Straightness error measuring system for linear rolling guide rail pair

Abstract

The invention relates to the field of linear rolling guide rails, and discloses a linear rolling guide rail pair straightness error measuring system which comprises a workbench, the top of the workbench is fixedly connected with a laser interferometer, the top of the workbench is fixedly connected with a measured guide rail, the top of the measured guide rail is slidably connected with a concave frame, and the concave frame is fixedly connected with the laser interferometer. The concave frame is provided with a CCD camera, a beam splitter prism and a moving target. A modularized framework of light source emission, light splitting, reflection receiving and data processing is adopted, all the modules are in precise cooperation, the laser interferometer provides a stable reference light path, the beam splitter prism constructs a high-precision interference light path, the CCD camera achieves sub-pixel-level light spot collection, and the laser interferometer provides a high-precision interference light path. The data processing module achieves real-time calculation of straightness errors through a gray scale weighted centroid method and a least square method, the dynamic measurement requirement of the high-speed guide rail (2 m / s) is met, the calculation efficiency and precision reach the industry leading level, and a reliable solution is provided for online detection and machining process control of the precise guide rail.

Description

Technical Field

[0001] This invention relates to the field of linear rolling guide technology, specifically to a linear rolling guide pair straightness error measurement system. Background Technology

[0002] Linear rolling guide pairs are a new type of mechanical transmission component. The guide rail and the slider are in contact through rolling elements, replacing the traditional sliding friction with rolling friction. Therefore, the coefficient of friction of linear rolling guides is one-fiftieth that of traditional sliding guides. This not only improves the motion accuracy and speed of the slider, but also extends the service life of linear rolling guide pairs.

[0003] Linear rolling guides, as fundamental components for precision positioning, automatic control, drive mechanisms, and motion conversion, possess advantages such as low friction coefficient, smooth motion, and high motion accuracy, thus finding widespread application in various fields of modern industry. With the rapid growth of my country's economy and the continuous expansion of various industrial sectors such as urban rail transit, automobiles, shipbuilding, and aerospace, a huge market space has been created for linear rolling guide products. As the demand for linear rolling guides continues to increase, the demand for higher-end products is also rising, thus placing higher demands on the manufacturing process of linear rolling guides.

[0004] Traditional straightness error measurement relies heavily on optical equipment such as laser interferometers and CCD cameras, calculating guide rail errors through spot offset. However, these techniques are susceptible to significant interference from dust contamination and environmental reflections. Dust contamination: In industrial settings, dust (such as metal processing debris and particulate pollutants in the air) easily adheres to the surface of optical lenses (beam splitters, mirrors), causing laser beam scattering and light intensity attenuation. This results in low-grayscale occlusion areas in the acquired spot image, causing centroid calculation deviations. For example, when the dust coverage area on the mirror surface reaches 10%, the measurement error can increase to 3-5 times the normal level. Environmental reflections: Interference from workshop lighting and reflections from equipment metal surfaces introduces background noise, leading to blurred spot edges and abnormal grayscale values. Traditional methods struggle to dynamically distinguish between the real signal and reflected noise. In highly reflective environments (such as direct sunlight or high-brightness LED lighting), the spot center positioning error can exceed 5μm, severely impacting measurement reliability. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a linear rolling guide pair straightness error measurement system. This system solves the problem that traditional straightness error measurements often rely on optical equipment such as laser interferometers and CCD cameras, calculating guide rail errors through spot offset. However, such technologies are susceptible to significant interference from dust contamination and environmental reflections.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a linear rolling guide pair straightness error measuring device, comprising a worktable, a laser interferometer fixedly connected to the top of the worktable, a guide rail to be measured fixedly connected to the top of the worktable, a concave frame slidably connected to the top of the guide rail to be measured, and a CCD camera, a beam splitter prism, and a moving target arranged on the concave frame.

[0007] Preferably, the concave frame is provided in two sets, one set of which has a CCD camera fixedly connected to the top, and the other set of which has a moving target fixedly connected to the top.

[0008] Preferably, a beam splitter is fixedly connected to the top of the CCD camera.

[0009] A linear rolling guide pair straightness error measurement system includes a light source emitting module, a light source beam splitting module connected to the light source emitting module, a reflection and receiving module connected to the light source beam splitting module, a data processing module connected to the reflection and receiving module, and an external environment anti-interference module connected to the reflection and receiving module.

