Map quality intelligent detection method based on multi-modal perception

Abstract

The application discloses a map quality intelligent detection method based on multi-modal perception and relates to the technical field of defect detection, which comprises the following steps: an environment factor detection module is used to acquire map control parameters associated with environment factors; a closed-loop detection optimization method is constructed based on the map control parameters associated with the environment factors; finally, when the map device is running, the map process is intervened, and the map quality detection result is acquired after the intervention; the application is used to solve the problems that, in the existing map quality intelligent detection method, in the joint analysis of environment and process, secondary re-inspection after the regulation and intervention of the environment and quality comparison before and after are lacked, so that the map defect causes cannot be traced back, the quality detection result is one-sided, and the map quality improvement effect cannot be quantitatively evaluated.

Description

Technical Field

[0001] This invention relates to the field of defect detection technology, specifically to an intelligent detection method for texture quality based on multimodal perception. Background Technology

[0002] Texture mapping refers to the process of precisely attaching materials with patterns, logos, or functional coatings to the surface of a product using automated equipment. It is a high-precision surface decoration or information marking technology in physical production. Intelligent texture mapping quality inspection refers to the process of automatically identifying and evaluating whether the texture mapping effect of the product surface pattern meets the quality standards using artificial intelligence and machine vision technology.

[0003] Existing methods for intelligent texture quality inspection typically rely on visual inspection technology and feature comparison to perform flow-through inspections throughout the entire product manufacturing process. This allows for the acquisition of abnormal status information during production, thereby obtaining texture quality inspection results. While this improved method can replace manual inspection and increase the efficiency of single-piece inspection, it lacks comprehensive environmental and process analysis. It only performs single-time defect identification based on environmental data during the production process, lacking secondary inspections and before-and-after quality comparisons after environmental control and intervention. This leads to problems such as the inability to trace the causes of texture defects, biased quality inspection results, and the inability to quantify the effectiveness of texture quality improvement. For example, in publication number CN1208... Patent application 69255A discloses a method for inspecting the quality of textured films. This method uses a streamlined inspection process and a feature database to automatically match inspection parameters and record various abnormal status information during inspection, facilitating subsequent traceability. Other methods for intelligent inspection of texture quality typically improve the accuracy of defect identification. However, they lack the ability to conduct secondary inspections and compare the quality before and after the intervention of environmental control in the joint analysis of environment and process. This results in problems such as the inability to trace the causes of texture defects, one-sided quality inspection results, and the inability to quantify the improvement effect of texture quality. Therefore, it is necessary to improve the existing intelligent inspection methods for texture quality. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art by proposing a multimodal perception-based intelligent texture quality detection method. This method addresses the shortcomings of existing intelligent texture quality detection methods in the joint analysis of environment and process. These methods lack secondary inspection and before-and-after quality comparison after environmental control and intervention, resulting in the inability to trace the causes of texture defects, biased quality detection results, and the inability to quantify and evaluate the texture quality improvement effect.

[0005] To achieve the above objectives, this application provides a multimodal perception-based intelligent texture quality detection method, comprising the following steps: Based on the environment in which the texturing device is located, an environmental factor detection module is placed around the texturing device, and texturing control parameters associated with environmental factors are obtained based on the environmental factor detection module. Based on the texture control parameters associated with environmental factors, a closed-loop detection optimization method is constructed. When the texturing device is running, the texturing process is intervened based on the texturing control parameters associated with environmental factors and the closed-loop detection optimization method, and the texturing quality detection results are obtained after the intervention.

[0006] Furthermore, based on the environment in which the texturing device is located, an environmental factor detection module is placed around the texturing device, and texturing control parameters associated with environmental factors are obtained based on the environmental factor detection module, including: An environmental factor detection module is placed in the texture area of ​​the texture device. The environmental factor detection module includes a temperature sensor, a humidity sensor, and a wind speed sensor. Devices in the environment where the texture device is located that can control the temperature, humidity, and wind speed are referred to as environmental factor control devices. The bonding pressure, running speed, and heating temperature of the texturing device during the texturing process in the texturing area are respectively recorded as the first texturing parameter, the second texturing parameter, and the third texturing parameter.

[0007] Furthermore, the mapping control parameters associated with environmental factors obtained based on the environmental factor detection module also include: When the mapping device is in normal operation, the environmental factor detection module performs environmental detection on the mapping area for a duration of T. The closed interval formed by the minimum and maximum temperature values ​​in the detection results is denoted as [W]. min W max The closed interval formed by the minimum and maximum humidity values ​​in the detection results is denoted as [S]. min S max The closed interval formed by the minimum and maximum wind speed values ​​in the detection results is denoted as [F]. min F max ]; Without activating the environmental factor control equipment, use the visual film quality inspection equipment to perform quality inspection on the texture in the discharge area under normal operating conditions, and record all types of defects in the inspection results as normal operating defects; for any type of normal operating defect, record the total area occupied by the normal operating defect on the texture surface and the number of independent defects as the defect coverage area and defect coverage number of the normal operating defect, respectively.

