A visual inspection system for high-pressure compressor impellers

Abstract

The present invention belongs to the technical field of impeller detection, and specifically relates to a high-pressure compressor impeller visual inspection system, which realizes dynamic lighting adjustment, high-quality image capture and accurate color reproduction by introducing a reflection control module, a contrast enhancement module, an exposure correction module and a color balance module. The reflection control module adjusts the lighting angle and intensity in real time according to the material characteristics of the impeller to ensure that the light evenly covers the entire impeller surface and reduces the interference caused by high reflectivity; the contrast enhancement module and the exposure correction module work together to ensure that the image has a high dynamic range and an appropriate exposure level under different lighting conditions to avoid overexposure and underexposure; the color balance module ensures accurate color reproduction and further improves image quality. These optimization processes enable the defect recognition module to more accurately locate and classify potential defects on the impeller surface, thereby significantly improving the overall performance and reliability of the visual inspection system.

Description

Technical Field

[0001] The invention belongs to the technical field of impeller detection, and in particular relates to a high-pressure compressor impeller visual detection system. Background Art

[0002] During the manufacturing and maintenance of high-pressure compressor impellers, visual inspection is a key step to ensure their quality. Existing visual inspection systems usually include a light source module, an image capture module, and some basic image processing modules. The basic workflow of these systems is to provide illumination through the light source module, then use the image capture module to obtain an image of the impeller surface, and finally perform preliminary analysis through a simple image processing algorithm.

[0003] However, the existing technology has a significant technical problem: due to the high reflectivity of the impeller material to a specific spectrum, overexposure or insufficient contrast often occurs during image capture. This phenomenon causes loss of image details, making subsequent defect identification difficult or even unreliable. Specifically, when the lighting angle and intensity are not appropriate, the highly reflective areas will be overexposed, while the low-reflective areas may be underexposed, thus affecting the overall image quality and detection accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a high-pressure compressor impeller visual inspection system, which not only includes a traditional light source module and an image capture module, but also introduces a reflection control module, a contrast enhancement module, an exposure correction module and a color balance module; through the coordinated work of these newly added modules, the system can dynamically adjust the lighting conditions and optimize the dynamic range, exposure parameters and color reproduction of the image, thereby effectively avoiding the problems of overexposure and insufficient contrast.

[0005] To achieve the above object, the present invention adopts the following technical solution: a high-pressure compressor impeller visual inspection system, comprising:

[0006] A light source module is used to provide lighting; a reflection control module connected to the light source module is used to adjust the lighting angle and intensity to adapt to the material characteristics of the impeller; the reflection control module is connected to the image capture module, which is responsible for capturing the detailed image of the impeller surface; the output end of the image capture module is connected to the contrast enhancement module, which is used to adjust the dynamic range according to the captured image data; the contrast enhancement module signal is transmitted to the exposure correction module, which is used to optimize the exposure parameters to avoid overexposure; the exposure correction module is connected to the color balance module to ensure accurate color reproduction under different light; the color balance module is connected to the defect recognition module to locate the potential defects on the impeller surface.

[0007] Preferably, the light source module is further used to perform the steps:

[0008] Receive initial lighting parameters including light intensity I_initial and angle A_initial;

[0009] Based on the received initial illumination parameters, the ideal illumination conditions adapted to the material characteristics of the impeller are calculated, where the light intensity adjustment coefficient K satisfies: K=(I_target / I_initial), where I_target is the target light intensity;

[0010] Using the calculated ideal lighting conditions, adjust the position and angle of the optical elements in the reflection control module so that the light can irradiate the impeller surface at the optimal angle; the optimal angle A_optimal satisfies: A_optimal=A_initial+ΔA, ΔA is the angle adjustment amount;

[0011] The image data captured under the adjusted lighting conditions is passed to the image capture module.

[0012] Preferably, the reflection control module is further used to execute the steps:

[0013] Receiving a light signal under optimized lighting conditions from a light source module, including an adjusted light intensity I_adjusted and an optimal angle A_optimal;

[0014] Based on the received light signal, the lighting angle and intensity are monitored and adjusted in real time;

[0015] Using the real-time monitoring data, the position and angle of the optical elements in the reflection control module are dynamically adjusted. The new angle adjustment ΔA_new satisfies: ΔA_new=A_optimal-A_current, where A_current is the current actual angle.

[0016] The image data captured under the adjusted lighting conditions is passed to the image capture module, and the working status of the reflection control module is continuously optimized through a real-time monitoring feedback loop.

