A computer mainboard visual inspection device

Abstract

The application is suitable for the technical field of electronic manufacturing quality detection, and provides a computer mainboard visual detection equipment, which comprises a detection box, a camera main body is arranged at the top center position of the detection box, a lighting mechanism is arranged around the camera main body, a clamping mechanism for fixing the computer mainboard is arranged at the bottom of the detection box, and a detection effect optimization system is further arranged, which comprises a shooting state regulation module: continuously monitoring a shooting effect index; when the shooting effect index is less than a preset critical threshold, the system analyzes and adjusts related influence parameters in a mainboard state evaluation module, a lighting condition evaluation module and a camera state evaluation module. The computer mainboard visual detection equipment can effectively solve the problem of lack of intelligent diagnosis and priority sorting in the adjusting mechanism of the prior art.

Description

Technical Field

[0001] This invention belongs to the field of electronic manufacturing quality inspection technology, and in particular relates to a computer motherboard visual inspection device. Background Technology

[0002] With the continuous development of the electronics and information industry, the manufacturing quality of computer motherboards, as core hardware components, directly affects the overall performance and operational stability of the system. At the final stage of the production line, visual inspection technology is widely used to identify quality problems such as motherboard solder joint defects, component placement deviations, and circuit anomalies.

[0003] Current mainstream inspection methods rely on industrial cameras to acquire images, followed by algorithmic defect analysis. However, significant technical bottlenecks exist in actual inspection processes, severely limiting the reliability and consistency of inspection results.

[0004] Computer motherboards are susceptible to external vibration interference during testing, causing dynamic blurring of images; the specular reflection characteristics of surface components can lead to local overexposure or dark areas, interfering with the identification of key features; and angular deviations between the motherboard and the camera's imaging plane during placement can cause image distortion and scaling issues. These inherent problems of the tested object and environmental factors are deeply intertwined, forming a complex coupling effect.

[0005] Under the existing technological framework, these multi-dimensional influencing factors are treated piecemeal, lacking a unified quantitative evaluation mechanism. When image quality deteriorates, traditional methods can only implement single parameter adjustments. This isolated control strategy is difficult to cope with scenarios where multiple factors deteriorate concurrently, often requiring manual intervention for troubleshooting, significantly reducing detection efficiency and increasing the risk of production interruption. Summary of the Invention

[0006] The purpose of this invention is to provide a computer motherboard visual inspection device, which aims to solve the problem that traditional methods can only implement single parameter adjustments, and this isolated control strategy is difficult to cope with scenarios where multiple factors deteriorate simultaneously.

[0007] This invention is implemented as follows: a computer motherboard visual inspection device includes an inspection box, a camera body disposed at the top center of the inspection box, an illumination mechanism disposed around the camera body, a clamping mechanism for fixing the computer motherboard disposed at the bottom of the inspection box, and an inspection effect optimization system, which includes:

[0008] Motherboard Status Assessment Module: Based on the vibration amplitude of the computer motherboard, the interference of specular reflection, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane, a motherboard status assessment model is constructed, and the motherboard status index is output.

[0009] Lighting condition assessment module: Constructs a lighting condition assessment model based on the lighting uniformity, irradiation stability, and illuminance of the lighting facility, and outputs a lighting health index;

[0010] Shooting Condition Assessment Module: The shooting condition assessment model is constructed based on the motherboard status index and light health index as the main conditions, and the ambient humidity and ambient temperature as influencing factors, and the shooting condition index is output.

[0011] Camera status assessment module: Based on the camera's own noise level, lens cleanliness, and effective focus sharpness, a camera status assessment model is constructed, and a camera status index is output.

[0012] Shooting effect evaluation module: Constructs a shooting effect evaluation model based on shooting condition index and camera status index, and outputs the shooting effect index;

[0013] Shooting status control module: continuously monitors the shooting effect index; when the shooting effect index is less than the preset critical threshold, the system analyzes and adjusts the relevant influencing parameters in the motherboard status evaluation module, the lighting condition evaluation module, and the camera status evaluation module.

[0014] A further technical solution is that the lighting mechanism includes a fixing ring, a connecting rod, a No. 1 motor, and a lamp;

[0015] The fixing ring is fixedly connected to the inner top surface of the testing box. Multiple connecting rods are fixedly connected to the fixing ring at equal intervals. One end of each connecting rod is rotatably connected to the output shaft of motor number one. Each motor number one is fixedly connected to a lamp, and all lamps are set at equal angles.

[0016] A further technical solution is that the clamping mechanism includes an electric push rod, a fixed frame, a second motor, and a clamping plate;

[0017] Two electric push rods are arranged opposite each other on the inner side wall of the testing box. The telescopic ends of the electric push rods are fixedly connected to a fixing frame. The fixing frame is fixedly connected to a second motor. The output shaft of the second motor is fixedly connected to a clamping plate. The clamping plate can clamp and fix the computer motherboard.

[0018] A further technical solution involves substituting the vibration amplitude of the computer motherboard, the specular reflection interference, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane into a maximum-minimum normalization formula for normalization, thereby generating vibration amplitude index, specular reflection interference index, and angular deviation index in sequence; in the motherboard condition assessment model:

[0019] The vibration amplitude index, specular reflection interference index, and angle deviation index are weighted and summed with preset vibration amplitude weighting coefficients, specular reflection interference weighting coefficients, and angle deviation weighting coefficients, respectively. The motherboard status index is then calculated by subtracting the weighted sum from 1.

[0020] A further technical solution involves substituting the illumination uniformity, illumination stability, and luminance of the illumination mechanism sequentially into a maximum-minimum normalization formula for normalization, thereby generating illumination uniformity index, illumination stability index, and luminance index in sequence; in the illumination condition evaluation model:

[0021] First, the illumination uniformity index, illumination stability index, and illumination brightness index are weighted and summed with preset illumination uniformity weighting coefficients, illumination stability weighting coefficients, and illumination brightness weighting coefficients respectively to obtain the basic illumination index. Then, based on the degree to which the product of the illumination uniformity index and the illumination stability index deviates from the ideal value of 1, the basic illumination index is reduced by multiplying it by a penalty term adjusted by a preset synergy coefficient, thereby calculating the illumination health index.

[0022] A further technical solution involves substituting the ambient humidity and temperature sequentially into a maximum-minimum normalization formula for normalization, and then generating humidity and temperature indices accordingly; in the shooting condition evaluation model:

[0023] First, the motherboard status index and the lighting health index are weighted and summed to obtain the basic condition index;

[0024] Then, an environmental factor is constructed using an exponential decay function. This environmental factor is a decay coefficient between 0 and 1. It is calculated by using the natural constant e as the base, and the sum of the negative humidity influence weighting coefficient multiplied by (the square of the difference between the humidity index and the optimal humidity threshold) and the negative temperature influence weighting coefficient multiplied by (the square of the difference between the temperature index and the optimal temperature threshold) as the exponent, and then performing a power operation.

[0025] The final shooting condition index is the product of the basic condition index and the environmental factor.