[0008] Furthermore, the mapping control parameters associated with environmental factors obtained based on the environmental factor detection module also include: Perform k parameter correlation simulations and obtain k sets of texture control parameters associated with environmental factors based on the results of the parameter correlation simulations; Establish a spatial coordinate system with the X-axis unit of ℃, the Y-axis unit of %RH, and the Z-axis unit of m / s, and denote it as the parameter positioning coordinate system; for any parameter correlation simulation: denote the points in the parameter positioning coordinate system whose x-coordinate, y-coordinate, and Z-axis coordinates are respectively Wt, St, and Ft in the parameter correlation simulation as adjustable points; Parameter correlation simulation includes: in [W min W max ]、[S min S max ] and [F min F max Randomly obtain one parameter from each of the following parameters and record them as Wt, St, and Ft; activate the environmental factor control device and adjust the temperature, humidity, and wind speed of the texture area to Wt, St, and Ft, respectively; Start the texture mapping equipment and use the visual texture mapping quality inspection equipment to perform quality inspection on the texture mapping in the discharge area. Record all types of defects in the inspection results as simulated operation defects. For any type of simulated operation defect, record the total area occupied by the simulated operation defect on the texture mapping surface and the number of independent simulated operation defects as the simulated defect area and the simulated defect number, respectively.

[0009] Furthermore, parameter correlation simulation also includes: For any simulated operational defect, the minimum value between the simulated defect area and the defect coverage area is recorded as the target adjustment area, and the minimum value between the defect coverage number and the simulated defect number is recorded as the target adjustment number. Obtain the target adjustment area and target adjustment quantity for all simulated operational defects.

[0010] Furthermore, parameter correlation simulation also includes: The allowable adjustment ranges of the first texture parameter, the second texture parameter, and the third texture parameter in the texture device are obtained respectively, and are denoted as the first parameter range, the second parameter range, and the third parameter range respectively; Perform steps V1 to V5; Step V1: Randomly select a value from the first parameter interval, the second parameter interval, and the third parameter interval, and record them as the first target value, the second target value, and the third target value, respectively; Step V2: Adjust the first, second, and third mapping parameters in the film application device to the first, second, and third target values, respectively; Step V3: Start the mapping equipment and use the visual mapping quality inspection equipment to inspect the mapping in the output area. Record all defects in the inspection results as adjustment operation defects. If any adjustment operation defect is not recorded as a simulation operation defect, repeat steps V1 to V5. Step V4: For any adjustment operation defect, the total area occupied by the adjustment operation defect on the texture surface and the number of independent defects are respectively recorded as the adjustment defect area and the adjustment defect number. When the adjustment defect area is less than the target adjustment area of ​​the adjustment operation defect and the adjustment defect number is less than the target adjustment number of the adjustment operation defect, the adjustment operation defect is recorded as a perfection defect. Step V5: When all operational defects are recorded as improvement defects, the first target value, the second target value, and the third target value are recorded as texture control parameters associated with the environmental factors [Wt, St, Ft]. When any operational defect is not recorded as a improvement defect, steps V1 to V5 are executed again.

[0011] Furthermore, the loop closure detection optimization method includes: When performing texture quality inspection on the texture device, the textures in the texture area of ​​the texture device that are being textured are recorded as the pre-analysis textures; when the pre-analysis textures move from the texture area to the discharge area, the detection results of temperature, humidity and wind speed by the environmental factor detection module are recorded as real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed, respectively. When the pre-analysis texture is removed from the discharge area, a visual texture quality inspection device is used to inspect the quality of the pre-analysis texture, and all types of defects in the inspection results are recorded as pre-analysis defects. When the number of defects in the current analysis is 0, the texture quality test result is recorded as good.

[0012] Furthermore, the loop closure detection optimization method also includes: When the number of pre-analysis defects is greater than 0, for any pre-analysis defect, the total area of ​​the area occupied by the pre-analysis defect on the texture surface and the number of independent pre-analysis defects are respectively recorded as the pre-analysis defect area and the pre-analysis defect number. In the parameter positioning coordinate system, obtain the points with the horizontal, vertical and Z-axis coordinates of real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed respectively, and record them as real-time intervention points; record the controllable point in the parameter positioning coordinate system that is closest to the real-time intervention point as the parameter control point, and record the horizontal, vertical and Z-axis coordinates of the parameter control point as X1, Y1 and Z1 respectively.