[0017] Preferably, the image capture module is further used to perform the steps:

[0018] receiving a light signal under the lighting conditions optimized by the reflection control module;

[0019] Based on the received light signal, adjust the camera parameters in the image capture module, including the exposure time T and the gain value G. The exposure time T satisfies: T=T_base*(I_target / I_received), G=G_base*(I_target / I_received), where T_base and G_base are initial values, I_target is the target light intensity, and I_received is the actual light intensity received;

[0020] Using the adjusted camera parameters, the image capture module is started to capture the detailed image of the impeller surface;

[0021] The captured image data is passed to the contrast enhancement module, and the working status of the image capture module is monitored and adjusted in real time through the internal calibration mechanism.

[0022] Preferably, the contrast enhancement module is further used to perform the steps:

[0023] receiving image data from an image capture module;

[0024] Based on the received image data, the brightness value distribution of each pixel in the image is calculated to determine the minimum brightness L_min and the maximum brightness L_max for subsequent dynamic range adjustment;

[0025] Using the calculated brightness value distribution, the global contrast enhancement method is applied to adjust the overall contrast of the image. The new brightness value L_new satisfies:

[0026] L_new=(L-L_min)*(L_max_new-L_min_new) / (L_max-L_min)+L_min_new, where L is the original brightness value, L_max_new and L_min_new are the maximum and minimum values ​​of the target brightness;

[0027] Pass the processed image to the exposure correction module.

[0028] Preferably, the exposure correction module is further used to perform the steps:

[0029] receiving image data processed by a contrast enhancement module;

[0030] Based on the received image data, the actual brightness value of each pixel in the image is calculated, and the over-exposed area and the under-exposed area are determined, and the over-exposed threshold is LO_over and the under-exposed threshold is LO_under;

[0031] Using the determined over-exposed and under-exposed area information, adjust the exposure parameters in the exposure correction module, and the new exposure time T_new satisfies: T_new=T_old*(L_target / L_current), where T_old is the original exposure time, L_target is the target brightness value, and L_current is the current actual brightness value;

[0032] The adjusted exposure parameters are applied to subsequent image capture processes, and the image brightness distribution is continuously monitored.

[0033] Preferably, the color balancing module is further used to perform the steps:

[0034] receiving image data processed by the exposure correction module;

[0035] Based on the received image data, analyze the brightness distribution of the red, green and blue color channels in the image, and determine the average brightness values ​​R_avg, G_avg and B_avg of each channel;

[0036] Using the average brightness value of each channel, adjust the gain coefficient of each channel. The new gain coefficients are Gain_R, Gain_G and Gain_B, which satisfy: Gain_R = (R_target / R_avg), Gain_G = (G_target / G_avg), Gain_B = (B_target / B_avg), where R_target, G_target and B_target are the target brightness values;

[0037] The adjusted gain factor is applied to the color channel of the image and the color reproduction is continuously optimized through real-time monitoring to ensure accurate and consistent color reproduction under different lighting conditions. The processed image is passed to the defect recognition module.

[0038] Preferably, the defect identification module is further used to execute the steps:

[0039] Receiving image data processed by the color balance module;

[0040] Preprocess the received image data, including noise reduction and edge enhancement. The noise level N satisfies: N=Σ|I(x,y)-I_avg| / (M*N), where I(x,y) is the pixel brightness value, I_avg is the average brightness value of the image, and M and N are the width and height of the image respectively;

[0041] Using the preprocessed image data, the normal area and the potential defect area are distinguished by setting a threshold. The threshold T satisfies: T=μ+kσ, where μ is the mean brightness of the local area of ​​the image, σ is the standard deviation, and k is a constant;

[0042] The identified potential defect areas are compared with the known defect pattern library, the defect types are located and classified, and the results are passed to the report generation module.

[0043] Technical effects and advantages of the present invention: Compared with the prior art, the high-pressure compressor impeller visual inspection system proposed by the present invention has the following advantages:

[0044] The present invention realizes dynamic lighting adjustment, high-quality image capture and accurate color reproduction by introducing a reflection control module, a contrast enhancement module, an exposure correction module and a color balance module. The reflection control module adjusts the lighting angle and intensity in real time according to the material characteristics of the impeller to ensure that the light evenly covers the entire impeller surface and reduces the interference caused by high reflectivity; the contrast enhancement module and the exposure correction module work together to ensure that the image has a high dynamic range and an appropriate exposure level under different lighting conditions to avoid overexposure and underexposure; the color balance module ensures accurate color reproduction and further improves image quality. These optimization processes enable the defect recognition module to more accurately locate and classify potential defects on the impeller surface, thereby significantly improving the overall performance and reliability of the visual inspection system. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a block diagram of the high-pressure compressor impeller visual inspection system of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0047] The present invention provides Figure 1 A high-pressure compressor impeller visual inspection system shown in the figure realizes dynamic lighting adjustment, high-quality image capture and accurate color reproduction by introducing a reflection control module, a contrast enhancement module, an exposure correction module and a color balance module, as follows:

[0048] Exemplarily, the light source module is used to provide lighting; specifically includes:

[0049] Receive initial lighting parameters including light intensity I_initial and angle A_initial; this step ensures that the system can obtain the basic parameters of the current light source and provide basic data for subsequent lighting adjustments. By accurately measuring the initial lighting conditions, the system can be optimized in a targeted manner.