[0026] A further technical solution involves substituting the camera's inherent noise level, lens cleanliness, and effective focus sharpness sequentially into a maximum-minimum normalization formula for normalization, thereby generating noise intensity, lens cleanliness, and focus sharpness indices. In the camera state evaluation model:

[0027] First, the product of the lens cleanliness index and the focus sharpness index is used as a co-term, and its deviation from the ideal value of 1 is calculated.

[0028] Then, the noise intensity index, lens cleanliness index, and focus sharpness index are weighted and summed with preset noise impact weighting coefficients, lens impact weighting coefficients, and focus impact weighting coefficients respectively to obtain the basic camera index.

[0029] Finally, the deviation of the collaborative term is amplified by a preset collaborative influence factor, and the basic camera index is reduced to calculate the camera state index.

[0030] Further technical solutions, in the shooting effect evaluation model:

[0031] First, the product of the shooting condition index and the camera state index is used as a co-term, and its deviation from the ideal value of 1 is calculated.

[0032] Then, the shooting condition index and camera status index are weighted and summed with the preset shooting condition weight coefficient and camera status weight coefficient respectively to obtain the basic effect index.

[0033] Finally, the deviation of the collaborative item is amplified by a preset collaborative penalty factor, and the basic effect index is reduced to calculate the shooting effect index.

[0034] A further technical solution is provided, where the control logic in the shooting status control module is as follows:

[0035] S1: Analyze which of the three indices—motherboard status index, lighting health index, and camera status index—deviates the most from its own baseline value;

[0036] S2: Lock the module with the largest deviation as the primary adjustment target and call the corresponding automatic adjustment program;

[0037] S3: After adjustment, re-evaluate the shooting effect index. If it still does not meet the standard, adjust it in the following order: motherboard status evaluation module, lighting condition evaluation module, and camera status evaluation module.

[0038] S4: If automatic adjustment is ineffective, the shooting status control module will trigger the alarm and request manual intervention.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] The computer motherboard visual inspection device of this application effectively solves the problem of the lack of intelligent diagnosis and priority ranking in the adjustment mechanism of the prior art. When the shooting effect index drops, the system no longer performs blind or isolated adjustments, but first intelligently identifies the main cause of the decline in shooting effect by quantitatively analyzing the deviation of the motherboard status index, lighting health index, and camera status index. This intelligent diagnostic capability enables the system to concentrate adjustment resources on the most critical problem points, significantly improving the efficiency and accuracy of troubleshooting.

[0041] This application transforms key factors affecting visual inspection results—such as the vibration amplitude of the computer motherboard, specular reflection interference, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane—from raw data of different dimensions into a unified, quantifiable, standardized index. This standardization process eliminates the differences between different physical dimensions, allowing each factor to participate fairly in the comprehensive evaluation. Based on this, by constructing a weighted motherboard condition assessment model, the overall inspection suitability of the motherboard can be accurately quantified, outputting a motherboard condition index. This provides a systematic, objective, and adjustable evaluation mechanism for the motherboard condition assessment module, solving the problem of traditional methods' difficulty in comprehensively quantifying and evaluating multi-dimensional and highly coupled factors. As a crucial input to the imaging condition assessment module, this motherboard condition index significantly improves the robustness and accuracy of the entire inspection effect optimization system, ensuring that the computer motherboard visual inspection equipment can still obtain high-quality images in complex and changing environments, thereby improving the reliability of defect detection.

[0042] This application enables a more accurate evaluation of the overall imaging performance of computer motherboard vision inspection equipment, resolving the inaccurate evaluation problem in existing technologies caused by neglecting the synergistic influence of shooting condition indices and camera state indices. This model effectively captures the interdependence between the shooting environment and the equipment's own state, avoiding misjudgments when a single factor performs well but the overall effect is poor, thus providing a more precise basis for subsequent intelligent diagnosis and control. This significantly improves the robustness and automation level of the vision inspection system, reduces the need for manual intervention, and ensures the stability and reliability of the inspection results. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the structure of the present invention;

[0044] Figure 2 This is a schematic diagram of the illumination mechanism in this invention;

[0045] Figure 3 This is a schematic diagram of the clamping mechanism in this invention;

[0046] Figure 4 This is a schematic diagram of the detection effect optimization system in this invention.

[0047] In the attached diagram: 1. Detection box; 2. Camera body; 3. Illumination mechanism; 31. Fixing ring; 32. Connecting rod; 33. Motor No. 1; 34. Lamp; 4. Clamping mechanism; 41. Electric push rod; 42. Fixing frame; 43. Motor No. 2; 44. Clamping plate. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0049] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0050] like Figure 1 , Figure 4 As shown, a computer motherboard visual inspection device according to an embodiment of the present invention includes an inspection box 1, a camera body 2 disposed at the top center of the inspection box 1, an illumination mechanism 3 disposed around the camera body 2, a clamping mechanism 4 for fixing the computer motherboard disposed at the bottom of the inspection box 1, and an inspection effect optimization system, which includes:

[0051] Motherboard Status Assessment Module: Based on the vibration amplitude of the computer motherboard, the interference of specular reflection, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane, a motherboard status assessment model is constructed, and the motherboard status index is output.

[0052] Lighting condition assessment module: Based on the light uniformity, irradiation stability and light intensity of the lighting mechanism 3, a lighting condition assessment model is constructed, and the light health index is output;

[0053] Shooting Condition Assessment Module: The shooting condition assessment model is constructed based on the motherboard status index and light health index as the main conditions, and the ambient humidity and ambient temperature as influencing factors, and the shooting condition index is output.

[0054] Camera status assessment module: Based on the noise level of the camera body 2, the cleanliness of the lens, and the effective focus sharpness, a camera status assessment model is constructed, and the camera status index is output.

[0055] Shooting effect evaluation module: Constructs a shooting effect evaluation model based on shooting condition index and camera status index, and outputs the shooting effect index;

[0056] Shooting status control module: continuously monitors the shooting effect index; when the shooting effect index is less than the preset critical threshold, the system analyzes and adjusts the relevant influencing parameters in the motherboard status evaluation module, the lighting condition evaluation module, and the camera status evaluation module.

[0057] This embodiment proposes a computer motherboard visual inspection device, which achieves comprehensive optimization of the inspection process by integrating multiple evaluation modules and a control module.

[0058] The camera body 2 can consist of a fixedly mounted industrial camera with its lens facing the bottom of the inspection box 1. An illumination mechanism 3 is fixed around the camera body 2 to provide constant illumination. A clamping mechanism 4 for securing the computer motherboard is provided at the bottom of the inspection box 1.

[0059] This embodiment also includes a detection effect optimization system, which can be implemented as an independent control unit, such as an embedded system or an industrial computer, running specialized software programs.

[0060] The motherboard status assessment module in the detection effect optimization system is responsible for evaluating the physical condition of the computer motherboard. This module can receive data from external sensors, process the raw data, and output a motherboard status index.

[0061] The lighting condition assessment module is used to evaluate the lighting quality provided by the lighting mechanism 3. This module can receive data from the lighting sensor, and after these measurement data are processed, a lighting health index is output.