[0010] Preferably, the external environment anti-interference module includes: a reflection suppression module; the reflection suppression module suppresses reflection interference in the external environment through an algorithm.

[0011] Preferably, the algorithm in the reflection suppression module is as follows:

[0012]

[0013] in:

[0014] I raw (x,y) represents the original light intensity signal, which is the original grayscale value collected by the CCD camera at pixel (x,y), containing the actual laser spot signal and reflection noise interference.

[0015] B(x,y0) represents dynamic background noise. The system continuously acquires background images when there is no laser by using a noise sensor, calculates the average gray value of each pixel, and characterizes the base noise of environmental reflection.

[0016] δ represents the noise compensation factor, which is the minimum offset used to prevent the denominator from approaching zero. It is usually set to the standard deviation of the system noise (5-10).

[0017] K p Represents the polarization correction coefficient, a dynamic coefficient determined by the polarizer angle;

[0018] `max` represents the maximum function, compared to the current light intensity I. raw The larger of the two values, B and B, is used as the denominator.

[0019] I final (x,y) represents the calculated value of reflective grayscale suppression.

[0020] Preferably, the external environment anti-interference module includes a dust compensation module, which compensates for dust interference in the external environment through an algorithm.

[0021] Preferably, the algorithm in the dust compensation module is as follows:

[0022] W(x,y)=max(0,I(x,y)-γ·μ env )

[0023] Where: I(x,y) represents the light intensity distribution containing dust, the gray value of the spot image after reflection suppression, including the dust-occluded area (low gray value part);

[0024] γ represents the dust detection threshold coefficient, which controls the sensitivity of dust detection. The larger the value, the easier it is to detect dust (1.2 to 1.5 is recommended).

[0025] μ env The ambient light intensity is estimated by the average gray value of the background environment and the average gray value of the ring-shaped area around the spot (1.5 times the diameter of the spot), and is used to compare and determine dust occlusion.

[0026] W(x,y) represents the effective region weight matrix, which suppresses gray levels below the threshold γ·μ. env The pixels (i.e., those marked as dust areas) retain valid spot signals for centroid calculation;

[0027] x c ,y c Represents the corrected centroid coordinates, calculated only based on the effective spot area (high grayscale portion) to eliminate the offset caused by dust occlusion.

[0028] This invention provides a system for measuring the straightness error of a linear rolling guide pair. It has the following advantages:

[0029] 1. This invention accurately separates the laser spot signal from environmental reflection noise through dynamic background noise modeling and a real-time reflection suppression algorithm. It acquires background images in real-time when there is no laser, calculates the average grayscale value of each pixel as dynamic background noise, and combines a polarization correction coefficient and a noise compensation factor to suppress reflection interference pixel by pixel.

[0030] 2. The dust compensation module of this invention sets the dust detection threshold based on the average grayscale value of the annular region surrounding the light spot, uses a weight matrix to shield dust-occluded areas, and calculates the centroid coordinates only through the effective light spot area. This is applicable when the dust concentration is 5 mg / m³. 3In this environment, the measurement error is reduced by 80% compared to the uncompensated scheme. Even when the spot is obscured by dust by 30%, the deviation is still less than 2μm. By adjusting the dust detection threshold coefficient (1.2-1.5), the false detection and false negative issues under different dust environments can be balanced.

[0031] 3. This invention adopts a modular architecture encompassing light source emission, beam splitting, reflection reception, and data processing, with each module working in precise coordination: a laser interferometer provides a stable reference optical path, a beam splitter prism constructs a high-precision interference optical path, a CCD camera achieves sub-pixel-level spot acquisition, and the data processing module uses the gray-scale weighted centroid method and least squares method to calculate straightness error in real time (single frame processing time <10ms). Combined with the dynamic algorithm of the external environment anti-interference module, the system ensures a measurement accuracy of ±1μm while meeting the dynamic measurement requirements of high-speed guide rails (2m / s). Its computational efficiency and accuracy reach industry-leading levels, providing a reliable solution for online inspection and processing control of precision guide rails. Attached Figure Description

[0032] Figure 1 This is a perspective view of the device of the present invention;

[0033] Figure 2 This is a schematic diagram of the system flow of the present invention.

