Aluminum alloy component surface flaw detection device based on visual inspection

Abstract

The invention provides an aluminum alloy component surface flaw detection device based on visual inspection, which relates to the technical field of surface flaw detection and comprises a guard plate, a guide plate, a conveying mechanism, a fixing mechanism, an adjusting mechanism and a flaw detection system. Wherein the flaw detection system comprises an acquisition module, an image processing module, an acquisition and allocation module, a result output module and an environment adaptation module. The environment adaptation module collects illumination and temperature parameters in real time, automatically adjusts image preprocessing parameters and working parameters of a visual detector when the parameters exceed the range, and counteracts environment interference. The acquisition and allocation module automatically calculates the adaptive height of the visual detector and generates an adjustment instruction in combination with the component detection requirement and the state of the adjustment mechanism, and manual adjustment is not needed; and the image processing module accurately identifies the types and positions of the flaws through multi-feature extraction and weighted decision. The device solves the problems that existing equipment is poor in flexible adaptation, unstable in detection precision and prone to misjudgment of flaws, and efficient and accurate detection of the flaws on the surface of the aluminum alloy component is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of surface flaw detection, and in particular to an aluminum alloy component surface flaw detection device based on visual detection. BACKGROUND

[0002] In the fields of high-end manufacturing such as aerospace, automobile manufacturing, and rail transportation, aluminum alloy components have become the preferred material for key structural parts such as aircraft fuselage skins, automobile aluminum alloy hubs, and high-speed train body frames due to their core advantages of "lightweight (density is only 1 / 3 of steel), high strength (tensile strength can reach more than 600 MPa), and corrosion resistance". The surface flaws (such as cracks, scratches, and depressions) of such components directly affect the safety and service life of the products, and the industry's control requirements for the surface quality of aluminum alloy components are becoming increasingly stringent, pushing the surface flaw detection technology based on visual detection to become the mainstream development direction.

[0003] However, existing aluminum alloy component surface flaw detection devices based on visual detection have significant shortcomings: first, the height and detection angle of the visual detector are mostly fixed designs, which can only adapt to components of a single thickness or size. When detecting different specifications of components, manual disassembly and adjustment are required, which takes 40-60 minutes per single changeover, and cannot meet the flexible production needs of "multiple varieties and small batches". Second, there is a lack of environmental adaptation mechanism. Fluctuations in light intensity and temperature changes in the detection area can easily cause image blurring or reduced contrast, resulting in missed or false judgments of flaws, poor detection accuracy and stability, and difficulty in adapting to complex workshop conditions. Therefore, it is necessary to provide an aluminum alloy component surface flaw detection device based on visual detection to solve the above technical problems. SUMMARY

[0004] The present application provides an aluminum alloy component surface flaw detection device based on visual detection to solve the problems raised in the background art.

[0005] To solve the above technical problems, the aluminum alloy component surface flaw detection device based on visual detection provided by the present application comprises: horizontally opposite installed guard plates, one end of which is inclined and fixed with a material guide plate, and a conveying mechanism is installed between the other end of the material guide plate and the guard plates, a fixed mechanism is vertically upwardly installed in the middle of the conveying mechanism, and an adjusting mechanism is installed inside the fixed mechanism, characterized in that it further comprises a flaw detection system; the flaw detection system comprises a collection module, an image processing module, a collection and deployment module, and a result output module. The collection module is used to collect image data of the surface of the aluminum alloy component, as well as operation state data of the conveying mechanism and position state data of the adjusting mechanism. The image processing module is used to process the image data through a preset image analysis algorithm, extract the surface features of the component, and identify the type and position of the flaw to generate a flaw identification result. The collection and adjustment module is used for adjusting the height of the visual detector by a preset parameter adjustment algorithm, combining the detection requirement of the component with the state data, calculating the adaptive height parameter of the visual detector, and generating the action control instruction of the adjusting mechanism.

[0006] Preferably, the lower end of the same side of the guard plate is mounted with two support columns by bolts, one end of the support column is fixed to the bottom plate frame, and the lower end of the bottom plate frame is mounted with four universal wheels in a rectangular shape.

[0007] Preferably, the conveying mechanism comprises a roller rotatably mounted between the guard plates, a conveying belt is sleeved on the roller, a servo motor is mounted on the guard plate at one end of one of the rollers, and the output end of the servo motor is coaxially fixed to the end of the roller through the guard plate.