[0062] The shooting condition assessment module comprehensively evaluates the motherboard status and lighting conditions, taking environmental factors into account. This module receives motherboard status index and lighting health index as primary inputs, while also receiving data from environmental sensors. After processing this input data, it outputs a shooting condition index.

[0063] The camera status assessment module is used to assess the operational status of the camera body 2 itself. This module can receive data from the camera's internal diagnostic system, process these assessment results, and output a camera status index.

[0064] The shooting effect evaluation module quantifies the overall shooting effect. This module receives a shooting condition index and a camera state index as input, and performs a comprehensive calculation based on preset logic or algorithms to output a shooting effect index. This index directly reflects the quality level of the image acquired by the current visual inspection.

[0065] The shooting status control module is responsible for intelligent adjustments based on the shooting effect index. This module continuously monitors the shooting effect index, and when the index falls below a preset critical threshold, the control module analyzes relevant influencing parameters from the motherboard status assessment module, lighting condition assessment module, and camera status assessment module. For example, if the motherboard status index is too low, the control module can issue a command to instruct the operator to check the fixation of the clamping mechanism 4; if the lighting health index is too low, the control module can prompt the operator to check the status of the lighting mechanism 3; if the camera status index is too low, the control module can suggest cleaning the lens or refocusing. These analyses and adjustments aim to restore the shooting effect index to an acceptable level.

[0066] like Figure 2 As shown, in a preferred embodiment of the present invention, the lighting mechanism 3 includes a fixing ring 31, a connecting rod 32, a first motor 33, and a lamp 34;

[0067] The fixing ring 31 is fixedly connected to the inner top surface of the detection box 1. The fixing ring 31 is fixedly connected to multiple connecting rods 32 at equal intervals. One end of each connecting rod 32 is rotatably connected to the output shaft of the first motor 33. Each first motor 33 is fixedly connected to a lamp 34. All lamps 34 are set at equal angles.

[0068] In this embodiment, the illumination mechanism 3 solves the problems of illumination uniformity and angular flexibility, thereby improving the accuracy of the illumination health index. Specifically, the fixing ring 31 is fixedly connected to the top surface of the detection box 1, providing a stable installation base and ensuring that the entire illumination mechanism 3 does not shift during the detection process, avoiding illumination fluctuations caused by vibration. The fixing ring 31 is fixedly connected to multiple connecting rods 32 at equal intervals, achieving uniform distribution of the lamps 34, making the light coverage more balanced and reducing local shadows or overexposure. One end of the connecting rod 32 is rotatably connected to the output shaft of the first motor 33, allowing the angle of the lamps 34 to be dynamically adjusted by the motor drive, adapting to different motherboard positions and reflection characteristics. The first motor 33 is fixedly connected to the lamps 34, achieving precise angle control of the lamps 34 and ensuring that the illumination direction can be optimized in real time. All lamps 34 are set at equal angles, ensuring the symmetry and consistency of light projection, thereby improving the overall illumination uniformity and stability, and providing more reliable data input for the illumination condition assessment module. This adjustable illumination mechanism 3 works in conjunction with the illumination condition evaluation module in the detection effect optimization system. When the illumination condition evaluation module detects that the illumination uniformity, illumination stability, or illumination intensity does not meet the requirements, the shooting state control module can instruct the first motor 33 to adjust the angle of the lamp 34 to optimize the illumination conditions, thereby improving the illumination health index and thus enhancing the shooting effect index.

[0069] like Figure 3 As shown, in a preferred embodiment of the present invention, the clamping mechanism 4 includes an electric push rod 41, a fixing frame 42, a second motor 43, and a clamping plate 44.

[0070] Two electric push rods 41 are arranged opposite each other on the inner side wall of the testing box 1. The telescopic ends of the electric push rods 41 are fixedly connected to the fixing frame 42. The fixing frame 42 is fixedly connected to the second motor 43. The output shaft of the second motor 43 is fixedly connected to the clamping plate 44. The clamping plate 44 can clamp and fix the computer motherboard.

[0071] In this embodiment, two electric push rods 41 are arranged at opposite positions on the inner sidewall of the testing box 1. When it is necessary to adjust the horizontal position of the computer motherboard, the control system drives the two electric push rods 41 to extend and retract synchronously or independently. The extension and retraction ends of the electric push rods 41 are firmly connected to the fixing frame 42, so the linear movement of the electric push rods 41 can accurately drive the fixing frame 42 to move horizontally. A second motor 43 is fixedly connected to each fixing frame 42, and the output shaft of the second motor 43 is directly fixedly connected to the clamping plate 44. When the computer motherboard is placed between the clamping plates 44, the second motor 43 can be driven to rotate its output shaft, thereby driving the clamping plate 44 to adjust its angle. In this way, the clamping plate 44 can clamp and fix the computer motherboard, and the overall horizontal position of the motherboard can be adjusted by the electric push rods 41 according to the testing requirements, while the tilt angle of the motherboard can be adjusted by the second motor 43. This integrated clamping mechanism 4 design makes the computer motherboard firmly fixed during the testing process, effectively suppressing the motherboard vibration caused by external vibration or the movement of the equipment itself. Meanwhile, through the coordinated action of the electric push rod 41 and the second motor 43, the position and angle of the main board can be precisely and dynamically adjusted, ensuring that the measured surface of the main board always maintains the best relative angle with the imaging plane of the camera body 2, thereby minimizing the impact of specular reflection interference and angle deviation on image quality.

[0072] In a preferred embodiment of the present invention, the vibration amplitude of the computer motherboard, the specular reflection interference degree, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane are sequentially substituted into the maximum-minimum normalization formula for normalization processing, and the vibration amplitude index, specular reflection interference degree index, and angular deviation index are generated sequentially; in the motherboard condition assessment model:

[0073] ;

[0074] in This is the vibration amplitude weighting coefficient. This is the weighting coefficient for specular reflection interference. This is the angle deviation weighting coefficient. , , as well as All are greater than 0. The amplitude index is the vibration index. The specular reflection interference index. The angle deviation index, This is the motherboard status index.

[0075] In this embodiment, the vibration amplitude of the computer motherboard refers to the degree of minute mechanical vibration that occurs during the testing process. The vibration amplitude can be measured in various ways. For example, a high-precision accelerometer can be directly mounted on the clamping mechanism 4 or the edge of the motherboard for real-time monitoring, or a laser vibrometer can be used to non-contactly measure the vibration displacement of the motherboard surface. Specular reflection interference refers to the intensity and range of specular reflection from the computer motherboard surface (such as solder joints, chip packages, metal connectors, etc.) to a light source. This interference can be evaluated by analyzing the pixel saturation and uniformity of brightness distribution in high-brightness areas of the image. The relative angular deviation between the motherboard surface under test and the camera imaging plane refers to the non-perpendicular or non-parallel relationship between the motherboard surface under test and the imaging plane of the camera body 2. This deviation can be calculated by acquiring the three-dimensional point cloud data of the motherboard using a multi-view stereo vision system, and then calculating its orientation relative to the camera imaging plane.