[0050] Based on the received initial illumination parameters, the ideal illumination conditions adapted to the material characteristics of the impeller are calculated, where the light intensity adjustment coefficient K satisfies: K=(I_target / I_initial), where I_target is the target light intensity (the ideal light intensity determined according to the material characteristics of the impeller). By calculating the light intensity adjustment coefficient K, the system can adjust the output intensity of the light source according to the target light intensity I_target, thereby ensuring that no overexposure or underexposure occurs during image capture.

[0051] Using the calculated ideal lighting conditions, the position and angle of the optical elements in the reflection control module are adjusted so that the light can illuminate the impeller surface at the optimal angle; the optimal angle A_optimal satisfies: A_optimal=A_initial+ΔA, ΔA is the angle adjustment amount; by adjusting the position and angle of the optical elements, the system can ensure that the light evenly covers the entire impeller surface, reduce the interference caused by high reflectivity, and improve image quality.

[0052] The image data captured under the adjusted lighting conditions are passed to the image capture module. After the lighting conditions are optimized, the image capture module can obtain high-quality images of the impeller surface details, providing a reliable data basis for subsequent processing.

[0053] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0054] Receive initial lighting parameters:

[0055] The measured initial light intensity I_initial = 500 lux, and the initial light angle A_initial = 30 degrees.

[0056] Calculate ideal lighting conditions:

[0057] According to the material properties of the impeller, the target light intensity I_target=800lux is determined.

[0058] Calculate the light intensity adjustment factor K:

[0059] K=I_target / I_initial=800 / 500=1.6;

[0060] This means that the output intensity of the light source needs to be increased by 1.6 times.

[0061] Adjust the position and angle of optical components:

[0062] Based on the material properties, the angle adjustment ΔA is calculated to be 10 degrees.

[0063] Calculate the optimal lighting angle A_optimal:

[0064] A_optimal=A_initial+ΔA=30+10=40 degrees;

[0065] Adjust the optics in the Reflection Control Module so that it illuminates the impeller surface at a 40 degree angle.

[0066] Passing image data:

[0067] Under the adjusted lighting conditions, an image capture module is used to capture detailed images of the impeller surface and pass them to the subsequent processing module.

[0068] Exemplarily, the light source module is connected to a reflection control module for adjusting the illumination angle and intensity to adapt to the material characteristics of the impeller; specifically comprising:

[0069] Receive the light signal under the optimized lighting conditions from the light source module, including the adjusted light intensity I_adjusted and the optimal angle A_optimal, to provide basic data for subsequent dynamic adjustment. This step ensures that the system can accurately adjust the lighting according to the impeller material characteristics.

[0070] Based on the received light signal, the lighting angle and intensity are monitored and adjusted in real time; by monitoring the current lighting conditions in real time and making dynamic adjustments as needed, the light is ensured to cover the entire impeller surface and be evenly distributed. This step improves the quality of image capture and reduces image problems caused by uneven lighting.

[0071] Using real-time monitoring data, the position and angle of the optical elements in the reflection control module are dynamically adjusted. The new angle adjustment value ΔA_new satisfies: ΔA_new=A_optimal-A_current, where A_current is the current actual angle. By dynamically adjusting the position and angle of the optical elements, the system can ensure that the light is irradiated to the impeller surface at the optimal angle, reduce the interference caused by high reflectivity, and thus improve image quality.

[0072] The image data captured under the adjusted lighting conditions is passed to the image capture module, and the working state of the reflection control module is continuously optimized through the real-time monitoring feedback loop. After the lighting conditions are optimized, the image capture module can obtain high-quality images of the impeller surface details. At the same time, through the real-time monitoring feedback loop, the system can continuously optimize the working state of the reflection control module to ensure that the image quality is always at the best level.

[0073] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0074] Receive optimized lighting conditions:

[0075] The optimized light intensity of the light source module is I_adjusted = 800 lux (initial light intensity 500 lux, adjustment coefficient 1.6).