[0008] Preferably, the fixing mechanism comprises a detection frame fixed to the upper end of the guard plates on both sides, an adjusting cavity is vertically and downwardly formed in the detection frame, a touch screen is mounted on one side of the detection frame close to one end of the guide plate, a baffle is mounted on the guard plate at one end of the guide plate through bolts, and a sealing plate with a self-locking function is mounted on one side of the detection frame at the upper end of the guard plate.

[0009] Preferably, the adjusting mechanism comprises an electric telescopic rod mounted on the upper end of the detection frame, a U-shaped fixing frame is threadedly connected to the output end of the electric telescopic rod through the detection frame, a fixing cylinder is mounted between the fixing frames through bolts, and sliding blocks are symmetrically fixed to the two sides of the fixing cylinder.

[0010] Preferably, a guide rod is vertically and downwardly arranged in each of the two sliding blocks, the top end of the guide rod is fixed to the inner wall of the detection frame, a limiting block is fixed to the bottom end of the guide rod, a cross positioning frame is horizontally mounted on the inner side of the fixing cylinder, a detection hole is formed in the middle of the cross positioning frame, a visual detector is mounted on the upper end of the cross positioning frame, a positioning plate is mounted on the fixing cylinder at the upper end of the visual detector through bolts, and a wire port is formed in one side of the positioning plate.

[0011] Preferably, the image processing module is used for processing the image data by a preset image analysis algorithm, extracting the surface features of the component, and identifying the type and position of the defect, and specifically comprises the following steps: Step one: image adaptive pre-processing, according to the adaptive relationship between the preset standard light intensity and the real-time light intensity of each pixel point of the image, the light compensation is first performed for the strong light reflection characteristics of the surface of the aluminum alloy component, and then the noise in the image is removed through filtering processing; Step two: multi-dimensional feature extraction, synchronously extracting three types of core features of the surface of the component to distinguish the flaw area from the normal surface texture, which are: gray gradient features obtained by gradient calculation, texture entropy features obtained by texture statistical analysis, and geometric moment features obtained by image pixel distribution calculation; Step three: flaw classification and identification, comprehensively judging the extracted multi-dimensional features by a preset weighted feature decision logic in combination with standard feature parameters of the normal aluminum alloy surface to determine whether the image pixel area is a flaw area; If it is determined to be a flaw area, the extracted multi-dimensional features are matched with a preset multi-type flaw feature library to determine the specific type and pixel-level position of the flaw, and finally an identification result containing the flaw type and position boundary information is generated, which is recorded as the flaw identification result.

[0012] Preferably, the process of the acquisition and adjustment module calculating the adaptive height parameter and generating the action control instruction through the preset parameter adjustment algorithm comprises the following steps: Step one: obtaining the component detection requirement related parameters and the state data of the adjustment structure; wherein the component detection requirement related parameters include the characteristic parameters of the aluminum alloy component, the height correlation coefficient matched with the preset detection precision level, and the basic compensation height of the visual detector; the state data of the adjustment structure includes the actual height of the visual detector corresponding to the current position of the adjustment structure; Step two: theoretical adaptive height calculation, combining the component detection requirement related parameters, determining the theoretical adaptive height of the visual detector through parameter coordination calculation, and dynamically adjusting the height correlation coefficient according to the detection precision level during the calculation process; Step three: mechanical state optimization adjustment, optimizing and correcting the theoretical adaptive height in combination with the actual height state of the adjustment structure; Step four: action control instruction generation, determining the action stroke of the adjustment structure according to the difference between the optimized adaptive height and the current height of the adjustment structure, combining the action characteristics of the adjustment structure, determining the action direction according to the size relationship between the final adaptive height and the current height, and finally generating the adjustment structure action control instruction containing the action direction and the action stroke.

[0013] Preferably, the flaw detection system comprises an environment adaptation module, which acquires the environmental parameters of the detection area in real time, automatically adjusts the preprocessing parameters of the image analysis algorithm and the working parameters of the visual detector when the environmental parameters exceed the preset adaptive range, and the environmental parameters include the environmental light intensity and the environmental temperature, and the preprocessing parameters include the image denoising parameters and the contrast enhancement parameters.