[0013] Furthermore, the loop closure detection optimization method also includes: The bonding pressure, running speed, and heating temperature of the mapping device during the mapping process in the mapping area are adjusted to mapping control parameters associated with environmental factors [X1, Y1, Z1], and the mapping completed in the mapping area after the bonding pressure, running speed, and heating temperature are adjusted is recorded as the closed-loop detection mapping. The quality of the closed-loop detection texture is inspected using a visual texture quality inspection device. When there are defects in the inspection results that are not related to the defects in the previous analysis, the texture quality inspection result is recorded as poor quality and abnormal optimization adjustment. When all defects in the quality inspection results are pre-analysis defects, the defects within the closed-loop detection map are recorded as closed-loop analysis defects. For any closed-loop analysis defect, when the area of ​​the closed-loop analysis defect within the adjustment defect area is smaller than the area of ​​the pre-defect of the closed-loop analysis defect, and the number of independent closed-loop analysis defects within the adjustment defect area is smaller than the number of pre-defects of the closed-loop analysis defect, the closed-loop analysis defect is recorded as a closed-loop optimization defect.

[0014] Furthermore, the loop closure detection optimization method also includes: When all closed-loop analysis defects are recorded as closed-loop optimization defects, the texture quality detection result is recorded as poor quality, can be optimized and has been optimized and adjusted. When there are closed-loop analysis defects that are not recorded as closed-loop optimization defects, the texture quality test result is recorded as poor quality, can be optimized but not fully optimized. When none of the closed-loop analysis defects are recorded as closed-loop optimization defects, the texture quality test result is recorded as poor quality and cannot be optimized.

[0015] The beneficial effects of this invention are as follows: Firstly, based on the environment in which the texturing device is located, an environmental factor detection module is placed around the texturing device, and texturing control parameters associated with environmental factors are obtained based on the environmental factor detection module. The advantage of this is that by obtaining the texturing control parameters based on the environmental factor detection module, parameters that can control the texturing device and effectively optimize defects in the texturing can be obtained when the environment in which the texturing device is located has different temperatures, humidity, and wind speeds. This allows the texturing device to be adjusted using the corresponding texturing control parameters based on the environment in which the texturing device is located during subsequent analysis when defects exist in the texturing in the discharge area, thereby providing comparative data for secondary re-inspection and before-and-after quality comparison. This application also constructs a closed-loop detection optimization method based on texture control parameters associated with environmental factors. Finally, when the texture mapping device is running, the texture mapping process is intervened based on the texture control parameters associated with environmental factors and the closed-loop detection optimization method, and texture quality detection results are obtained after the intervention. The advantage of this is that by constructing a closed-loop detection optimization method, the texture mapping device can be controlled based on environmental parameters when detecting texture quality, and secondary reconstruction and before-and-after quality comparison can be performed after control, thereby obtaining more comprehensive texture quality detection results, providing direction for obtaining the causes of texture defects, and achieving the effect of texture quality improvement. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of the method of the present invention; Figure 2 This is a flowchart of steps V1 to V5 of the present invention; Figure 3 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0017] 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.

[0018] Example 1, please refer to Figure 1 As shown, this application provides an intelligent texture quality detection method based on multimodal perception, including the following steps: Step S1: Based on the environment in which the texturing device is located, place an environmental factor detection module around the texturing device, and obtain the texturing control parameters associated with the environmental factors based on the environmental factor detection module. Step S1 includes: Step S101, placing an environmental factor detection module in the mapping area of ​​the mapping device, wherein the environmental factor detection module includes a temperature sensor, a humidity sensor and a wind speed sensor; the devices in the environment where the mapping device is located that can control the temperature, humidity and wind speed are denoted as environmental factor control devices. Step S102: The bonding pressure, running speed and heating temperature of the mapping device when performing the mapping process in the mapping area are recorded as the first mapping parameter, the second mapping parameter and the third mapping parameter, respectively.