[0034] The components include: 1. worktable; 2. concave frame; 3. guide rail to be measured; 4. moving target; 5. beam splitter; 6. laser interferometer; and 7. CCD camera. Detailed Implementation

[0035] The technical solutions in the embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings.

[0036] Example 1:

[0037] Please see the appendix Figure 1 -Appendix Figure 2This invention provides a device for measuring the straightness error of a linear rolling guide pair, including a worktable 1. The worktable 1 provides support for the entire device, ensuring its stability during use. A laser interferometer 6 is fixedly connected to the top of the worktable 1, providing support for the laser interferometer 6 and ensuring its stability during use. A guide rail 3 is fixedly connected to the top of the worktable 1, providing guidance for a concave frame 2. The concave frame 2 is slidably connected to the top of the guide rail 3. Two sets of concave frames 2 are provided. A CCD camera 7 is fixedly connected to the top of one set of concave frames 2. When one set of concave frames 2 moves, it drives the CCD camera 7 and the beam splitter prism 5 located in the middle to move laterally simultaneously, achieving the effect of changing their lateral position. The top of the CCD camera 7 is fixedly connected to... A beam splitter 5 is used, with its splitting surface at a 45° angle to the horizontal plane. A CCD camera 7 is mounted directly below the beam splitter 5. This ensures that one laser beam emitted by the laser interferometer 6 is reflected after passing through the beam splitter 5, while the other beam, after propagating back through the optical path, lands on the photosensitive surface of the CCD camera 7. A movable target 4 is fixedly connected to the top of another set of concave supports 2. Moving the movable target 4 by moving the other set of concave supports 2 causes it to move laterally. A plane mirror is fixedly connected to the movable target 4 on the guide rail, allowing the light to propagate parallel to the guide rail and be reflected to the CCD camera 7 for reception. The coordinates of the light spot center are extracted using image processing methods to obtain the relative offset between the light spot center at each measurement position and the initial position. The straightness error of the measured guide rail is then calculated using the least squares method.

[0038] A linear rolling guide pair straightness error measurement system includes a light source emitting module, which generates a stable laser beam to provide a reference optical path for straightness measurement, and is usually set in a laser interferometer (6);

[0039] A light source emission module is connected to a light source beam splitting module, which in turn is connected to a reflection and receiving module, and the reflection and receiving module is connected to a data processing module. The light source beam splitting module divides the laser beam into a measurement beam and a reference beam, constructing an interference optical path, specifically as follows:

[0040] One beam of light is reflected back, while the other, according to the principles of light propagation, reaches the plane mirror and returns along the same path. It is then received by the CCD camera via the beam splitter, acquiring an initial light spot. After the plane mirror moves a certain distance with the moving target mounted on the slider, it acquires another light spot. Theoretically, when the linear guide pair has no straightness error, the two light spots can perfectly overlap. However, in actual measurement, due to the straightness of the guide rails, the light spots cannot perfectly overlap; there is a certain offset d between the centers of the light spots.

[0041] d=D·tanθ

[0042] Where D represents the optical path of the laser, and θ represents the deflection angle at the measurement position. The basic principle is to obtain the straightness error information of the guide rail based on the moving distance of the plane mirror and the offset between the center of the laser spot at each measurement point and the center of the laser spot at the initial position. The offset d between the centers of the two laser spots can be calculated using the following formula:

[0043]

[0044] In the formula, (x i ,y i Let (x0, y0) be the coordinates of the center of the light spot at any position on the guide rail; and (x0, y0) be the coordinates of the center of the light spot at the initial position. When evaluating the straightness error of the guide rail, the center of the light spot measured at each measurement point on the guide rail is parallel to the center line where the center of the light spot is located at the initial position, and is defined by d... max Within a cylindrical surface with radius [radius];

[0045] An external environment anti-interference module is connected to the reflection and reception module. The external environment anti-interference module includes: a reflection suppression module.

[0046] The reflection suppression module uses an algorithm to suppress reflective interference from the external environment. The algorithm is as follows:

[0047]

[0048] in:

[0049] I raw (x,y) represents the original light intensity signal, which is the original grayscale value collected by the CCD camera at pixel (x,y), containing the actual laser spot signal and reflection noise interference.

[0050] Example: If the grayscale value at the center of the laser spot is 200, and the grayscale value of the ambient reflective area is 80, then I raw (200, 150) = 200;

[0051] B(x,y0) represents dynamic background noise. The system continuously acquires background images when there is no laser by using a noise sensor, calculates the average gray value of each pixel, and characterizes the base noise of environmental reflection.

