paint mesh detection surface image data The data is matched and fused with the corresponding body paint surface grid detection surface image positioning data.

5. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 4 is characterized in that: The S4 comprises the following steps: S41. Establishing a vehicle body paint defect type image data set , ;in Indicates Types of car body paint defect image data, Indicates the maximum number of car body paint defect types; S42: locating the vehicle body paint surface grid detection surface image data set Car body paint surface grid detection surface image positioning data The image data set of the vehicle body paint surface defect type is ordered according to the vehicle body paint surface grid detection surface number. Car body paint defect type image data Perform image matching to search for the vehicle body paint surface grid detection surface image positioning data The type of vehicle body paint defect is searched for the vehicle body paint grid detection surface image positioning data The specific steps for the types of body paint defects are as follows: S421, initializing algorithm parameters, population size N, maximum number of iterations T; S422, initializing the population, calculating fitness, and determining defect recognition pathfinders and defect recognition followers; S423, collecting vehicle body paint surface defect type image data Update defect identification pathfinder position in search space; S424, updating the defect recognition follower in the body paint defect type image data set Position in the search space; S425, calculate the fitness value and update the global optimal value, that is, in the body paint defect type image data set Search the search space to find the image positioning data of the vehicle body paint surface grid detection surface Matching the vehicle body paint defect type image data Corresponding body paint defect type; S426: When the maximum number of iterations is met, output the image positioning data of the vehicle body paint surface grid detection surface searched out in step S425 Matching body paint defect types and constructing a body paint grid detection surface defect recognition data set ,in Represents the body paint surface grid detection surface image positioning data Corresponding vehicle body paint surface grid detection surface defect identification data; the vehicle body paint surface grid detection surface defect identification data includes the presence of specific defect types on the vehicle body paint surface and the absence of specific defect types on the vehicle body paint surface; S43, using a breadth-first search algorithm to search for the body paint surface grid detection surface defect recognition data set according to the body paint surface defect type keyword Defect recognition data of the body paint surface grid detection surface in which the body paint surface has a specific defect type And perform counting processing to generate the number of individual defects on the body paint mesh detection surface , .

6. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 5 is characterized in that: The S5 comprises the following steps: S51, based on the vehicle body paint surface grid detection surface defect recognition data set Get the total number of mesh detection surfaces on the car body paint surface ; S52, the number of individual defects on the vehicle body paint surface grid detection surface The total number of mesh detection surfaces with the vehicle body paint surface Perform numerical quotient processing to calculate the defect rate data of the body paint surface inspection surface .

7. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 6 is characterized in that: The S6 comprises the following steps: S61. Establish qualified defect rate data for vehicle body paint inspection ,in It indicates the minimum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification. It indicates the maximum value of the qualified defect rate data of the body paint surface inspection that meets the requirements of the automobile body spraying process specification; S62, the vehicle body paint surface detection surface defect rate data The body paint inspection qualified defect rate data Compare defect rate values ​​and build body paint defect detection evaluation data based on defect rate value comparison results ; when ∈ , indicating that the body paint surface detection defect rate meets the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output To be qualified; when ∉ , indicating that the body paint surface defect rate does not meet the requirements of the automobile body spraying process specification, then the body paint surface defect detection evaluation data is output Is unqualified.

8. The method for real-time monitoring of automobile spraying defects at the user end of the Internet of Things according to claim 7 is characterized in that: The S7 comprises the following steps: S71, the vehicle body paint surface grid detection surface defect recognition data set Car body paint surface mesh detection surface defect recognition data The vehicle body paint surface grid detection surface image positioning data set Car body paint surface grid detection surface image positioning data According to the body paint surface grid detection surface number, data matching and fusion processing are performed to generate the body paint surface grid detection surface image positioning and defect recognition data set ,in Represents the body paint surface mesh detection surface defect recognition data and body paint mesh detection surface image positioning data The data is matched and integrated with the corresponding vehicle body paint surface grid detection surface image positioning and defect recognition data; S72: locating the vehicle body paint grid detection surface image and collecting defect recognition data Car body paint surface grid detection surface image positioning and defect recognition data The spatial model data of the vehicle body paint surface is projected onto the vehicle body paint surface using the spatial rectangular coordinate system as the reference system. Corresponding to the three-dimensional model surface of the car body paint, construct the three-dimensional result data of the car body paint defect detection and display the output through a display device; S73, the vehicle body paint surface defect detection evaluation data Pushed to the vehicle body quality inspection management center through the Internet of Things communication network.

9. A real-time monitoring system for automobile spraying defects at the user end of the Internet of Things, used to implement a real-time monitoring method for automobile spraying defects at the user end of the Internet of Things according to any one of claims 1 to 8, characterized in that: The system comprises a vehicle body inspection point construction module, a vehicle body inspection point paint surface defect analysis module, and a vehicle body inspection point paint surface defect monitoring and display module.

Images