[0076] The max-min normalization formula is a commonly used data preprocessing method that linearly transforms the original data to a preset range, such as [0,1]. This formula is typically expressed as X... normalized = (X - X) min ) / (X max -X min ), where X is the original data, X min and X max These represent the minimum and maximum values ​​of the original data, respectively. This normalization process eliminates differences between different physical dimensions, allowing the vibration amplitude, specular reflection interference, and angular deviation parameters to be compared and weighted on a unified scale. This prevents any single parameter from dominating subsequent calculations due to its large numerical range, ensuring the fairness and accuracy of the assessment. The vibration amplitude index, specular reflection interference index, and angular deviation index are dimensionless values ​​obtained after processing with the maximum-minimum normalization formula. They represent the standardized state of the computer motherboard in terms of vibration, specular reflection, and angular deviation, respectively. These indices typically range from 0 to 1, where 0 may represent the ideal state (e.g., no vibration, no reflection) and 1 represents the worst state (e.g., maximum permissible vibration, strongest reflection). These indices serve as input variables in the motherboard condition assessment model, providing a unified and quantifiable data foundation for subsequent comprehensive evaluation.

[0077] The motherboard condition assessment model is a mathematical expression used to comprehensively quantify the overall condition of a computer motherboard. This model employs a weighted summation method, linearly combining the vibration amplitude index, specular reflection interference index, and angular deviation index, and calculates the motherboard condition index by subtracting the weighted sum from one. This design makes A higher value indicates a better motherboard condition, which is more beneficial for visual inspection. (Weight coefficients in the model) , as well as These correspond to the influence of vibration amplitude, specular reflection interference, and angular deviation on the detection effect, respectively. The sum of these weighting coefficients is 1, and each coefficient is greater than 0, ensuring that all factors are considered. These coefficients can be adjusted according to actual detection needs and experience to reflect the relative importance of different factors in specific detection scenarios. For example, in detection tasks sensitive to minute defects, a higher weight may be assigned to vibration amplitude. Motherboard Status Index This is the final output of the model, a comprehensive index between 0 and 1 that intuitively reflects the overall suitability of the current computer motherboard for testing.

[0078] In this model, the pre-set weighting coefficients , as well as (The sum of these factors is 1 and all are greater than 0) reflects the relative importance of each factor to the final image quality. For example, if experience shows that vibration has the greatest impact on image sharpness, then... It can be set relatively large. The model comprehensively considers the cumulative effect of these three factors through a weighted summation, and uses a "1 minus" approach to make the final output motherboard status index... It can intuitively represent the quality of the motherboard's condition, that is... The higher the value, the more ideal the motherboard's detection status. The value is then output by the motherboard status assessment module, serving as one of the key input conditions for the imaging condition assessment module. In this way, this solution transforms the originally complex and interconnected physical phenomena into quantifiable digital indicators, providing the motherboard status assessment module with a precise and unified evaluation basis, thus solving the problem of parameter standardization and comprehensive evaluation in traditional methods. This systematic evaluation method enables computer motherboard visual inspection equipment to more accurately grasp the real-time status of the motherboard under test, providing a solid data foundation for the subsequent optimization and control of the motherboard status assessment module.

[0079] In a preferred embodiment of the present invention, the illumination uniformity, illumination stability, and illumination intensity of the illumination mechanism 3 are sequentially substituted into the maximum-minimum normalization formula for normalization processing, and the illumination uniformity index, illumination stability index, and illumination intensity index are generated sequentially; in the illumination condition evaluation model:

[0080] ;

[0081] in This is the weighting coefficient for illumination uniformity. This is the irradiation stability weighting coefficient. This is the luminance weighting coefficient. ,and , as well as All are greater than 0. For the synergy coefficient, ,, The light uniformity index. The irradiation stability index. The luminance index is the light intensity index. This refers to the light health index.

[0082] In this embodiment, the illumination condition assessment module can receive data from illumination sensors, such as measuring illumination uniformity through multiple photodiode arrays, measuring illumination stability through continuous sampling, and measuring illumination intensity through a luminance meter. These measurement data are then processed to output a illumination health index.

[0083] The illumination uniformity index, illumination stability index, and illumination intensity index are normalized quantified values, representing the performance of illumination mechanism 3 in terms of illumination uniformity, illumination time stability, and illumination intensity, respectively. The illumination uniformity index reflects the uniformity of illumination distribution within the measured area; a high index indicates a more consistent illumination distribution. The illumination stability index reflects the degree of fluctuation in illumination intensity over time; a high index indicates more stable illumination. The illumination intensity index reflects the overall intensity level of illumination; a high index indicates more sufficient illumination. These indices serve as the basic input for subsequent illumination condition evaluation models, ensuring the standardization and comparability of the evaluation.

[0084] The light condition assessment model is used to comprehensively evaluate the health level of light conditions and output the Light Health Index (LHI). The core of the model lies in combining linear weighting and synergistic effects. Specifically, The light uniformity index. The irradiation stability index. These are the luminance indexes, which are the input parameters after normalization. , as well as These are the weighting coefficients for illumination uniformity, illumination stability, and illumination intensity, used to characterize the relative importance of these three factors in overall illumination health, and their sum is 1, ensuring a reasonable allocation of weights. The synergy coefficients, ranging from (0,1), are used to quantify the degree of synergy or mutual influence between illumination uniformity and illumination stability. These weighting coefficients and synergy coefficients can be trained and optimized based on practical application scenarios, expert experience, or through machine learning methods (such as regression analysis). When either illumination uniformity or illumination stability performs poorly, the synergy term... Will have an impact on the whole This produces a punitive effect, thus more accurately reflecting the true health status of lighting conditions.

[0085] The illumination condition assessment module of this application uses a systematic approach to comprehensively and quantitatively assess illumination factors affecting the imaging quality of visual inspection. First, the system collects raw data such as illumination uniformity, illumination stability, and illuminance of the illumination mechanism 3 in real time or periodically. These raw data may have different dimensions and numerical ranges. To ensure the fairness and accuracy of the assessment, these data are sequentially substituted into a maximum-minimum normalization formula to generate dimensionless illumination uniformity, illumination stability, and illuminance indices, unifying all illumination-related parameters within the range of [0,1], laying the foundation for subsequent comprehensive assessment. Subsequently, these normalized indices are input into the illumination condition assessment model to calculate the illumination health index. The model first uses a weighted sum to comprehensively consider the independent contributions of illumination uniformity, illumination stability, and illumination intensity to illumination health. The weighting coefficients are... , as well as The relative importance of each factor can be adjusted based on actual needs or experience. Furthermore, the model introduces a synergistic term. This synergy term is expressed through the synergy coefficient. Captured illumination uniformity and irradiation stability The interaction between them. When either the uniformity of illumination or the stability of illumination performs poorly (i.e., or (When the value is low), the value of the synergistic term will decrease, thus affecting the overall light health index. This creates a penalty effect. This design avoids the bias that may result from isolating and evaluating a single factor in traditional methods, and can more realistically reflect the overall health of lighting conditions. In particular, when a key lighting parameter has a deficiency, it can promptly reflect its negative impact on the overall lighting quality.