[0052] Example: If the average background value of a point for 5 consecutive frames is 50, then B(200,150) = 50;

[0053] δ represents the noise compensation factor, which is the minimum offset used to prevent the denominator from approaching zero. It is usually set to 5-10 of the standard deviation of the system noise.

[0054] K p Represents the polarization correction coefficient, a dynamic coefficient determined by the polarizer angle;

[0055] `max` represents the maximum function, compared to the current light intensity I. raw The larger of the two values, B and B, is used as the denominator.

[0056] I final (x,y) represents the calculated value of reflective grayscale suppression;

[0057] The above formula is used to calculate the reflection grayscale suppression value, which is then used to adjust the reflection and receiving modules to achieve the effect of anti-reflection interference.

[0058] Beneficial effects of Example 1: Example 1 accurately separates the laser spot signal from environmental reflective noise through a reflective suppression algorithm. The system acquires background images in real time when there is no laser, calculates dynamic background noise, and suppresses reflective interference pixel by pixel by combining polarization correction coefficients and noise compensation factors. This design enables the system to control the measurement error within ±0.5μm in environments where the reflective intensity reaches 30% of the original signal, improving accuracy by more than 50% compared to traditional methods. It can also adapt to changes in ambient light, significantly reducing measurement drift, and improving the edge sharpness of the laser spot by 40% in highly reflective scenes, balancing computational efficiency and anti-interference performance.

[0059] Example 2: Based on Example 1, Example 2 modifies the external environment anti-interference module to provide a linear rolling guide pair straightness error measurement system, which includes an external environment anti-interference module, including a dust compensation module.

[0060] The dust compensation module compensates for dust interference from the external environment using an algorithm. The algorithm is as follows: W(x,y)=max(0,I(x,y)-γ·μ env )

[0061] Where: I(x,y) represents the light intensity distribution containing dust, the gray value of the spot image after reflection suppression, which may include the low gray value part of the dust-occluded area;

[0062] γ represents the dust detection threshold coefficient, which controls the sensitivity of dust detection. The larger the value, the easier it is to detect dust. A value of 1.2 to 1.5 is recommended.

[0063] μ env The ambient light intensity is estimated by taking the mean gray value of the background light and using the mean gray value of the outer ring area around the light spot (1.5 times the diameter of the light spot) as a reference. This is used to compare and determine dust occlusion.

[0064] W(x,y) represents the effective region weight matrix, which suppresses gray levels below the threshold γ·μ. env The pixels are marked as dust areas, and the effective spot signals are retained for centroid calculation;

[0065] x c ,y c Represents the corrected centroid coordinates, calculated only based on the high grayscale portion of the effective spot area, eliminating the offset caused by dust occlusion;

[0066] The corrected centroid coordinates are calculated using the above formula, and the reflection and receiving modules are then corrected according to these corrected centroid coordinates, thereby achieving the effect of resisting dust interference.

[0067] Example 2's beneficial effects: The dust detection and centroid correction algorithm effectively addresses dust occlusion interference. The system sets a dust detection threshold based on the average grayscale value of the annular region surrounding the light spot, uses a weight matrix to shield dust-occluded areas, and calculates the centroid coordinates only from the effective light spot. Experiments show that at a dust concentration of 5 mg / m³... 3 In this environment, the measurement error is reduced by 80% compared to the uncompensated scheme. When the area of ​​the light spot being blocked reaches 30%, the deviation is still less than 2μm. Furthermore, by adjusting the threshold coefficient, the false detection and missed detection problems under different dust environments can be balanced. After working together with the reflection suppression module, the measurement uncertainty in the compound interference environment is as low as 0.8μm, and the anti-interference capability is significantly improved.

[0068] Comparative experiment:

[0069] To verify the advantages of Example 1 (reflection suppression algorithm) and Example 2 (dust compensation algorithm) over the prior art, the following comparative experiments were designed, covering scenarios with reflective interference, dust occlusion, and combined interference.

[0070] I. Experimental Preparation

[0071] Test device

[0072] Test object: Linear rolling guide pair (length 2m, accuracy class ISO P3).

[0073] Control group:

[0074] Existing technology: Spot extraction algorithm based on average background subtraction (no polarization correction, no dynamic threshold compensation).