[0019] Step S1 further includes: Step S103, when the mapping device is in normal operation, the environmental factor detection module performs environmental detection on the mapping area for a duration of T, and the closed interval formed by the minimum and maximum temperature values ​​in the detection results is denoted as [W]. min W max The closed interval formed by the minimum and maximum humidity values ​​in the detection results is denoted as [S]. min S max The closed interval formed by the minimum and maximum wind speed values ​​in the detection results is denoted as [F]. min F max ]; In practical implementation, the value of T can be determined based on the fluctuation of environmental parameters within the space where the mapping device is located. If the fluctuation of environmental parameters within the space where the mapping device is located is large, the value of T can be increased to ensure [W min W max ]、[S min S max ] and [F min F max It can effectively reflect the temperature, humidity and wind speed variation range in the environment where the mapping device is located, thus making the subsequent data analysis more comprehensive; in the data analysis of this embodiment, the value of T is set to 24h; Step S104: Without starting the environmental factor control equipment, use the visual film quality inspection equipment to perform quality inspection on the texture in the discharge area under normal operation, and record all types of defects in the inspection results as normal operation defects; for any type of normal operation defect, record the total area occupied by the normal operation defect on the texture surface and the number of independent defects as the defect coverage area and defect coverage number of the normal operation defect, respectively. In the data analysis of this embodiment, for example, during a single data analysis, after performing an environmental detection on the texture area for 24 hours using the environmental factor detection module, the obtained [W] min W max ]、[S min S max ] and [F min F max The values ​​were [18℃, 26℃], [40%RH, 60%RH], and [0.2m / s, 0.8m / s], respectively. When identifying defects during normal operation, one of the defects was an air bubble. Using a visual film quality inspection device, the defect coverage area and number of air bubbles were found to be 10cm² and 15, respectively. In the specific implementation process, the visual film quality inspection equipment can include 2D vision, structured light, line scan camera and AI detection, so as to effectively detect defects such as bubbles, edge lifting, wrinkles, glue overflow and bonding misalignment, and directly output the defect category, physical coordinates, defect area outline, defect area and the location of the defect area.

[0020] Step S1 also includes: Step S104, performing k parameter correlation simulations, and obtaining k sets of texture control parameters associated with environmental factors based on the results of the parameter correlation simulations; In the specific implementation process, the value of k can be determined according to the actual data analysis capability. Alternatively, after performing k parameter correlation simulations, k parameter correlation simulations can be performed again at intervals to make the obtained texture control parameters related to environmental factors more consistent with the environment of the current texture device. In the data analysis of this embodiment, the value of k is set to 20. Step S105: Establish a spatial coordinate system with X-axis unit of ℃, Y-axis unit of %RH, and Z-axis unit of m / s, and denote it as the parameter positioning coordinate system; For any parameter correlation simulation: denote the points in the parameter positioning coordinate system whose x-coordinate, y-coordinate, and Z-axis coordinates are respectively Wt, St, and Ft in the parameter correlation simulation as adjustable points; Step S106, parameter correlation simulation includes: Step S1061, in [W min W max ]、[S min S max ] and [F min F max Randomly obtain one parameter from each of the following parameters and record them as Wt, St, and Ft; activate the environmental factor control device and adjust the temperature, humidity, and wind speed of the texture area to Wt, St, and Ft, respectively; In the data analysis of this embodiment, for example, in a parameter correlation simulation, the obtained Wt, St and Ft are 20℃, 50%RH and 0.4m / s respectively. After starting the environmental factor control device, the environmental factor control device should be used to adjust the temperature, humidity and wind speed of the texture area to 20℃, 50%RH and 0.4m / s respectively. Step S1062: Start the mapping equipment and use the visual mapping quality inspection equipment to perform quality inspection on the mapping in the output area. Record all types of defects in the inspection results as simulated operation defects. For any simulated operation defect, record the total area occupied by the simulated operation defect on the mapping surface and the number of independent simulated operation defects as the simulated defect area and the simulated defect number, respectively.

[0021] The parameter correlation simulation also includes: step S1063, for any kind of simulated operation defect, the minimum value of the simulated defect area and the defect coverage area of ​​the simulated operation defect is recorded as the target adjustment area, and the minimum value of the defect coverage number and the simulated defect number of the simulated operation defect is recorded as the target adjustment number. In the data analysis of this embodiment, for example, in a parameter correlation simulation where Wt, St, and Ft are 20℃, 50%RH, and 0.4m / s respectively, the defect coverage area and defect coverage number for the simulated operational defect "bubble" are 23cm² and 10 respectively. Since the defect coverage area and defect coverage number for the bubble obtained in the above analysis are 10cm² and 15 respectively, the data analysis shows that the target adjustment area and target adjustment number for the "bubble" are 10cm² and 10 respectively. That is, after adjusting the first, second, and third texture parameters, the total area and the number of independent "bubbles" on the texture surface in the obtained texture should be less than 10cm² and 10 respectively. This shows that the defective "bubble" can be optimized by adjusting the first, second, and third texture parameters. Step S1064: Obtain the target adjustment area and target adjustment quantity for all simulated operational defects.