[0076] The calculated optimal illumination angle A_optimal = 40 degrees (initial angle 30 degrees, angle adjustment amount 10 degrees).

[0077] Monitor and adjust lighting angle and intensity in real time:

[0078] Monitor current lighting conditions in real time and obtain the current actual light intensity and angle.

[0079] Assume that the current actual light intensity is 780 lux and the current actual angle is 38 degrees.

[0080] Dynamically adjust the position and angle of optical elements in the Reflection Control Module:

[0081] Calculate the new angle adjustment ΔA_new:

[0082] ΔA_new=A_optimal-A_current=40-38=2 degrees;

[0083] According to the calculation results, the optical elements in the reflection control module are adjusted to adjust the lighting angle from 38 degrees to 40 degrees to ensure that the light illuminates the impeller surface at the optimal angle.

[0084] Pass image data and continue to optimize:

[0085] Under the adjusted lighting conditions, an image capture module is used to capture detailed images of the impeller surface and pass them to the subsequent processing module.

[0086] Through a real-time monitoring feedback loop, the system continuously monitors lighting conditions and further adjusts the position and angle of optical components as needed, ensuring that image quality always remains optimal.

[0087] Exemplarily, the reflection control module is connected to the image capture module, and is responsible for capturing the detailed image of the impeller surface; specifically, it includes:

[0088] Receive the light signal from the reflection control module under the lighting conditions optimized by the reflection control module; this step ensures that the image capture module can obtain the lighting parameters optimized by the reflection control module, providing accurate basic data for subsequent camera parameter adjustment. This step ensures that the system can capture images under the best lighting conditions.

[0089] Based on the received light signal, adjust the camera parameters in the image capture module, including the exposure time T and the gain value G. The exposure time T satisfies: T=T_base*(I_target / I_received), G=G_base*(I_target / I_received), where T_base and G_base are initial values, I_target is the target light intensity, and I_received is the actual light intensity received. By adjusting the exposure time and gain value of the camera, the system can optimize the quality of image capture according to the target light intensity, avoid overexposure or underexposure, and ensure that the image details are clearly visible.

[0090] Using the adjusted camera parameters, the image capture module is started to capture the detailed image of the impeller surface, providing a reliable data basis for subsequent processing. This step ensures the accuracy and reliability of image capture.

[0091] The captured image data is passed to the contrast enhancement module, and the working status of the image capture module is monitored and adjusted in real time through the internal calibration mechanism.

[0092] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0093] Receive optimized lighting conditions:

[0094] The light intensity after optimization by the reflection control module is I_received=780lux (target light intensity I_target=800lux).

[0095] Adjust camera parameters:

[0096] The initial exposure time T_base=10ms, and the initial gain value G_base=1.5.

[0097] Calculate the adjusted exposure time T:

[0098] T=T_base*(I_target / I_received)=10*(800 / 780)≈10.26ms;

[0099] Calculate the adjusted gain value G:

[0100] G=G_base*(I_target / I_received)=1.5*(800 / 780)≈1.54;

[0101] Start the image capture module to capture images:

[0102] Using the adjusted exposure time of 10.26 ms and gain value of 1.54, the image capture module was started to capture detailed images of the impeller surface.

[0103] Pass image data and continuously monitor:

[0104] The captured image data is passed to the contrast enhancement module.

[0105] Monitor the working status of the image capture module in real time and make adjustments as needed to ensure that image quality is always maintained at the best level.

[0106] Exemplarily, the output end of the image capture module is connected to a contrast enhancement module for adjusting the dynamic range according to the captured image data; specifically comprising:

[0107] Receive image data from the image capture module; this step ensures that the contrast enhancement module can obtain high-quality impeller surface detail images, providing accurate basic data for subsequent contrast adjustment. This step ensures that the system can perform image processing under optimal lighting conditions.

[0108] Based on the received image data, the brightness value distribution of each pixel in the image is calculated to determine the minimum brightness L_min and the maximum brightness L_max for subsequent dynamic range adjustment; this step helps to improve the dynamic range of the image and make the details in the image more clearly visible.

[0109] Using the calculated brightness value distribution, the global contrast enhancement method is applied to adjust the overall contrast of the image. The new brightness value L_new satisfies:

[0110] L_new=(L-L_min)*(L_max_new-L_min_new) / (L_max-L_min)+L_min_new, where L is the original brightness value, L_max_new and L_min_new are the maximum and minimum values ​​of the target brightness; by applying the global contrast enhancement method, the system can adjust the overall contrast of the image according to the target brightness range, make the details in the image clearer, and improve the overall quality of the image.