[0075] Experimental Group 1: Example 1 (Reflection Suppression Algorithm, K) p =0.8, δ=5)

[0076] Experimental Group 2: Example 1 + Example 2 (working together, γ = 1.3)

[0077] Instruments and equipment:

[0078] Laser interferometer (reference accuracy 0.1μm).

[0079] CCD camera (resolution 2448×2048, frame rate 200Hz).

[0080] Simulated reflector (surface specular reflectivity >90%).

[0081] Adjustable dust generator.

[0082] Interference environment configuration

[0083] Reflective interference scenario: A reflector is placed 30cm above the guide rail to simulate the reflection of a metal surface (light intensity interference of about 30%).

[0084] Dust-obstructed scenario: Spray standard dust (particle size 20μm, concentration 5mg / m³) onto the light spot area. 3 ).

[0085] Composite interference scenario: simultaneous application of reflective and dust interference.

[0086] Parameter benchmarking

[0087]

[0088] II. Experimental Procedure

[0089] Scene 1: Comparison of Reflective Noise Suppression

[0090] step:

[0091] 1. Turn off the dust generator and turn on the reflector to simulate a strong reflective environment.

[0092] 2. Drive the target to move at a constant speed along the guide rail 10 times (stroke 100mm, speed 5mm / s).

[0093] 3. Record the deviations between the measurement trajectories of the existing technology and Example 1 and the reference values ​​of the laser interferometer. Key indicators:

[0094] Maximum error at a single point.

[0095] Mean absolute error (MAE) over the entire process.

[0096] Spot signal-to-noise ratio (SNR).

[0097] Scenario 2: Comparison of Dust Occlusion Compensation

[0098] step:

[0099] 1. Turn off the reflector and turn on the dust generator to spray dust with a particle size of 20μm (concentration 5mg / m³). 3 ).

[0100] 2. Manually block the central area of ​​the light spot (accounting for 30%).

[0101] 3. Repeatedly move the target and compare the centroid offset of the uncompensated target with that of Example 2.

[0102] Key metrics:

[0103] Centroid offset (μm) of the occluded area.

[0104] Corrected deviation reduction rate.

[0105] Scenario 3: Composite Interference Test

[0106] step:

[0107] 1. Simultaneously turn on the reflector and dust generator (concentration 20mg / m³). 3 ).

[0108] 2. The target moves at a speed of 5 mm / s, randomly triggering light spot occlusion (accounting for 15-30%).

[0109] 3. Record the cumulative error of the three algorithms under combined interference.

[0110] 4. Analyze the dynamic anti-interference stability (error standard deviation).

[0111] III. Experimental Results and Data Analysis

[0112] 1. Comparison of reflectivity suppression effects

[0113]

[0114] in conclusion:

[0115] The reflection noise suppression algorithm in Example 1 significantly reduces the error (accuracy improvement ≥65%) and improves SNR by 14dB.

[0116] The increased latency stems from dynamic background calculations, but it still meets real-time requirements at a frame rate of 200Hz.

[0117] 2. Comparison of Dust Compensation Effects

[0118]

[0119] in conclusion:

[0120] The dust compensation algorithm in Example 2 achieves an 80% correction rate for 30% occlusion and improves dynamic occlusion stability by 75%.

[0121] 3. Comprehensive Comparison of Composite Interference

[0122]

[0123]

[0124] in conclusion:

[0125] By using the two algorithms together, the measurement uncertainty under combined interference is reduced to ±0.8μm, and the standard deviation of the error is reduced to ±0.3μm.

[0126] IV. Core Parameters and Phenomenon Analysis

[0127] 1. Polarization correction in reflection suppression

[0128] When the polarizer is not aligned (K p When the SNR is 0.5, the SNR drops to 25 dB and the error increases to ±2.1 μm.

[0129] Optimize K p After reaching 0.8, the signal-to-noise ratio and positioning accuracy improved simultaneously, verifying the necessity of polarization angle calibration.

[0130] 2. Dust compensation threshold sensitivity

[0131] When γ = 1.5, the false negative rate for dust particles ≤ 5μm reaches 20%; when γ = 1.2, the false negative rate is < 5%, but the false positive rate increases to 8%.

[0132] The optimal balance is achieved when γ = 1.3 (false negative rate 7%, false positive rate 3%).