[0014] Compared with the related art, the aluminum alloy component surface flaw detection device based on visual detection provided by the present application has the following beneficial effects: 1、The present application solves the problem of flexible production adaptation by collecting and adjusting module combined with component detection requirements and adjusting mechanism state data, automatically calculating the visual detector adaptation height and generating the electric telescopic rod action instruction, without manual disassembly and adjustment, solves the problem of flexible production adaptation, the existing device needs manual adjustment due to the fixed height of the visual detector, and cannot adapt to components of multiple specifications.

[0015] 2、The present application solves the problem of lack of environmental adaptation mechanism in the existing device by real-time collection of environmental parameters by the environmental adaptation module, automatic adjustment of image denoising, contrast enhancement parameters and visual detector exposure time when out of range, offsetting environmental interference, ensuring detection accuracy stability under different working conditions, and improving detection accuracy stability.

[0016] 3、The present application solves the problem of single feature recognition leading to defect misjudgment by first performing adaptive illumination compensation and filter preprocessing on the image, then synchronously extracting three types of core features of gray gradient, texture entropy and geometric moment, and combining normal aluminum alloy surface standard parameters to make a comprehensive judgment through weighted decision logic, which can accurately distinguish between defect areas and normal rolling textures, and improve the accuracy of defect recognition. In summary, the present application solves the problem of lack of flexible production adaptation of the existing device and the problem of unstable accuracy caused by environmental interference through the triple design of automatic height adaptation, environmental dynamic adjustment and multi-feature accurate recognition, and solves the problem of single feature recognition misjudgment, realizes efficient, accurate and stable detection of aluminum alloy component surface defects. BRIEF DESCRIPTION OF DRAWINGS

[0017] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings: Figure 1 The overall three-dimensional structure schematic diagram of the present application is shown in the figure; Figure 2 Another side of the overall three-dimensional structure schematic diagram of the present application is shown in the figure; Figure 3 The overall three-dimensional structure schematic diagram of the present application is shown in the figure; Figure 4 The adjusting mechanism three-dimensional structure schematic diagram of the present application is shown in the figure; Figure 5 The adjusting mechanism three-dimensional structure schematic diagram of the present application is shown in the figure; Figure 6 The principle block diagram of the defect detection system of the present application is shown in the figure.

[0018] The figure sequence number: 1, the guard plate; 2, the support column; 3, the bottom plate frame; 4, servo motor; 5, the conveying belt; 6, the guide plate; 7, the barrier plate; 8, the detection frame; 9, the touch screen; 10, the electric telescopic rod; 11, the closing plate; 12, the guide rod; 13, the fixed frame; 14, the fixed cylinder; 15, the positioning plate; 16, the sliding block; 17, the visual detector; 18, the cross positioning frame; 19, the limiting block. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0020] The terms used in the present disclosure are merely for the purpose of describing specific embodiments and are not intended to limit the present disclosure. The singular forms "a," "an," and "the" used in the present disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0021] It should be understood that although the terms first, second, third, etc. can be used in the present disclosure to describe various information, these information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "when" or "upon determination" or "in response to determining".

[0022] Please refer to Figures 1-6 The aluminum alloy component surface flaw detection device based on visual detection comprises guard plates 1 installed horizontally opposite to each other, a guide plate 6 is fixedly connected between one end of the guard plates 1, a conveying mechanism is installed between the other end of the guide plate 6 and the guard plates 1, and the guard plates 1, the guide plate 6, the conveying mechanism, a fixing mechanism and an adjusting mechanism are used to facilitate the erection of the overall frame of the aluminum alloy component surface flaw detection device and realize the guiding feeding, conveying, fixing and detection position adjusting of the component, thereby laying a foundation for subsequent visual detection. The fixing mechanism is installed vertically upward in the middle of the conveying mechanism, the adjusting mechanism is installed in the fixing mechanism, and the aluminum alloy component surface flaw detection device further comprises a flaw detection system. The acquisition module is used to collect image data of the aluminum alloy component surface, as well as the operating status data of the conveying mechanism and the position status data of the adjusting mechanism. In the acquisition module, the image data of the aluminum alloy component surface is directly captured by the vision detector 17 installed on the cross positioning frame 18; the operating status data of the conveying mechanism is collected through the feedback interface of the servo motor 4 that drives the roller, such as the speed and start / stop status; the position status data of the adjusting mechanism is collected through the position sensor mounted on the sliding block 16, combined with the scale markings of the guide rod 12 or the displacement detection element, such as the real-time displacement of the sliding block 16. The image processing module is used to process image data through a preset image analysis algorithm, extract surface features of components, and identify the type and location of defects in order to generate defect identification results; The data acquisition and adjustment module is used to adjust the algorithm with preset parameters, combine the component inspection requirements and status data, calculate the adaptation height parameters of the vision detector 17, and generate the motion control instructions of the adjustment mechanism; the result output module is used to integrate the defect identification results, inspection time, and mechanical operation parameters into a standardized inspection report and output it.