[0022] The parameter association simulation also includes: step S1065, obtaining the adjustable ranges of the first texture parameter, the second texture parameter, and the third texture parameter in the texture device, and recording them as the first parameter range, the second parameter range, and the third parameter range, respectively; For step S1066, please refer to [link / reference]. Figure 2 As shown, execute steps V1 to V5; Step V1: Randomly select a value from the first parameter interval, the second parameter interval, and the third parameter interval, and record them as the first target value, the second target value, and the third target value, respectively; Step V2: Adjust the first, second, and third mapping parameters in the film application device to the first, second, and third target values, respectively; Step V3: Start the mapping equipment and use the visual mapping quality inspection equipment to inspect the mapping in the output area. Record all defects in the inspection results as adjustment operation defects. If any adjustment operation defect is not recorded as a simulation operation defect, repeat steps V1 to V5. In the data analysis of this embodiment, if any adjustment operation defect is not recorded as a simulated operation defect, it means that after adjusting the first, second, and third mapping parameters in the film application device, new defects appear in the completed mapping. This indicates that after adjusting the first, second, and third mapping parameters to the current first, second, and third target values, the defects were not effectively optimized, but instead new defects appeared in the mapping. Therefore, step V1 should be restarted directly to obtain a new set of first, second, and third target values. Step V4: For any adjustment operation defect, the total area occupied by the adjustment operation defect on the texture surface and the number of independent defects are respectively recorded as the adjustment defect area and the adjustment defect number. When the adjustment defect area is less than the target adjustment area of ​​the adjustment operation defect and the adjustment defect number is less than the target adjustment number of the adjustment operation defect, the adjustment operation defect is recorded as a perfection defect. Step V5: When all operational defects are recorded as improvement defects, the first target value, the second target value, and the third target value are recorded as texture control parameters associated with the environmental factors [Wt, St, Ft]. When any operational defect is not recorded as a improvement defect, steps V1 to V5 are re-executed. In the data analysis of this embodiment, for example, in a parameter correlation simulation where Wt, St, and Ft are 20℃, 50%RH, and 0.4m / s respectively, when the first target value, the second target value, and the third target value are 0.5MPa, 5m / min, and 50℃ respectively, the adjustment defect area and the number of adjustment defects of the "bubble" are 5cm² and 2 respectively. From the data obtained above, it can be seen that the adjustment defect area of ​​the bubble is smaller than the target adjustment area, and the number of adjustment defects is smaller than the target adjustment number. This indicates that by adjusting the first texture parameter, the second texture parameter, and the third texture parameter to 0.5MPa, 5m / min, and 50℃ respectively... After reaching a certain temperature, bubbles in the texture can be effectively optimized, so these bubbles can be considered as defects that improve the texture. If, after analyzing the data of other defects, all operational defects are considered as defects that improve the texture, it means that after adjusting the first, second, and third texture parameters to 0.5MPa, 5m / min, and 50℃, defects in the texture can be effectively optimized when the ambient temperature, humidity, and wind speed are 20℃, 50%RH, and 0.4m / s, respectively. Therefore, 0.5MPa, 5m / min, and 50℃ can be considered as texture control parameters associated with environmental factors [20℃, 50%RH, 0.4m / s].

[0023] Step S2: Based on the texture control parameters associated with environmental factors, construct a closed-loop detection optimization method; The closed-loop detection optimization method includes: step S201, when performing texture quality detection on the texture device, the texture in the texture area of ​​the texture device that is in the texture-being-applied state is recorded as the pre-analysis texture; when the pre-analysis texture moves from the texture area to the discharge area, the detection results of temperature, humidity and wind speed by the environmental factor detection module are recorded as real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed, respectively. Step S202: When the pre-analysis texture is removed from the discharge area, a visual film quality inspection device is used to inspect the quality of the pre-analysis texture, and all types of defects in the inspection results are recorded as pre-analysis defects. Step S203: When the number of defects in the current analysis is 0, the texture quality detection result is recorded as good.