[0111] The processed images are passed to the exposure correction module. After contrast enhancement, the image quality is significantly improved and the details are clearer. These high-quality images are passed to the exposure correction module to further optimize the image exposure parameters to ensure that the image can maintain good visual effects under different lighting conditions.

[0112] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0113] Receiving image data:

[0114] Detailed image data of the impeller surface is received from the image capture module.

[0115] Calculate the brightness value distribution:

[0116] Calculate the brightness value distribution of each pixel in the image.

[0117] Assume that the calculated minimum brightness L_min=50 and the maximum brightness L_max=200.

[0118] Apply a global contrast enhancement method:

[0119] The maximum value L_max_new of the target brightness is set to 255 and the minimum value L_min_new is set to 0.

[0120] Adjust the brightness value L of each pixel using the formula:

[0121] L_new=(L-L_min)*(L_max_new-L_min_new) / (L_max-L_min)+L_min_new;

[0122] For example, for a pixel with brightness value L=100:

[0123] L_new=(100-50)*(255-0) / (200-50)+0

[0124] =50*255 / 150

[0125] =85;

[0126] Pass the processed image:

[0127] The image after contrast enhancement processing is passed to the exposure correction module to further optimize the exposure parameters of the image to ensure that the image can maintain good visual effects under different lighting conditions.

[0128] Exemplarily, the contrast enhancement module signal is transmitted to the exposure correction module to optimize the exposure parameters to avoid overexposure; specifically including:

[0129] Receive the image data processed by the contrast enhancement module to provide accurate basic data for subsequent exposure parameter adjustment. This step ensures that the system can be further optimized under the best lighting conditions.

[0130] Based on the received image data, calculate the actual brightness value of each pixel in the image, and determine the overexposed area and underexposed area, with the overexposure threshold being LO_over and the underexposure threshold being LO_under; this step helps to identify the areas that need adjustment, thus avoiding overexposure or underexposure phenomena.

[0131] Using the determined overexposure and underexposure area information, adjust the exposure parameters in the exposure correction module. The new exposure time T_new satisfies: T_new = T_old * (L_target / L_current), where T_old is the original exposure time, L_target is the target brightness value, and L_current is the current actual brightness value; by adjusting the exposure time, the system can optimize the overall exposure effect of the image according to the target brightness value, avoiding overexposure or underexposure phenomena, and thus ensuring the image quality.

[0132] Apply the adjusted exposure parameters to the subsequent image capture process, continuously monitor the image brightness distribution, and ensure that the image can maintain a good visual effect under different lighting conditions.

[0133] Suppose there is a high-pressure compressor impeller whose material has a high reflectivity to a specific spectrum. To ensure high-quality visual inspection, the following steps are required:

[0134] Receive the processed image data:

[0135] Receive the processed impeller surface detail image data from the contrast enhancement module.

[0136] Calculate the brightness value and determine the overexposed and underexposed areas:

[0137] Calculate the actual brightness value of each pixel in the image.

[0138] Set the overexposure threshold LO_over = 240 and the underexposure threshold LO_under = 10.

[0139] For a pixel with an actual brightness value L = 250, since L > LO_over, this pixel is in the overexposed area.

[0140] For another pixel with an actual brightness value L = 5, since L < LO_under, this pixel is in the underexposed area.

[0141] Adjust the exposure parameters:

[0142] Suppose the original exposure time T_old = 10ms, the target brightness value L_target = 180, and the current actual brightness value L_current = 200.

[0143] Calculate the new exposure time T_new:

[0144] T_new=T_old*(L_target / L_current)=10*(180 / 200)=9ms;

[0145] Apply adjusted exposure parameters and continuously monitor:

[0146] The adjusted exposure time of 9ms is applied to subsequent image capture processes.

[0147] By real-time monitoring of image brightness distribution, it ensures that the image maintains good visual effects under different lighting conditions.

[0148] Exemplarily, the exposure correction module is connected to the color balance module to ensure accurate color reproduction under different light conditions; specifically, it includes:

[0149] Receive the processed image data from the exposure correction module to provide accurate basic data for subsequent color channel analysis and gain adjustment. This step ensures that the system can optimize color reproduction under the best lighting conditions.

[0150] Based on the received image data, the brightness distribution of the red, green, and blue color channels in the image is analyzed to determine the average brightness values ​​R_avg, G_avg, and B_avg of each channel; this step helps to identify color imbalance problems and provide basic data for subsequent color correction.