[0023] Specifically, two support columns 2 are installed at equal intervals on the lower end of the same side guard plate 1 by bolts. One end of the support column 2 is fixed to the base plate frame 3. The lower end of the base plate frame 3 is equipped with four universal wheels in a rectangular shape. The support columns 2, the base plate frame 3, and the universal wheels facilitate stable support of the entire device. At the same time, the universal wheels enable flexible movement of the device to adapt to different testing operation scenarios.

[0024] Specifically, the conveying mechanism includes rollers rotatably mounted between guard plates 1, with a conveyor belt 5 fitted on the rollers. A servo motor 4 is mounted on one end of one of the guard plates 1. The output end of the servo motor 4 passes through the guard plate 1 and is coaxially fixed to one end of the roller. Through the rollers, the conveyor belt 5, and the servo motor 4, the servo motor 4 can drive the conveyor belt 5 to run, thereby driving the aluminum alloy components to be stably conveyed along a set path, providing a conveying guarantee for the continuous detection of surface defects of the components.

[0025] Specifically, the fixing mechanism includes a detection frame 8 fixed to the upper end of the two side guard plates 1. The detection frame 8 has an adjustment cavity that runs vertically downward. A touch screen 9 is installed on one side of the detection frame 8 near the guide plate 6. A baffle plate 7 is installed on the guard plate 1 at one end of the guide plate 6 by bolts. A sealing plate 11 with a self-locking function is installed on one side of the detection frame 8 at the upper end of the guard plate 1. The detection frame 8, touch screen 9, baffle plate 7, and sealing plate 11 facilitate the installation of visual inspection-related components. The touch screen 9 can display detection data and results. The baffle plate 7 restricts the displacement of the component conveying. The sealing plate 11 protects the internal components of the detection frame 8 and prevents external interference.

[0026] Specifically, the adjustment mechanism includes an electric telescopic rod 10 installed on the upper end of the inspection frame 8. The output end of the electric telescopic rod 10 passes through the inspection frame 8 and is connected to a U-shaped fixed frame 13 by threads. Fixed cylinders 14 are installed between the fixed frames 13 by bolts. Sliding blocks 16 are symmetrically fixed on both sides of the fixed cylinders 14. Through the electric telescopic rod 10, fixed frame 13, fixed cylinder 14, and sliding blocks 16, it is convenient to use the electric telescopic rod 10 to drive the fixed cylinder 14 to move up and down, adjust the height of the fixed cylinder 14 and the internal visual inspection components, and adapt to the inspection needs of aluminum alloy components of different thicknesses.

[0027] Specifically, guide rods 12 are vertically installed through both sliding blocks 16. The top ends of the guide rods 12 are fixed to the inner wall of the detection frame 8, and the bottom ends of the guide rods 12 are fixed to limit blocks 19. A cross positioning frame 18 is horizontally installed inside the fixed cylinder 14. A detection hole is opened in the middle of the cross positioning frame 18, and a vision detector 17 is installed on the upper end of the cross positioning frame 18. A positioning plate 15 is bolted to the fixed cylinder 14 above the vision detector 17. A wire opening is opened on one side of the positioning plate 15. Through the guide rods 12, limit blocks 19, cross positioning frames 18, vision detectors 17, positioning plates 15, and wire opening, the guide rods 12 ensure the stable movement of the fixed cylinder 14, the limit blocks 19 prevent excessive movement, the vision detector 17 detects surface defects of the components, the positioning plate 15 fixes the detector, and the wire opening facilitates the threading of connecting wires to ensure stable operation of the detection function.