[0086] In a preferred embodiment of the present invention, the ambient humidity and ambient temperature are sequentially substituted into the maximum-minimum normalization formula for normalization, and the humidity index and temperature index are generated sequentially; in the shooting condition evaluation model:

[0087] ;

[0088] in This is the motherboard status weighting coefficient. The weighting factor for the influence of illumination. As the weighting factor for the effect of humidity, The weighting factor for the effect of temperature. , ,and , , as well as All are greater than 0. This is the motherboard status index. For the light health index, Humidity index The optimal humidity threshold is set. , Temperature index To achieve the optimal temperature threshold, , This refers to the shooting conditions index.

[0089] In this embodiment, ambient humidity and ambient temperature are important environmental parameters affecting the imaging quality of visual inspection. Ambient humidity refers to the water vapor content in the air inside or around the inspection chamber 1, usually expressed as a relative humidity percentage. Ambient temperature refers to the air temperature inside or around the inspection chamber 1. These parameters can be measured by deploying corresponding sensors inside or near the inspection equipment. For example, capacitive or resistive humidity sensors can be used to monitor ambient humidity in real time, while thermistors or thermocouples can be used to monitor ambient temperature in real time. These sensors can be integrated into the control system of the inspection chamber 1 to periodically collect data. The maximum-minimum normalization formula is a commonly used data preprocessing method. The purpose of using this formula is to eliminate the dimensional differences between different physical quantities, making all input parameters comparable in the model, avoiding the dominance of certain parameters with large numerical ranges in the calculation, thereby improving the stability and accuracy of the model. The humidity index and temperature index are quantitative representations of ambient humidity and ambient temperature after maximum-minimum normalization. These indices typically take values ​​between 0 and 1, providing standardized input for the shooting condition evaluation model. The shooting condition assessment model is a comprehensive mathematical model used to quantitatively evaluate the combined impact of the current environment and the state of the object under test on the shooting effect of visual inspection. This model outputs a single shooting condition index by combining the motherboard condition index, lighting health index, ambient humidity index, and ambient temperature index.

[0090] Motherboard status weighting coefficient Weighting coefficients for the influence of light Used to balance motherboard status index and light health index The relative importance of these weighting coefficients in the shooting condition assessment model is typically derived from experience, expert knowledge, or by training and optimizing historical testing data using machine learning methods. Humidity influence weighting coefficients. And the weighting coefficient of temperature effect The factors used to adjust for the attenuation of the Shooting Condition Index (SCI) when ambient humidity and temperature deviate from their optimal values ​​can also be determined through experimental data analysis, expert experience, or optimization algorithms. Optimal humidity threshold. and optimal temperature threshold These represent the ideal ambient humidity and temperature conditions for achieving optimal image quality during visual inspection. These optimal thresholds are typically determined through experimental testing, industry standards, or recommendations from equipment manufacturers. (Exponential function) It is a key component of the shooting condition assessment model, used to simulate the nonlinear decay effect of environmental humidity and temperature deviating from their optimal thresholds on the shooting condition index. When the humidity index or temperature index deviates from its corresponding optimal threshold... or When the environmental conditions deteriorate, the value of the function will decrease, thus accurately reflecting the negative impact of the deterioration of environmental conditions on the shooting quality.

[0091] The imaging condition evaluation module in this application aims to comprehensively quantify the impact of environmental factors on the imaging quality of computer motherboard visual inspection. Among these factors, the motherboard state weighting coefficient... Weighting coefficients for the influence of light This is used to adjust the relative importance of motherboard status and lighting conditions within the overall shooting conditions, ensuring a reasonable consideration of key influencing factors. More importantly, the model introduces an exponential decay term. The attenuation term affects the weighting coefficient through humidity. Temperature influence weighting coefficient and the preset optimal humidity threshold and optimal temperature threshold This precisely captures the non-linear negative impact on image quality when ambient humidity and temperature deviate from ideal conditions. When ambient humidity or temperature deviates from its optimal threshold, the negative value of the exponential term increases, causing the overall value of the exponential function to decrease, thus affecting the final image quality index. The shooting condition index decreases. This design allows the model to sensitively reflect the potential impact of deteriorating environmental conditions on shooting results and quantify it in the shooting condition index. Through the above mechanism, this shooting condition evaluation module not only comprehensively considers the motherboard's own condition and lighting conditions, but also creatively incorporates the nonlinear effects of environmental humidity and temperature into the evaluation system, making the shooting condition index more effective. It can more comprehensively and accurately reflect the overall conditions of current visual inspection. Compared with solutions that rely solely on motherboard status and lighting conditions for evaluation, it can identify potential risks of image quality degradation caused by environmental factors earlier and more accurately, providing a more reliable basis for subsequent intelligent adjustment.

[0092] In a preferred embodiment of the present invention, the noise level of the camera body 2, the lens cleanliness, and the effective focus sharpness are sequentially substituted into the maximum-minimum normalization formula for normalization processing, and noise intensity index, lens cleanliness index, and focus sharpness index are generated sequentially; in the camera state evaluation model:

[0093] ;

[0094] in The weighting coefficients are for the impact of noise. The lens influence weighting coefficient, The focusing effect weighting coefficient, ,and , as well as All are greater than 0. As a synergistic influencing factor, , Noise intensity index, Lens cleanliness index This refers to the focus sharpness index. This is the camera status index.

[0095] In this embodiment, the indices are noise intensity index, lens cleanliness index, and focus sharpness index. These indices are normalized values ​​representing the state of the camera body 2 in specific aspects, typically ranging from 0 to 1. Their purpose is to transform raw, potentially complex physical measurements into unified, intuitive quantitative indicators, facilitating input and calculation for subsequent evaluation models and enabling unified processing and comparison of camera state information across different dimensions. These indices can be directly derived from normalized values, or further mapped or transformed based on normalization to better reflect actual impacts.

[0096] The camera condition assessment model combines noise intensity index, lens cleanliness index, and focus sharpness index, and uses a weighting system... , as well as and synergistic influencing factors This model quantifies the impact of these factors on the camera's condition. It provides a quantitative and comprehensive index that accurately reflects the health status of the camera subject 2 in real time, overcoming the limitations of traditional methods that isolate and evaluate parameters, and especially considering the synergistic effects between key parameters. The model can be implemented using software algorithms, for example, in the control system of a detection device, by using a programming language to write functions or modules that receive normalized indices as input and calculate the results according to the above formula. value.

[0097] in, The weighting coefficients are for the impact of noise. The lens influence weighting coefficient, These are the focusing influence weighting coefficients. These weighting coefficients represent the relative importance of different factors in camera state evaluation. They are values ​​between 0 and 1, and their sum is 1, ensuring a reasonable distribution of the contributions of each factor. This allows the system to flexibly adjust the influence of different camera state parameters on the final evaluation result based on actual application scenarios and experience. These weighting coefficients can be set via a configuration file during system initialization, allowing users or maintenance personnel to adjust them according to actual detection needs and experience; they can also be optimized through machine learning methods, for example, by collecting a large amount of historical detection data and corresponding image quality evaluation results to train a model to automatically learn and determine the optimal weighting coefficients.