[0024] The closed-loop detection optimization method also includes: step S204, when the number of pre-analysis defects is greater than 0, for any pre-analysis defect, the total area of ​​the area occupied by the pre-analysis defect on the texture surface and the number of independent pre-analysis defects are respectively recorded as the pre-analysis defect area and the pre-analysis defect number. Step S205: Obtain the points in the parameter positioning coordinate system whose horizontal coordinate, vertical coordinate, and Z-axis coordinate are respectively the real-time intervention temperature, real-time intervention humidity, and real-time intervention wind speed, and record them as real-time intervention points; record the controllable point in the parameter positioning coordinate system that is closest to the real-time intervention point as the parameter control point, and record the horizontal coordinate, vertical coordinate, and Z-axis coordinate of the parameter control point as X1, Y1, and Z1, respectively. In the specific implementation process, the real-time intervention point can be connected to all adjustable points respectively. The adjustable point corresponding to the shortest line segment among all the line segments obtained after connection is the adjustable point closest to the real-time intervention point. By obtaining the parameter adjustment point, the data that is closest to the real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed in the existing data can be obtained, so that the subsequent texture control parameters are in line with the environment of the current texture device.

[0025] The closed-loop detection optimization method also includes: step S206, adjusting the bonding pressure, running speed and heating temperature of the mapping device when performing the mapping process in the mapping area to the mapping control parameters associated with the environmental factors [X1, Y1, Z1], and recording the mapping completed in the mapping area after the bonding pressure, running speed and heating temperature are adjusted as the closed-loop detection mapping; Step S207: Use a visual film quality inspection device to perform quality inspection on the closed-loop detection texture. When there are defects in the inspection results that are not related to the defects in the previous analysis, record the texture quality inspection result as poor quality and abnormal optimization adjustment. Step S208: When all defects in the quality inspection results are pre-analysis defects, the defects in the closed-loop detection map are recorded as closed-loop analysis defects; for any closed-loop analysis defect, when the area of ​​the closed-loop analysis defect in the adjustment defect area is smaller than the area of ​​the pre-defect of the closed-loop analysis defect, and the number of independent closed-loop analysis defects in the adjustment defect area is smaller than the number of pre-defects of the closed-loop analysis defect, the closed-loop analysis defect is recorded as a closed-loop optimization defect. In the data analysis of this embodiment, for example, during a single data analysis, for the closed-loop analysis defect "bubble", the data analysis shows that the area and number of preceding defects corresponding to the bubble are 8 cm² and 5, respectively. If the area of ​​the bubble within the adjusted defect area and the number of bubbles existing independently within the adjusted defect area are 10 cm² and 3, it indicates that even if the number of bubbles in the texture is reduced, the area occupied by the bubbles in the texture is larger. Therefore, the control of the texture device through the texture control parameters fails to effectively optimize the bubbles in the texture, and the bubbles cannot be recorded as closed-loop optimization defects. If there are closed-loop analysis defects that are not recorded as closed-loop optimization defects, it indicates that the defects in the texture can be optimized, but not completely optimized. The closed-loop analysis defects that are not recorded as closed-loop optimization defects can be used as the direction for subsequent optimization and the causes of defects can be further analyzed.

[0026] The closed-loop detection optimization method also includes: step S209, when all closed-loop analysis defects are recorded as closed-loop optimization defects, the texture quality detection result is recorded as poor quality, can be optimized and has been optimized and adjusted. Step S210: When there is a closed-loop analysis defect that is not recorded as a closed-loop optimization defect, the texture quality detection result is recorded as poor quality, can be optimized but not fully optimized. Step S211: When all closed-loop analysis defects are not recorded as closed-loop optimization defects, the texture quality detection result is recorded as poor quality and cannot be optimized.

[0027] Step S3: When the texture mapping device is running, the texture mapping process is intervened based on the texture mapping control parameters associated with environmental factors and the closed-loop detection optimization method, and the texture mapping quality detection results are obtained after the intervention.

[0028] Example 2, please refer to Figure 3 As shown, Figure 3A schematic diagram of an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps such as those in a multimodal perception-based intelligent texture quality detection method to achieve the following functions: First, based on the environment in which the texture device is located, an environmental factor detection module is placed around the texture device, and texture control parameters associated with the environmental factors are obtained based on the environmental factor detection module; then, based on the texture control parameters associated with the environmental factors, a closed-loop detection optimization method is constructed; finally, when the texture device is running, the texture process is intervened based on the texture control parameters associated with the environmental factors and the closed-loop detection optimization method, and texture quality detection results are obtained after the intervention.

[0029] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0030] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the intelligent texture quality detection method based on multimodal perception provided by the above methods. The method includes: first, based on the environment in which the texture device is located, placing an environmental factor detection module around the texture device, and obtaining texture control parameters associated with the environmental factors based on the environmental factor detection module; then, constructing a closed-loop detection optimization method based on the texture control parameters associated with the environmental factors; finally, when the texture device is running, intervening in the texture process based on the texture control parameters associated with the environmental factors and the closed-loop detection optimization method, and obtaining texture quality detection results after intervention.