[0028] Specifically, the image processing module is used to process image data using a preset image analysis algorithm, extract surface features of the component, and identify the type and location of defects. This includes the following steps: Step 1: Adaptive Image Preprocessing. Addressing the strong reflective properties of aluminum alloy components, illumination compensation is first performed based on the relationship between a preset standard illumination intensity and the real-time illumination intensity of each pixel in the image. The expression is: ,in To compensate for the grayscale value at pixel (x,y) in the image after processing, Let be the gray value of the original image at (x, y). To preset the standard light intensity, The real-time illumination intensity at pixel (x,y) is collected by the built-in light sensor of the visual detector 17, and k is a preset anti-zero offset coefficient; this formula eliminates the interference of ambient light fluctuations on image quality. Then, noise in the image is removed through filtering. The filtering expression is as follows: ,in, Let be the gray value of the filtered image at pixel (x, y). is the standard deviation of the Gaussian kernel, m and n are the half-size of the Gaussian kernel, and i and j are the loop variables for the Gaussian filtering summation operation, used to traverse the pixel positions within the Gaussian kernel and calculate the weighted average of the original image pixel and its neighboring pixels to achieve smoothing filtering; this filtering operation ensures the image sharpness. Step Two: Multi-dimensional Feature Extraction. Three core features of the component surface are extracted simultaneously to distinguish defective areas from normal surface textures. These are: Gray-level gradient features obtained through gradient calculation The formula is calculated using the Sobel operator. ,in These are the gradients in the x and y directions, respectively. The gray-scale gradient features can reflect the contour changes of the component surface. Texture entropy features obtained through texture statistical analysis The formula is calculated using the gray-level co-occurrence matrix. Where L is the gray level (value 256). represents the probability of gray values ​​h and g appearing simultaneously in the gray-level co-occurrence matrix. h and g are index variables of the gray level, used to traverse all possible combinations of gray values. The texture entropy feature can reflect the regularity of surface texture. Geometric moment features calculated from image pixel distribution The formula is , where q and p are the order of moments, and W and H are the image width and height, respectively. The geometric moment features reflect the regional morphology of the component surface. Step 3: Defect classification and identification. This involves using a pre-defined weighted feature decision logic, combined with standard feature parameters of a normal aluminum alloy surface (including standard texture entropy, standard geometric moments, etc.). The formula is as follows: Where D is the decision value, , , These are the weights of the grayscale gradient feature, texture entropy feature, and geometric moment feature, respectively, used in the weighted decision function for defect recognition to quantify the influence of each feature on the defect judgment result. The standard texture entropy of a normal aluminum alloy surface. The standard geometric moments of a normal aluminum alloy surface; A comprehensive judgment is made on the extracted multi-dimensional features, and a preset decision threshold T is set. When the region is identified as a defective area, it is then analyzed using its feature vector. The defect is matched with a pre-defined multi-type defect feature library (including feature information corresponding to cracks, scratches, and dents) to determine the defect type and pixel-level location coordinates, and a recognition result containing the defect type and location bounding box is generated. This recognition result is recorded as the defect recognition result.

[0029] Specifically, the process by which the data acquisition and allocation module calculates the adaptation height parameters and generates motion control commands through a preset parameter adjustment algorithm includes the following steps: Step 1: Obtain relevant parameters for component inspection requirements and adjust the structural status data; among which, the relevant parameters for component inspection requirements include the actual thickness of the aluminum alloy component. (Image size conversion acquired by visual detector 17 or input via human-computer interaction component), and a high correlation coefficient matching the preset detection accuracy level. (The higher the detection accuracy, the larger the value of this coefficient, which is determined by the device through calibration using multi-specification aluminum alloy components.) The base compensation height of the vision detector 17. (Used to offset the basic distance between the detection hole and the component surface, and to preset fixed parameters for the device); The status data of the adjustment mechanism includes the actual height of the vision detector 17 corresponding to the current position of the adjustment mechanism. (Data is collected in real time by the displacement detection element mounted on the sliding block 16). Step 2: Calculate the theoretical adaptation height. Based on the relevant parameters of the component inspection requirements, determine the theoretical adaptation height of the vision detector 17 through parameter collaborative calculation. The formula is ,in This represents the theoretical fit height without considering the current mechanical state. The high correlation coefficient is dynamically adjusted based on the detection accuracy level during the calculation process. The detection accuracy is classified into levels (such as industrial grade, automotive grade, and aerospace grade), and a basic high correlation coefficient is preset for each level. Then through the formula ,in For accuracy adjustment coefficients, This represents the deviation between the actual detection accuracy and the accuracy of the grade standard. By dynamically adjusting the height correlation coefficient, we can ensure that the theoretical height is adapted to the characteristics of the component and the detection accuracy requirements, thus meeting the detection clarity requirements in different scenarios. Step 3: Mechanical condition optimization and adjustment. Based on the current actual height of the adjusting mechanism, the theoretical suitable height is optimized and corrected. The final suitable height is calculated using the anti-overshoot optimization formula. The optimized formula is as follows: In the formula, This is a mechanical state coordination coefficient used to reduce the influence of the current position on the optimal height, balancing adjustment speed and accuracy. hour, To be positive in order to reduce the downlink travel, when hour, To reduce the upward stroke, the value is set to positive. By optimizing the mechanical state, the unnecessary stroke or overshoot of the adjustment mechanism is effectively avoided, balancing the adjustment speed and position accuracy, and ensuring that the adjustment mechanism moves from the current position to the target height with the shortest stroke and the most stable movement. Step 4: Generate motion control commands based on the optimized adaptation height. With the current height of the regulating mechanism The difference, combined with the operating characteristics of the regulating mechanism, is used to calculate the operating stroke S of the regulating mechanism through the stroke-command conversion formula, which is: In the formula, The stroke conversion coefficient of the adjustment mechanism is determined by the pulse equivalent of the electric telescopic rod 10. For example, if 1 mm of travel of the electric telescopic rod corresponds to 100 pulses, then... ); At the same time, based on the optimized adaptation height Compared to the current height The size relationship clarifies the direction of the action. When extending the command, in The command is a retraction instruction, which ultimately generates an adjustment mechanism motion control instruction containing the motion direction and motion stroke.