[0098] also, Synergistic Influence Factor It is a parameter between 0 and 1, used to quantify the strength of the synergistic effect between lens cleanliness and focus sharpness. When A larger value indicates a more significant synergistic negative impact between the two parameters. This factor captures and quantifies the nonlinear interaction between camera state parameters, particularly when multiple key parameters are simultaneously in poor condition. Their negative impact on the overall camera state may not be a simple additive effect, but rather exhibit an amplified one. This makes the evaluation model more realistic and improves the accuracy of the evaluation. This factor can be set based on empirical values, for example, by using expert knowledge or experimental data analysis to determine the comprehensive impact on image quality when the lens is dirty and the focus is inaccurate; it can also be optimized using data-driven methods, such as through regression analysis or optimization algorithms, learning from actual test data to determine the factor that best reflects the synergistic effect. value.

[0099] final, As the final output, these indices provide crucial information for subsequent shooting effect evaluation and state adjustment. These indices are stored and transmitted in the system as digital variables, such as floating-point variables in memory or specific fields in data structures, for different modules to call and process.

[0100] The camera status assessment module of this application aims to provide a comprehensive and accurate assessment of the health status of the camera body 2. This module first acquires raw data related to the performance of the camera body 2, including its intrinsic noise level, lens cleanliness, and effective focus sharpness. Subsequently, it converts this data into a standardized dimensionless exponent using a maximum-minimum normalization formula. This model comprehensively considers various aspects of camera health through a weighted summation, where… Used to reflect the negative impact of noise levels on camera performance, when the lens is unclean ( Low value) and inaccurate focus ( When the value is low, this synergistic term significantly reduces the overall CSI value, thus more accurately reflecting the combined negative impact of multiple camera performance degradations on image quality. In this way, the camera state assessment module can output a real-time, quantified camera state index. This index is then used by the image capture performance evaluation module to comprehensively assess the overall quality of the captured images. This systematic evaluation method ensures that the condition of the camera body 2 is fully considered and its impact on the quality of the final inspected image is accurately quantified, thereby significantly enhancing the robustness and reliability of the entire computer motherboard vision inspection device.

[0101] In a preferred embodiment of the present invention, the shooting effect evaluation model includes:

[0102] ;

[0103] in For shooting conditions, weighting coefficients. These are the camera state weighting coefficients. ,and , All are greater than 0. For shooting conditions index, For camera status index, As a collaborative punishment factor, , This refers to the shooting effect index.

[0104] In this embodiment, the above-mentioned shooting effect evaluation model aims to comprehensively quantify the imaging quality of the computer motherboard visual inspection device under specific shooting conditions. The shooting condition weighting coefficient is used in this model. and camera state weight coefficient Used to allocate shooting condition index and camera status index The relative importance in the overall assessment. The sum of these two weighting coefficients is 1, and both are greater than 0, ensuring the balance and effectiveness of the assessment and avoiding the excessive dominance of a single factor. Shooting Conditions Index It is a comprehensive quantification of external environmental factors affecting image quality (such as motherboard status, lighting conditions, ambient temperature and humidity); a higher value indicates a more ideal external shooting environment. Camera Status Index This is a comprehensive quantification of the performance status of camera body 2 itself; a higher value indicates a better health condition for camera body 2. Cooperative penalty factor. It is a parameter between 0 and 1, used to measure the strength of the interaction between shooting conditions and camera status. When A larger value indicates a stronger penalty from the system for insufficient coordination between the two elements. (Shooting Effect Index) This is the final comprehensive evaluation result output by the model; the higher the value, the better the overall shooting effect.

[0105] The scheme in this application introduces a collaborative penalty factor. A system was constructed that fully considers shooting conditions. and camera status index A shooting effect evaluation model based on the mutual influence between elements. This model first uses weighting coefficients... and right and Weighted summation reflects the independent contributions of each factor to the shooting effect. Based on this, the model introduces a multiplication term. The core of this multiplication term lies in ,when and When both are relatively high, their product is close to 1. When the value is close to 0, the impact of the penalty is relatively small, resulting in a low shooting effect index. It is mainly determined by weighted sums. However, when or When any of the exponents is low, their product will decrease significantly, leading to Increase, thereby increasing the collaborative penalty factor Overall shooting effect index This design mechanism means that even if one single factor performs well, if another key factor performs poorly, the overall shooting effect index will be negatively impacted. It will also be significantly lowered, thus accurately reflecting the true situation of the interdependence and mutual constraints between shooting conditions and camera status.

[0106] For example, even if the lighting conditions and motherboard status are ideal (high... However, if the camera lens is dirty or the focus is inaccurate (low sensitivity), Even with these improvements, the final image quality will still be poor. This model can accurately capture this lack of coordination, avoiding potential misjudgments that may occur with traditional models. In this way, the model can more accurately evaluate the imaging performance of computer motherboard vision inspection equipment in actual operation, providing a more reliable basis for subsequent intelligent control.

[0107] In a preferred embodiment of the present invention, the control logic in the shooting state control module is as follows:

[0108] S1: Analyze which of the three indices—motherboard status index, lighting health index, and camera status index—deviates the most from its own baseline value;

[0109] S2: Lock the module with the largest deviation as the primary adjustment target and call the corresponding automatic adjustment program;

[0110] S3: After adjustment, re-evaluate the shooting effect index. If it still does not meet the standard, adjust it in the following order: motherboard status evaluation module, lighting condition evaluation module, and camera status evaluation module.

[0111] S4: If automatic adjustment is ineffective, the shooting status control module will trigger the alarm and request manual intervention.