[0031] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-described intelligent texture quality detection method based on multimodal perception to achieve the following functions: First, based on the environment in which the texture device is located, an environmental factor detection module is placed around the texture device, and texture control parameters associated with environmental factors are obtained based on the environmental factor detection module; then, based on the texture control parameters associated with environmental factors, a closed-loop detection optimization method is constructed; finally, when the texture device is running, the texture process is intervened based on the texture control parameters associated with environmental factors and the closed-loop detection optimization method, and texture quality detection results are obtained after the intervention.

[0032] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0033] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A texture quality intelligent detection method based on multimodal perception, characterized in that, Includes the following steps: Based on the environment in which the texturing device is located, an environmental factor detection module is placed around the texturing device, and texturing control parameters associated with environmental factors are obtained based on the environmental factor detection module. Based on the texture control parameters associated with environmental factors, a closed-loop detection optimization method is constructed. When the texturing device is running, the texturing process is intervened based on the texturing control parameters associated with environmental factors and the closed-loop detection optimization method, and the texturing quality detection results are obtained after the intervention.

2. The intelligent texture quality detection method based on multimodal perception according to claim 1, characterized in that, Based on the environment in which the texturing device is located, an environmental factor detection module is placed around the texturing device, and the texturing control parameters associated with the environmental factors are obtained based on the environmental factor detection module, including: An environmental factor detection module is placed in the texture area of ​​the texture device. The environmental factor detection module includes a temperature sensor, a humidity sensor, and a wind speed sensor. Devices in the environment where the texture device is located that can control the temperature, humidity, and wind speed are referred to as environmental factor control devices. The bonding pressure, running speed, and heating temperature of the texturing device during the texturing process in the texturing area are respectively recorded as the first texturing parameter, the second texturing parameter, and the third texturing parameter.

3. The intelligent texture quality detection method based on multimodal perception according to claim 2, characterized in that, The texture control parameters associated with environmental factors obtained based on the environmental factor detection module also include: When the mapping device is in normal operation, the environmental factor detection module performs environmental detection on the mapping area for a duration of T. The closed interval formed by the minimum and maximum temperature values ​​in the detection results is denoted as [W]. min W max The closed interval formed by the minimum and maximum humidity values ​​in the detection results is denoted as [S]. min S max The closed interval formed by the minimum and maximum wind speed values ​​in the detection results is denoted as [F]. min F max ]; Without activating the environmental factor control equipment, use the visual film quality inspection equipment to perform quality inspection on the texture in the discharge area under normal operating conditions, and record all types of defects in the inspection results as normal operating defects; for any type of normal operating defect, record the total area occupied by the normal operating defect on the texture surface and the number of independent defects as the defect coverage area and defect coverage number of the normal operating defect, respectively.

4. The intelligent texture quality detection method based on multimodal perception according to claim 3, characterized in that, The texture control parameters associated with environmental factors obtained based on the environmental factor detection module also include: Perform k parameter correlation simulations and obtain k sets of texture control parameters associated with environmental factors based on the results of the parameter correlation simulations; Establish a spatial coordinate system with the X-axis unit of ℃, the Y-axis unit of %RH, and the Z-axis unit of m / s, and denote it as the parameter positioning coordinate system; for any parameter correlation simulation: denote the points in the parameter positioning coordinate system whose x-coordinate, y-coordinate, and Z-axis coordinates are respectively Wt, St, and Ft in the parameter correlation simulation as adjustable points; Parameter correlation simulation includes: in [W min W max ]、[S min S max ] and [F min F max Randomly obtain one parameter from each of the following parameters and record them as Wt, St, and Ft; activate the environmental factor control device and adjust the temperature, humidity, and wind speed of the texture area to Wt, St, and Ft, respectively; Start the texture mapping equipment and use the visual texture mapping quality inspection equipment to perform quality inspection on the texture mapping in the discharge area. Record all types of defects in the inspection results as simulated operation defects. For any type of simulated operation defect, record the total area occupied by the simulated operation defect on the texture mapping surface and the number of independent simulated operation defects as the simulated defect area and the simulated defect number, respectively.

5. The intelligent texture quality detection method based on multimodal perception according to claim 4, characterized in that, Parameter correlation simulation also includes: For any simulated operational defect, the minimum value between the simulated defect area and the defect coverage area is recorded as the target adjustment area, and the minimum value between the defect coverage number and the simulated defect number is recorded as the target adjustment number. Obtain the target adjustment area and target adjustment quantity for all simulated operational defects.