[0030] The above process solves the problems of frequent operation, excessive invalid stroke, and mechanical state interference in the adaptation accuracy of the existing adjustment mechanism by combining component detection requirements with mechanical state dual-dimensional collaborative calculation and anti-overshoot optimization design, thereby improving the efficiency and accuracy of the highly adapted vision detector 17.

[0031] Specifically, the defect detection system includes an environment adaptation module, which collects environmental parameters of the detection area in real time. When the environmental parameters exceed the preset adaptation range, it automatically adjusts the preprocessing parameters of the image analysis algorithm and the working parameters of the visual detector 17. The environmental parameters include ambient light intensity and ambient temperature, and the preprocessing parameters include image denoising parameters and contrast enhancement parameters. The process of collecting environmental parameters of the detection area in real time and dynamically adjusting the relevant parameters includes the following steps: Step 1: Real-time acquisition of environmental parameters. The real-time ambient light intensity of the detection area is collected by the light sensor mounted on the detection frame 8. Real-time ambient temperature It also calls the device's preset environmental parameter adaptation range (including the standard range of light intensity). Temperature standard range (The calibration is determined by multiple environmental conditions before the device leaves the factory). Step Two: Adjusting lighting adaptation and contrast enhancement parameters. When the ambient light intensity exceeds the preset lighting intensity standard range, i.e. Dynamically calculate image contrast enhancement parameters The formula is , This represents the base contrast enhancement factor under standard lighting conditions. To preset the standard light intensity, This parameter is used to adjust the contrast enhancement parameter in the image analysis algorithm. It is used to adjust the degree of contrast enhancement in the image analysis algorithm to ensure that the image contrast is enhanced in low light conditions and that overexposure is suppressed in strong light conditions. Step 3: Temperature adaptation, noise reduction, and adjustment of the operating parameters of the visual detector 17. When the ambient temperature exceeds the preset temperature standard range, i.e. Two adjustments will be made simultaneously: Calculate the standard deviation of the Gaussian filter kernel using the correlation formula between temperature and noise reduction. The formula is ,in, The baseline noise reduction standard deviation at standard temperature. To preset standard temperature, This is a temperature influence coefficient, used to enhance the denoising effect as the temperature deviation increases, thus offsetting temperature-induced image noise. The exposure time of the visual detector 17 is adjusted using a temperature-exposure correlation formula. The formula is ,in The base exposure time at standard temperature. It is the exposure temperature coefficient, used to shorten the exposure time when the temperature rises to avoid thermal noise interference, and to appropriately extend the exposure time when the temperature falls to ensure image brightness. Step 4: Adjust the image preprocessing parameters (including...) The adjusted visual detector 17 is sent to the preprocessing module of the image analysis algorithm. The control unit of the visual detector 17 sends the data to automatically adjust the environmental parameters when they are out of range. The above process, through the correlation between illumination and contrast, and the correlation between temperature, noise reduction and exposure, uses a two-dimensional collaborative adaptation logic to solve the problem that existing technologies only adapt to illumination or ignore the influence of temperature, thus ensuring image quality and detection stability under different environments.