[0112] In this embodiment, the control logic addresses the lack of intelligent diagnosis and priority ranking in existing control mechanisms by defining a series of steps, thereby improving control efficiency and system reliability. Step S1 aims to quantitatively compare three core dimensions affecting shooting performance to identify the most significant performance bottleneck. The judgment of "maximum deviation from its own baseline value" can be achieved in various ways. For example, the absolute difference between each index and a preset ideal baseline value can be calculated, and the module corresponding to the index with the largest difference can be selected; alternatively, the percentage deviation of each index from its historical average or the upper / lower limit of the normal operating range can be calculated, and the module with the largest percentage deviation can be selected. The baseline value can be the historical average value of the system under ideal conditions, the optimal value determined through experiments, or a target threshold set according to actual application requirements. In this way, the system can intelligently diagnose the most pressing problem source from a data perspective. Step S2, based on the diagnostic results of S1, executes the highest-priority automated intervention. Locking down the module with the largest deviation means concentrating limited control resources on solving the most prominent problem. The corresponding automated control program can vary depending on the module type. For example, if the motherboard status assessment module deviates to the maximum, the automated adjustment program can control the clamping mechanism 4 to adjust the clamping force or position of the computer motherboard to reduce vibration or angular deviation; if the lighting condition assessment module deviates to the maximum, the automated adjustment program can adjust the brightness, uniformity, or angle of the lighting mechanism 3; if the camera status assessment module deviates to the maximum, the automated adjustment program can trigger the autofocus function of the camera body 2 (implemented by the camera's built-in function or external motor adjustment) or activate the lens cleaning mechanism. These programs are designed to perform preliminary optimization of problematic modules in a preset, non-manual manner. Step S3 reflects the iterative and comprehensive nature of the system adjustment. After the primary adjustment object is automatically adjusted, the system immediately recalculates the shooting effect index to verify the effectiveness of the adjustment. If the shooting effect index is still lower than the preset critical threshold (i.e., "not up to standard"), it indicates that a single adjustment is insufficient to solve the problem, or that there are multiple sources of problems. At this time, the system will automatically adjust other modules in sequence according to the preset priority order (e.g., motherboard status assessment module, lighting condition assessment module, camera status assessment module). This sequential adjustment can be determined based on factors such as experience, the coupling relationship between modules, or adjustment costs, ensuring that the system can systematically identify and resolve potential secondary problems until the shooting effect meets the standard or all automated adjustment methods have been tried. Step S4 is the system's final guarantee mechanism when automated adjustment fails to solve the problem. When all automated adjustment procedures have been tried, but the shooting effect index still fails to reach the preset standard, the system will determine that "automatic adjustment is ineffective." At this time, the shooting status control module will trigger an alarm, such as issuing an audible and visual alarm or displaying a warning message on the operation interface, and send a command to the operator requesting manual intervention.This ensures that in complex or abnormal situations, the system can promptly notify human operators to intervene and conduct deeper fault diagnosis and handling, avoiding interruptions or quality degradation of the detection task due to the limitations of automated adjustment.

[0113] The computer motherboard visual inspection device of this application, when continuously monitoring the shooting effect index, will activate the aforementioned intelligent control logic once it detects that the index is below a preset critical threshold, in order to systematically optimize shooting conditions. This control logic first analyzes three key indicators—motherboard status index, lighting health index, and camera status index—to accurately identify the source of the problem that has the greatest impact on shooting effect. Specifically, the system compares the degree to which each of these three indices deviates from its ideal benchmark value, thereby quantitatively determining which module's performance has declined most significantly. This diagnostic method based on quantitative data avoids the inefficiency of traditional blind or experience-based adjustments and can directly target the core of the problem. Once the module with the largest deviation is identified, the system will lock it as the primary adjustment target and immediately invoke the preset automatic adjustment program. This strategy of prioritizing the main problem can quickly improve the most critical performance bottleneck, thereby improving shooting effect in the shortest possible time. After the primary adjustment is completed, the system will immediately re-evaluate the shooting effect index. If the shooting effect index has recovered to the target level at this time, the control process ends, and the system resumes normal monitoring. However, if the shooting effect index still does not meet the target, the system will not stop but will automatically adjust other modules in a preset order. This iterative and sequential adjustment mechanism ensures that even with multiple interrelated or minor issues, the system can systematically identify and resolve them, thereby achieving comprehensive optimization of shooting conditions. Through this progressive adjustment, the system can effectively cope with complex and changing environments and equipment states, significantly improving the robustness and accuracy of detection. Ultimately, if the shooting effect index still fails to reach the preset standard after all automated adjustment attempts, the system will determine that automated adjustment can no longer solve the current problem. At this point, the shooting status control module will immediately control the alarm to sound and request manual intervention from the operator. This provides a reliable fault handling mechanism, ensuring that artificial intelligence can be introduced in a timely manner for diagnosis and resolution beyond the boundaries of automation capabilities, avoiding the system from falling into an ineffective loop and guaranteeing the continuity of the detection task and the final quality. The entire control logic works closely with the motherboard status evaluation module, lighting condition evaluation module, camera status evaluation module, and shooting effect evaluation module in the detection effect optimization system to form a closed-loop intelligent diagnosis and adjustment system, enabling the computer motherboard vision inspection equipment to adaptively optimize its working state in real time, thereby obtaining high-quality inspection images under various working conditions.

[0114] For example, suppose that during the operation of the computer motherboard visual inspection device, the shooting effect index output by the shooting effect evaluation module suddenly drops from 0.9 to 0.6, falling below the preset critical threshold of 0.7. At this time, the shooting state control module will activate its control logic. First, the system executes step S1 to analyze the motherboard status index, lighting health index, and camera status index. Assume that under normal operating conditions, the baseline values ​​for these three indices are 0.95, 0.90, and 0.92, respectively. Currently, the detected motherboard status index is 0.80 (deviation from the baseline value of 0.15), the lighting health index is 0.85 (deviation from the baseline value of 0.05), and the camera status index is 0.70 (deviation from the baseline value of 0.22). Through comparison, the system identifies that the camera status index deviates the most from its own baseline value. Next, the system executes step S2, locking the camera status evaluation module as the primary adjustment target. The shooting state control module will then invoke the automated adjustment program for camera body 2. For example, the program might first trigger the autofocus function of camera body 2 to ensure optimal focus sharpness. If the problem persists, it might activate the lens cleaning mechanism to clean the lens of camera body 2 to improve lens cleanliness. After completing the automated adjustment of the camera status, the system immediately executes step S3 to reassess the shooting effect index. Assuming that after focusing and cleaning, the shooting effect index rises to 0.68, but is still below the critical threshold of 0.7, the system will proceed with the next adjustment in a preset order. It will first check the motherboard status assessment module. Assuming the system detects slight vibrations in the computer motherboard, the shooting status control module will control the clamping mechanism 4 to fine-tune the clamping force on the computer motherboard to reduce the vibration amplitude. The shooting effect index is reassessed again. If the shooting effect index rises to 0.75 at this point, which is above the critical threshold of 0.7, the adjustment process is successful, and the system resumes normal monitoring. However, if the shooting effect index still does not meet the standard, the system will continue to check the lighting condition assessment module in sequence and perform corresponding automated adjustments to the lighting mechanism 3, such as adjusting the brightness or uniformity of the lamps 34. If the shooting effect index still fails to reach 0.7 after all these automatic adjustments, the system will execute step S4. The shooting status control module will control the alarm to issue an audible and visual alarm and display the message "Automatic adjustment is ineffective, please intervene manually" on the operation interface, requesting the operator to intervene and check, for example, whether there is a hardware failure in the camera body 2 or whether the lamp 34 of the lighting mechanism 3 is damaged.