6. The intelligent texture quality detection method based on multimodal perception according to claim 5, characterized in that, Parameter correlation simulation also includes: The allowable adjustment ranges of the first texture parameter, the second texture parameter, and the third texture parameter in the texture device are obtained respectively, and are denoted as the first parameter range, the second parameter range, and the third parameter range respectively; Perform steps V1 to V5; Step V1: Randomly select a value from the first parameter interval, the second parameter interval, and the third parameter interval, and record them as the first target value, the second target value, and the third target value, respectively; Step V2: Adjust the first, second, and third mapping parameters in the film application device to the first, second, and third target values, respectively; Step V3: Start the mapping equipment and use the visual mapping quality inspection equipment to inspect the mapping in the output area. Record all defects in the inspection results as adjustment operation defects. If any adjustment operation defect is not recorded as a simulation operation defect, repeat steps V1 to V5. Step V4: For any adjustment operation defect, the total area occupied by the adjustment operation defect on the texture surface and the number of independent defects are respectively recorded as the adjustment defect area and the adjustment defect number. When the adjustment defect area is less than the target adjustment area of ​​the adjustment operation defect and the adjustment defect number is less than the target adjustment number of the adjustment operation defect, the adjustment operation defect is recorded as a perfection defect. Step V5: When all operational defects are recorded as improvement defects, the first target value, the second target value, and the third target value are recorded as texture control parameters associated with the environmental factors [Wt, St, Ft]. When any operational defect is not recorded as a improvement defect, steps V1 to V5 are executed again.

7. The intelligent texture quality detection method based on multimodal perception according to claim 6, characterized in that, Optimization methods for closed-loop detection include: When performing texture quality inspection on the texture device, the textures in the texture area of ​​the texture device that are being textured are recorded as the pre-analysis textures; when the pre-analysis textures move from the texture area to the discharge area, the detection results of temperature, humidity and wind speed by the environmental factor detection module are recorded as real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed, respectively. When the pre-analysis texture is removed from the discharge area, a visual texture quality inspection device is used to inspect the quality of the pre-analysis texture, and all types of defects in the inspection results are recorded as pre-analysis defects. When the number of defects in the current analysis is 0, the texture quality test result is recorded as good.

8. The intelligent texture quality detection method based on multimodal perception according to claim 7, characterized in that, The closed-loop detection optimization methods also include: When the number of pre-analysis defects is greater than 0, for any pre-analysis defect, the total area of ​​the area occupied by the pre-analysis defect on the texture surface and the number of independent pre-analysis defects are respectively recorded as the pre-analysis defect area and the pre-analysis defect number. In the parameter positioning coordinate system, obtain the points with the horizontal, vertical and Z-axis coordinates of real-time intervention temperature, real-time intervention humidity and real-time intervention wind speed respectively, and record them as real-time intervention points; record the controllable point in the parameter positioning coordinate system that is closest to the real-time intervention point as the parameter control point, and record the horizontal, vertical and Z-axis coordinates of the parameter control point as X1, Y1 and Z1 respectively.

9. The intelligent texture quality detection method based on multimodal perception according to claim 8, characterized in that, The closed-loop detection optimization methods also include: The bonding pressure, running speed, and heating temperature of the mapping device during the mapping process in the mapping area are adjusted to mapping control parameters associated with environmental factors [X1, Y1, Z1], and the mapping completed in the mapping area after the bonding pressure, running speed, and heating temperature are adjusted is recorded as the closed-loop detection mapping. The quality of the closed-loop detection texture is inspected using a visual texture quality inspection device. When there are defects in the inspection results that are not related to the defects in the previous analysis, the texture quality inspection result is recorded as poor quality and abnormal optimization adjustment. When all defects in the quality inspection results are pre-analysis defects, the defects within the closed-loop detection map are recorded as closed-loop analysis defects. For any closed-loop analysis defect, when the area of ​​the closed-loop analysis defect within the adjustment defect area is smaller than the area of ​​the pre-defect of the closed-loop analysis defect, and the number of independent closed-loop analysis defects within the adjustment defect area is smaller than the number of pre-defects of the closed-loop analysis defect, the closed-loop analysis defect is recorded as a closed-loop optimization defect.

10. The intelligent texture quality detection method based on multimodal perception according to claim 9, characterized in that, The closed-loop detection optimization methods also include: When all closed-loop analysis defects are recorded as closed-loop optimization defects, the texture quality detection result is recorded as poor quality, can be optimized and has been optimized and adjusted. When there are closed-loop analysis defects that are not recorded as closed-loop optimization defects, the texture quality test result is recorded as poor quality, can be optimized but not fully optimized. When none of the closed-loop analysis defects are recorded as closed-loop optimization defects, the texture quality test result is recorded as poor quality and cannot be optimized.

Images