[0151] Using the average brightness value of each channel, adjust the gain coefficient of each channel. The new gain coefficients are Gain_R, Gain_G and Gain_B, which satisfy:

[0152] Gain_R=(R_target / R_avg);

[0153] Gain_G = (G_target / G_avg);

[0154] Gain_B=(B_target / B_avg);

[0155] Among them, R_target, G_target and B_target are the target brightness values; by adjusting the gain coefficient of each color channel, the system can optimize the color reproduction effect of the image according to the target brightness value, ensuring accurate and consistent color reproduction under different lighting conditions.

[0156] The adjusted gain factor is applied to the color channel of the image, and the color reproduction effect is continuously optimized through real-time monitoring to ensure accurate and consistent color reproduction under different lighting conditions. At the same time, the processed image is passed to the defect recognition module to further improve the accuracy of detection.

[0157] Among them, R_avg: average brightness value of the red channel, G_avg: average brightness value of the green channel, B_avg: average brightness value of the blue channel, Gain_R: red channel gain coefficient, Gain_G: green channel gain coefficient, Gain_B: blue channel gain coefficient, R_target: red channel target brightness value, G_target: green channel target brightness value, B_target: blue channel target brightness value.

[0158] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0159] Receive processed image data:

[0160] The processed impeller surface detail image data is received from the exposure correction module.

[0161] Analyze color channel brightness distribution:

[0162] Analyzes the brightness distribution of each color channel (red, green, blue) in an image.

[0163] Assume that the calculated average brightness value of the red channel R_avg=100, the average brightness value of the green channel G_avg=150, and the average brightness value of the blue channel B_avg=80.

[0164] Adjust the gain factor of each channel:

[0165] Set the target brightness values ​​R_target=120, G_target=120, B_target=120.

[0166] Calculate the new gain factor:

[0167] Gain_R=R_target / R_avg=120 / 100=1.2;

[0168] Gain_G=G_target / G_avg=120 / 150=0.8;

[0169] Gain_B=B_target / B_avg=120 / 80=1.5;

[0170] Apply the adjusted gain factor and continuously monitor:

[0171] Apply the adjusted gain factor to the color channel of the image to make the colors in the image more balanced and realistic. Continuously optimize the color reproduction effect through real-time monitoring to ensure accurate and consistent color reproduction under different lighting conditions. Pass the processed image to the defect recognition module to further improve the accuracy of detection.

[0172] Exemplarily, the color balancing module is connected to the defect recognition module to locate potential defects on the impeller surface; specifically, the color balancing module includes:

[0173] Receive image data processed by the color balance module to provide accurate basic data for subsequent preprocessing and defect detection. This step ensures that the system can identify defects under optimal lighting conditions.

[0174] The received image data is preprocessed, including noise reduction and edge enhancement. The noise level N satisfies: N=Σ|I(x,y)-I_avg| / (M*N), where I(x,y) is the pixel brightness value, I_avg is the average brightness value of the image, and M and N are the width and height of the image respectively. Through noise reduction and edge enhancement processing, the system can reduce noise interference in the image and highlight the edge features of the potential defect area. This step helps to improve the accuracy of subsequent defect detection.

[0175] Using the preprocessed image data, the normal area and the potential defect area are distinguished by setting a threshold. The threshold T satisfies: T=μ+kσ, where μ is the mean brightness of the local area of ​​the image, σ is the standard deviation, and k is a constant. By setting the threshold, the system can distinguish between normal areas and potential defect areas, thereby accurately locating and classifying defect types. This step improves the accuracy and reliability of defect detection.

[0176] The identified potential defect areas are compared with the known defect pattern library to locate and classify the defect types, and the results are passed to the report generation module. This step improves the accuracy and efficiency of defect identification, and the results are passed to the report generation module to generate a detailed inspection report.

[0177] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0178] Receive processed image data:

[0179] The processed impeller surface detail image data is received from the color balancing module.

[0180] Preprocessing:

[0181] The received image data is subjected to noise reduction and edge enhancement processing.

[0182] Calculate the noise level N of the image:

[0183] N=Σ|I(x,y)-I_avg| / (M*N);

[0184] Assume that the image size is 1000x1000 pixels, the image average brightness value I_avg=128, and the calculated noise level N=5.

[0185] Set thresholds and differentiate between normal and defective areas:

[0186] Set the local area brightness mean μ=130, standard deviation σ=10, and constant k=2.

[0187] Calculate the threshold T:

[0188] T = μ + kσ = 130 + 2 * 10 = 150;

[0189] The normal area and the potential defect area are distinguished by using the set threshold T=150. For example, for a pixel with a brightness value of L=160, since L>T, the pixel is located in the potential defect area.

[0190] Compare the defect pattern library and pass the results:

[0191] The identified potential defect areas are compared with the known defect pattern library to locate and classify the defect type. Assuming that the detected defect type is "crack", the results are passed to the report generation module to generate a detailed inspection report.