[0133] V. Experimental Conclusions

[0134] Example 1 (Reflection Suppression): In strong reflective scenarios, the error is reduced by more than 65% compared to traditional methods, and the SNR is improved to the 30dB level, supporting precision industrial measurement.

[0135] Example 2 (Dust Compensation): When there is a large area of ​​obstruction (30%), the centroid correction rate reaches 80%, which can cope with long-term stable operation in dusty environments.

[0136] Collaborative anti-interference: After the two algorithms are superimposed, the measurement uncertainty under combined interference is compressed to the submicron level (±0.8μm), which meets the detection requirements of high-precision guide rail straightness (ISO standard P1 level).

[0137] Further evidence shows that compared to existing technologies, this solution can increase the availability of the measurement system from 60% to over 95% in environments with interference such as reflection and dust, significantly reducing the cost of manual re-inspection and downtime.

[0138] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A device for measuring the straightness error of a linear rolling guide pair, characterized in that, The device includes a worktable (1), a laser interferometer (6) fixedly connected to the top of the worktable (1), a test guide rail (3) fixedly connected to the top of the worktable (1), a concave frame (2) slidably connected to the top of the test guide rail (3), and a CCD camera (7), a beam splitter (5) and a moving target (4) are provided on the concave frame (2).

2. The linear rolling guide pair straightness error measuring device according to claim 1, characterized in that, The concave frame (2) is provided in two sets. One set of the concave frame (2) has a CCD camera (7) fixedly connected to the top, and the other set of the concave frame (2) has a moving target (4) fixedly connected to the top.

3. The linear rolling guide pair straightness error measuring device according to claim 2, characterized in that, A beam splitter (5) is fixedly connected to the top of the CCD camera (7).

4. A linear rolling guide pair straightness error measurement system according to any one of claims 1-3, characterized in that, It includes a light source emitting module, a light source beam splitting module connected to the light source emitting module, a reflection and receiving module connected to the light source beam splitting module, a data processing module connected to the reflection and receiving module, and an external environment anti-interference module connected to the reflection and receiving module.

5. The linear rolling guide pair straightness error measurement system according to claim 4, characterized in that, The external environment anti-interference module includes a reflection suppression module; the reflection suppression module uses an algorithm to suppress reflection interference present in the external environment.

6. The linear rolling guide pair straightness error measurement system according to claim 5, characterized in that, The algorithm in the reflection suppression module is as follows: in: I raw (x,y) represents the original light intensity signal, which is the original grayscale value collected by the CCD camera at pixel (x,y), containing the actual laser spot signal and reflection noise interference. B(x,y0) represents dynamic background noise. The system continuously acquires background images when there is no laser by using a noise sensor, calculates the average gray value of each pixel, and characterizes the base noise of environmental reflection. δ represents the noise compensation factor, which is the minimum offset used to prevent the denominator from approaching zero. It is usually set to the standard deviation of the system noise (5-10). K p Represents the polarization correction coefficient, a dynamic coefficient determined by the polarizer angle; `max` represents the maximum function, compared to the current light intensity I. raw The larger of the two values, B and B, is used as the denominator. I final (x,y) represents the calculated value of reflective grayscale suppression.

7. The linear rolling guide pair straightness error measurement system according to claim 4, characterized in that, The external environment anti-interference module includes a dust compensation module, which compensates for dust interference in the external environment through an algorithm.

8. The linear rolling guide pair straightness error measurement system according to claim 7, characterized in that, The algorithm in the dust compensation module is as follows: Where: I(x,y) represents the light intensity distribution containing dust, the gray value of the spot image after reflection suppression, including the dust-occluded area (low gray value part); γ represents the dust detection threshold coefficient, which controls the sensitivity of dust detection. The larger the value, the easier it is to detect dust (1.2 to 1.5 is recommended). μ env The ambient light intensity is estimated by the average gray value of the background environment and the average gray value of the ring-shaped area around the spot (1.5 times the diameter of the spot), and is used to compare and determine dust occlusion. W(x,y) represents the effective region weight matrix, which suppresses gray levels below the threshold γ·μ. env The pixels (i.e., those marked as dust areas) retain valid spot signals for centroid calculation; x c ,y c Represents the corrected centroid coordinates, calculated only based on the effective spot area (high grayscale portion) to eliminate the offset caused by dust occlusion.

Images