[0115] Through the aforementioned control logic, by locking the module with the largest deviation as the primary control target and invoking the corresponding automated control program, this solution achieves rapid response and efficient intervention. This avoids ineffective adjustments made in traditional methods due to failure to identify the main problem, thereby shortening fault recovery time and improving the operating efficiency of the inspection equipment. Even if the primary control fails to completely resolve the issue, the system can automatically adjust other modules sequentially according to a preset order, ensuring comprehensive coverage and systematic resolution of potential secondary problems, greatly enhancing the system's robustness. Furthermore, when automated control fails to resolve the problem, the system can promptly issue an alarm and request manual intervention, providing a reliable safeguard mechanism for complex or abnormal situations. This human-machine collaborative fault handling mode leverages the efficiency advantages of automated systems while retaining the flexibility and in-depth diagnostic capabilities of manual intervention, ensuring high-quality inspection standards are maintained under all circumstances. Overall, this control logic endows the computer motherboard visual inspection equipment with stronger adaptive and fault self-healing capabilities, significantly improving the stability and automation level of the inspection process, thereby guaranteeing the inspection quality and production efficiency of computer motherboards.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A computer motherboard visual inspection device, comprising an inspection box (1), characterized in that, The detection box (1) has a camera body (2) at the top center, a lighting mechanism (3) around the camera body (2), and a clamping mechanism (4) for fixing the computer motherboard at the bottom. It also includes a detection effect optimization system, which comprises: Motherboard Status Assessment Module: Based on the vibration amplitude of the computer motherboard, the interference of specular reflection, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane, a motherboard status assessment model is constructed, and the motherboard status index is output. Lighting condition assessment module: Based on the light uniformity, irradiation stability and light intensity of the lighting mechanism (3), a lighting condition assessment model is constructed and the light health index is output; Shooting Condition Assessment Module: The shooting condition assessment model is constructed based on the motherboard status index and light health index as the main conditions, and the ambient humidity and ambient temperature as influencing factors, and the shooting condition index is output. Camera status assessment module: Based on the noise level of the camera body (2), the cleanliness of the lens and the effective focus sharpness, a camera status assessment model is constructed and the camera status index is output. Shooting effect evaluation module: Constructs a shooting effect evaluation model based on shooting condition index and camera status index, and outputs the shooting effect index; Shooting status control module: continuously monitors the shooting effect index; when the shooting effect index is less than the preset critical threshold, the system analyzes and adjusts the relevant influencing parameters in the motherboard status evaluation module, the lighting condition evaluation module, and the camera status evaluation module.

2. The computer motherboard visual inspection device according to claim 1, characterized in that, The lighting mechanism (3) includes a fixing ring (31), a connecting rod (32), a No. 1 motor (33), and a lamp (34); The fixing ring (31) is fixedly connected to the inner top surface of the detection box (1). The fixing ring (31) is fixedly connected to multiple connecting rods (32) at equal intervals. One end of each connecting rod (32) is rotatably connected to the output shaft of the No. 1 motor (33). Each No. 1 motor (33) is fixedly connected to a lamp (34). All lamps (34) are set at equal angles.

3. The computer motherboard visual inspection device according to claim 1, characterized in that, The clamping mechanism (4) includes an electric push rod (41), a fixed frame (42), a second motor (43), and a clamping plate (44). The inner side wall of the test box (1) is provided with two electric push rods (41) facing each other. The telescopic ends of the electric push rods (41) are fixedly connected to the fixing frame (42). The fixing frame (42) is fixedly connected to the second motor (43). The output shaft of the second motor (43) is fixedly connected to the clamping plate (44). The clamping plate (44) can clamp and fix the computer motherboard.

4. The computer motherboard visual inspection device according to claim 1, characterized in that, The vibration amplitude of the computer motherboard, the specular reflection interference, and the relative angular deviation between the motherboard's measured surface and the camera's imaging plane are successively substituted into the maximum-minimum normalization formula for normalization, and the vibration amplitude index, specular reflection interference index, and angular deviation index are generated sequentially; in the motherboard condition assessment model: The vibration amplitude index, specular reflection interference index, and angle deviation index are weighted and summed with preset vibration amplitude weighting coefficients, specular reflection interference weighting coefficients, and angle deviation weighting coefficients, respectively. The motherboard status index is then calculated by subtracting the weighted sum from 1.

5. The computer motherboard visual inspection device according to claim 1, characterized in that, The illumination uniformity, illumination stability, and luminance of the illumination mechanism (3) are successively substituted into the maximum-minimum normalization formula for normalization, and the illumination uniformity index, illumination stability index, and luminance index are generated in sequence; in the illumination condition evaluation model: First, the illumination uniformity index, illumination stability index, and illumination brightness index are weighted and summed with preset illumination uniformity weighting coefficients, illumination stability weighting coefficients, and illumination brightness weighting coefficients respectively to obtain the basic illumination index. Then, based on the degree to which the product of the light uniformity index and the irradiation stability index deviates from the ideal value of 1, the basic light index is reduced by multiplying it by a penalty term adjusted by a preset synergy coefficient, thereby calculating the light health index.

6. The computer motherboard visual inspection device according to claim 1, characterized in that, The ambient humidity and ambient temperature are successively substituted into the maximum-minimum normalization formula for normalization, and the humidity index and temperature index are generated accordingly; in the shooting condition evaluation model: First, the motherboard status index and the lighting health index are weighted and summed to obtain the basic condition index; Then, an environmental factor is constructed using an exponential decay function. This environmental factor is a decay coefficient between 0 and 1. It is calculated by using the natural constant e as the base, and the sum of the negative humidity influence weighting coefficient multiplied by (the square of the difference between the humidity index and the optimal humidity threshold) and the negative temperature influence weighting coefficient multiplied by (the square of the difference between the temperature index and the optimal temperature threshold) as the exponent, and then performing a power operation. The final shooting condition index is the product of the basic condition index and the environmental factor.

7. The computer motherboard visual inspection device according to claim 1, characterized in that, The noise level of the camera body (2), the lens cleanliness and the effective focus sharpness are substituted into the maximum-minimum normalization formula for normalization, and the noise intensity index, lens cleanliness index and focus sharpness index are generated in sequence. In the camera state evaluation model: First, the product of the lens cleanliness index and the focus sharpness index is used as a co-term, and its deviation from the ideal value of 1 is calculated. Then, the noise intensity index, lens cleanliness index, and focus sharpness index are weighted and summed with preset noise impact weighting coefficients, lens impact weighting coefficients, and focus impact weighting coefficients respectively to obtain the basic camera index. Finally, the deviation of the collaborative term is amplified by a preset collaborative influence factor, and the basic camera index is reduced to calculate the camera state index.

8. The computer motherboard visual inspection device according to claim 1, characterized in that, In the shooting effect evaluation model: First, the product of the shooting condition index and the camera state index is used as a co-term, and its deviation from the ideal value of 1 is calculated. Then, the shooting condition index and camera status index are weighted and summed with the preset shooting condition weight coefficient and camera status weight coefficient respectively to obtain the basic effect index. Finally, the deviation of the collaborative item is amplified by a preset collaborative penalty factor, and the basic effect index is reduced to calculate the shooting effect index.

9. The computer motherboard visual inspection device according to claim 1, characterized in that, The control logic in the shooting status control module is as follows: S1: Analyze which of the three indices—motherboard status index, lighting health index, and camera status index—deviates the most from its own baseline value; S2: Lock the module with the largest deviation as the primary adjustment target and call the corresponding automatic adjustment program; S3: After adjustment, re-evaluate the shooting effect index. If it still does not meet the standard, adjust it in the following order: motherboard status evaluation module, lighting condition evaluation module, and camera status evaluation module. S4: If automatic adjustment is ineffective, the shooting status control module will trigger the alarm and request manual intervention.

Images