[0032] The working principle proposed in this invention is as follows: When this device is used, the servo motor 4 drives the conveyor belt 5 to transport aluminum alloy components. After being fed by the guide plate 6 and limited by the barrier plate 7, the components enter the detection area. The environmental adaptation module on the detection frame 8 collects light and temperature parameters. When the parameters are out of range, it automatically adjusts the image denoising and contrast enhancement parameters and the exposure time of the visual detector 17 to ensure image quality. The acquisition module synchronously acquires the component surface image of the vision detector 17, the running status data of the servo motor 4, and the position data of the adjustment mechanism of the sliding block 16; The acquisition and allocation module combines component inspection requirements with mechanical status data, calculates the adaptation height of the vision detector 17, generates an action command for the electric telescopic rod 10, and drives the fixed cylinder 14 to move the vision detector 17 along the guide rod 12 to the adaptation height. After the image processing module extracts multi-dimensional features from the image, it matches and identifies the type and location with the defect feature library. The final detection result is output through the touch screen 9, realizing flexible and accurate detection of surface defects of aluminum alloy components.

[0033] The embodiments of the surface defect detection device for aluminum alloy components based on visual inspection of the present invention can be implemented in whole or in part through software, hardware, firmware, or any combination thereof. The core functions of the defect detection system, such as environmental adaptation, visual detector height calculation, and defect identification, if implemented in software, can be embodied in the form of a computer program product. This product contains computer instructions, which, after being loaded into a computer or programmable device, can execute detection processes such as environmental parameter adjustment, adaptation height calculation, and multi-feature defect identification.

[0034] Computer instructions can be stored in computer-readable storage media (such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, optical disks, etc.) or transmitted between storage media via wired / wireless means (infrared, microwave, etc.). The instructions are used to execute device detection-related steps.

[0035] In this embodiment of the invention, the execution order of each process is based on functional logic and is not limited; the division of system units is only a logical functional division and can be adjusted as needed (such as integration or splitting); the coupling or communication connection between components can be achieved through interfaces, which can be electrical, mechanical or other adaptation methods.

[0036] All formulas used in this device employ dimensionless numerical calculations (specific dimensionless calculations can be achieved through standardization and other means, which will not be elaborated here). The formulas are obtained through large-scale data simulation software to closely approximate real-world testing scenarios. The preset parameters in the formulas are set by those skilled in the art based on actual testing needs.

[0037] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0038] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A visual inspection-based surface defect detection device for aluminum alloy components, comprising horizontally mounted guard plates (1), a guide plate (6) obliquely fixed to one end of the guard plates (1), a conveying mechanism installed between the guard plates (1) at the other end of the guide plate (6), a fixing mechanism vertically mounted in the middle of the conveying mechanism, and an adjusting mechanism installed inside the fixing mechanism, characterized in that, It also includes a defect detection system; the defect detection system includes an acquisition module, an image processing module, an acquisition and allocation module, and a result output module; The acquisition module is used to acquire image data of the surface of the aluminum alloy component, as well as the operating status data of the conveying mechanism and the position status data of the adjusting mechanism. The image processing module is used to process the image data through a preset image analysis algorithm, extract the surface features of the component, and identify the type and location of defects to generate defect identification results; The acquisition and adjustment module is used to calculate the adaptation height parameters of the visual detector (17) by adjusting the algorithm with preset parameters, combining the component detection requirements and the state data, and generating the action control command of the adjustment mechanism; the result output module is used to integrate the defect identification results, detection time, and mechanical operation parameters into a standardized detection report and output it.

2. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 1, characterized in that, Two support columns (2) are installed at equal intervals on the lower end of the guard plate (1) on the same side by bolts. One end of the support column (2) is fixed to the bottom plate frame (3). The bottom plate frame (3) is rectangular and has four casters installed on its lower end.

3. The surface defect detection device for aluminum alloy components based on vision inspection according to claim 1, characterized in that, The conveying mechanism includes rollers rotatably mounted between guard plates (1), a conveyor belt (5) is sleeved on the rollers, and a servo motor (4) is mounted on the guard plate (1) at one end of one of the rollers. The output end of the servo motor (4) passes through the guard plate (1) and is coaxially fixed to one end of the roller.

4. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 3, characterized in that, The fixing mechanism includes a detection frame (8) fixed to the upper end of the two side guard plates (1). The detection frame (8) has an adjustment cavity that extends vertically downward. A touch screen (9) is installed on one side of the detection frame (8) near the guide plate (6). A barrier plate (7) is installed on the guard plate (1) at one end of the guide plate (6) by bolts. A sealing plate (11) with a self-locking function is installed on one side of the detection frame (8) at the upper end of the guard plate (1).

5. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 4, characterized in that, The adjustment mechanism includes an electric telescopic rod (10) installed on the upper end of the test frame (8). The output end of the electric telescopic rod (10) passes through the test frame (8) and is connected to a U-shaped fixing frame (13) by a thread. A fixing cylinder (14) is installed between the fixing frames (13) by bolts. Sliding blocks (16) are symmetrically fixed on both sides of the fixing cylinder (14).

6. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 5, characterized in that, Both sliding blocks (16) are vertically connected with guide rods (12). The top of each guide rod (12) is fixed to the inner wall of the detection frame (8), and the bottom of the guide rod (12) is fixed to a limit block (19). A cross positioning frame (18) is horizontally installed inside the fixed cylinder (14). A detection hole is opened in the middle of the cross positioning frame (18), and a visual detector (17) is installed on the upper end of the cross positioning frame (18). A positioning plate (15) is installed on the fixed cylinder (14) above the visual detector (17) by bolts. A wire opening is opened on one side of the positioning plate (15).

7. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 1, characterized in that, The image processing module is used to process the image data using a preset image analysis algorithm, extract surface features of the component, and identify the type and location of defects. Specifically, it includes the following steps: Step 1: Adaptive image preprocessing. In response to the strong reflective properties of aluminum alloy components, firstly, illumination compensation is performed based on the adaptation relationship between the preset standard illumination intensity and the real-time illumination intensity of each pixel in the image. Then, noise in the image is removed through filtering. Step 2: Multi-dimensional feature extraction. Simultaneously extract three types of core features from the component surface to distinguish defective areas from normal surface textures. These are: gray-level gradient features obtained through gradient calculation, texture entropy features obtained through texture statistical analysis, and geometric moment features obtained through image pixel distribution calculation. Step 3: Defect classification and identification. Through the preset weighted feature decision logic, combined with the standard feature parameters of normal aluminum alloy surface, the extracted multi-dimensional features are comprehensively judged to determine whether the image pixel area is a defect area. If a region is identified as a defect, the extracted multi-dimensional features are then matched with a pre-defined multi-type defect feature library to determine the specific type and pixel-level location of the defect. Finally, a recognition result containing defect type and location boundary information is generated, and this recognition result is recorded as the defect recognition result.

8. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 6, characterized in that, The process by which the acquisition and allocation module calculates the adaptation height parameters and generates motion control commands through a preset parameter adjustment algorithm specifically includes the following steps: Step 1: Obtain the relevant parameters for component inspection requirements and the status data of the adjustment mechanism; among which, the relevant parameters for component inspection requirements include the characteristic parameters of the aluminum alloy component, the height correlation coefficient matching the preset inspection accuracy level, and the basic compensation height of the vision detector (17); the status data of the adjustment mechanism includes the actual height of the vision detector (17) corresponding to the current position of the adjustment mechanism; Step 2: Calculate the theoretical adaptation height. Combine the relevant parameters of the component inspection requirements and determine the theoretical adaptation height of the vision detector through parameter collaborative calculation. During the calculation process, the height correlation coefficient is dynamically adjusted according to the detection accuracy level. Step 3: Optimize and adjust the mechanical condition. Based on the current actual height of the adjustment mechanism, optimize and correct the theoretically suitable height. Step 4: Generate motion control commands. Based on the difference between the optimized adaptation height and the current height of the adjustment mechanism, and combined with the motion characteristics of the adjustment mechanism, determine the motion stroke of the adjustment mechanism, and determine the motion direction based on the relationship between the final adaptation height and the current height. Finally, generate motion control commands for the adjustment mechanism that include motion direction and motion stroke.

9. The visual inspection-based surface defect detection device for aluminum alloy components according to claim 6, characterized in that, The defect detection system includes an environment adaptation module, which collects environmental parameters of the detection area in real time. When the environmental parameters exceed the preset adaptation range, it automatically adjusts the preprocessing parameters of the image analysis algorithm and the working parameters of the visual detector (17). The environmental parameters include ambient light intensity and ambient temperature. The preprocessing parameters include image denoising parameters and contrast enhancement parameters.

Images