[0192] Exemplarily, the defect identification module information is transmitted to the report generation module for generating a detailed inspection report; specifically including:

[0193] Receive defect information and classification results from the defect recognition module to provide accurate basic data for subsequent report generation. This step ensures that the system can generate detailed inspection reports under optimal lighting conditions.

[0194] Parse and organize the location X, Y, type Type, and severity of each defect detected; by parsing and organizing defect information, the system can clearly record the specific location, type, and severity of each defect. This step helps to generate a structured inspection report to ensure the accuracy and completeness of the information.

[0195] Using the sorted defect information, a structured inspection report template is generated, which includes the basic information of the impeller, the inspection date Date, and a detailed defect list. The report template format is: Report = {Basic information: {Model: Model, Batch number: Batch}, Inspection date: Date, Defect list: [{Location: (X, Y), Type: Type, Severity: Severity}]}; By generating a structured inspection report template, the system can integrate the basic information of the impeller, the inspection date, and the detailed defect list to form a complete inspection report. This step improves the readability and practicality of the report.

[0196] The generated structured report is converted into a user-friendly visual format and presented to the user through the user interface module for easy viewing and analysis. This step improves the user experience and ensures the ease of use of the report.

[0197] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0198] Receive defect information and classification results:

[0199] Receive detected defect information and classification results from the defect recognition module.

[0200] Assume that the detected defect information is as follows: [

[0202] {"Location":(100,200),"Type":"Crack","Severity":"Medium"},

[0203] {"Location":(300,400),"Type":"Pit","Severity":"Minor"} ];

[0205] Parse and organize defect information:

[0206] Analyze the location, type, and severity of each detected defect:

[0207] Defect 1: Location (100, 200), Type "crack", Severity "medium";

[0208] Defect 2: Location (300, 400), Type "Pit", Severity "Minor";

[0209] Generate a structured test report template:

[0210] Integrate the basic information of the impeller (such as model, batch number), inspection date and detailed defect list to generate a structured inspection report template:

[0211] Report={

[0212] Basic information: {model: "XYZ-123", batch number: "BATCH-001"},

[0213] Test date:"2025-01-01",

[0214] Defect list:

[0215] {"Location":(100,200),"Type":"Crack","Severity":"Medium"},

[0216] {"Location":(300,400),"Type":"Pit","Severity":"Minor"} ]

[0218] };

[0219] Convert to a user-friendly visualization format and display:

[0220] The generated structured report is converted into a user-friendly visual format, such as a table or chart, and is presented to the user through a user interface module.

[0221] Exemplarily, the report generation module is connected to the user interface module to display the test results and allow users to interactively query; specifically, it includes:

[0222] Receive structured test report data from the report generation module to provide accurate basic data for subsequent parsing and formatting. This step ensures that the system can generate detailed test reports under optimal lighting conditions and display them to users.

[0223] Parse and format the test results, converting the defect location X, Y, type Type and severity Severity information into a user-readable form; this step improves the user experience and enables users to quickly understand and view the test results.

[0224] An interactive view is created on the user interface, allowing the user to view detailed defect information by clicking or selecting different areas. The display element Display_element in the user interface satisfies: Display_element={position: (X, Y), content: Defect_info}; this interactive method improves the user experience and enables users to understand the inspection results more intuitively.

[0225] Providing query function, the system allows users to filter and view relevant defect information according to specific conditions (such as defect type, severity, etc.). This step improves the flexibility and practicality of the system, allowing users to quickly find the defect information of interest.

[0226] Consider a high-pressure compressor impeller whose material has high reflectivity for a specific spectrum of light. To ensure high-quality visual inspection, the following steps are required:

[0227] Receive structured test report data:

[0228] Receive detailed inspection report data from the report generation module. Assume that the detected defect information is as follows:

[0229] Report={

[0230] Basic information: {model: "XYZ-123", batch number: "BATCH-001"},

[0231] Test date:"2025-01-01",

[0232] Defect list:

[0233] {"Location":(100,200),"Type":"Crack","Severity":"Medium"},

[0234] {"Location":(300,400),"Type":"Pit","Severity":"Minor"} ]

[0236] };

[0237] Parse and format the test results:

[0238] Convert the detection results into a user-friendly format:

[0239] Impeller model: XYZ-123;

[0240] Batch number: BATCH-001;

[0241] Test date: 2025-01-01;

[0242] Defect list:

[0243] Position (X, Y) type Severity (100,200) crack medium (300,400) Pits slight

[0244] To create an interactive view:

[0245] Create an interactive view on the user interface, allowing users to view detailed defect information by clicking or selecting different areas. For example, define the display element Display_element:

[0246] Display_element={

[0247] Position:(100,200),

[0248] content:{

[0249] Type: "Crack",

[0250] Severity: "Medium"

[0251] }

[0252] };

[0253] Users can view detailed defect information by clicking or selecting different areas (such as location (100,200)):

[0254] Position: (100,200);

[0255] Type: Crack;

[0256] Severity: Moderate;

[0257] Provide query function:

[0258] Provides a query function that allows users to enter conditions to filter and view relevant defect information. For example, a user can select a defect of the "crack" type, and the system will return all defect information that meets this condition:

[0259] Input conditions: type = "crack";

[0260] Query results:

[0261] Position (X, Y) type Severity (100,200) crack medium

[0262] This series of operations significantly improves the overall performance and reliability of the visual inspection system, ensuring high-quality image capture, accurate defect identification, and detailed inspection report generation and presentation.

[0263] In summary, by introducing the reflection control module, contrast enhancement module, exposure correction module and color balance module, dynamic lighting adjustment, high-quality image capture and accurate color reproduction are achieved. The reflection control module adjusts the lighting angle and intensity in real time according to the material characteristics of the impeller to ensure that the light evenly covers the entire impeller surface and reduce the interference caused by high reflectivity; the contrast enhancement module and the exposure correction module work together to ensure that the image has a high dynamic range and appropriate exposure level under different lighting conditions to avoid overexposure and underexposure; the color balance module ensures accurate color reproduction and further improves image quality. These optimization processes enable the defect recognition module to more accurately locate and classify potential defects on the impeller surface, thereby significantly improving the overall performance and reliability of the visual inspection system.

[0264] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A high-pressure compressor impeller visual inspection system, characterized in that: include: A light source module, used for providing lighting; A reflection control module connected to the light source module, for adjusting the illumination angle and intensity to suit the impeller material characteristics; The reflection control module is connected to the image capture module and is responsible for capturing the detailed image of the impeller surface; The output end of the image capture module is connected to a contrast enhancement module, which is used to adjust the dynamic range according to the captured image data; the contrast enhancement module is also used to perform the steps of: receiving image data from the image capture module; calculating the brightness value distribution of each pixel in the image based on the received image data, and determining the minimum brightness L_min and the maximum brightness L_max for subsequent dynamic range adjustment; using the calculated brightness value distribution, applying a global contrast enhancement method, adjusting the overall contrast of the image, and the new brightness value L_new satisfies: L_new=(L-L_min)*(L_max_new-L_min_new) / (L_max-L_min)+L_min_new, where L is the original brightness value, L_max_new and L_min_new are the maximum and minimum values ​​of the target brightness; the processed image is passed to the exposure correction module; The contrast enhancement module signal is transmitted to the exposure correction module to optimize the exposure parameters to avoid overexposure; the exposure correction module is also used to perform the steps of: receiving the image data processed by the contrast enhancement module; calculating the actual brightness value of each pixel in the image based on the received image data, and determining the overexposed area and the underexposed area, the overexposure threshold is LO_over and the underexposure threshold is LO_under; using the determined overexposed and underexposed area information, adjusting the exposure parameters in the exposure correction module, the new exposure time T_new satisfies: T_new=T_old*(L_target / L_current), wherein T_old is the original exposure time, L_target is the target brightness value, and L_current is the current actual brightness value; applying the adjusted exposure parameters to the subsequent image capture process, and continuously monitoring the image brightness distribution; The exposure correction module is connected to the color balance module to ensure accurate color reproduction under different light conditions; The color balancing module is connected to the defect recognition module to locate potential defects on the impeller surface.

2. A high pressure compressor impeller visual inspection system according to claim 1, characterized in that: The light source module is also used to perform the steps: Receive initial lighting parameters including light intensity I_initial and angle A_initial; Based on the received initial illumination parameters, the ideal illumination conditions adapted to the material characteristics of the impeller are calculated, where the light intensity adjustment coefficient K satisfies: K=(I_target / I_initial), where I_target is the target light intensity; Using the calculated ideal lighting conditions, adjust the position and angle of the optical elements in the reflection control module so that the light can irradiate the impeller surface at the optimal angle; the optimal angle A_optimal satisfies: A_optimal=A_initial+ΔA, ΔA is the angle adjustment amount; The image data captured under the adjusted lighting conditions is passed to the image capture module.

3. A high pressure compressor impeller visual inspection system according to claim 2, characterized in that: The reflection control module is also used to execute the steps: Receiving a light signal under optimized lighting conditions from a light source module, including an adjusted light intensity I_adjusted and an optimal angle A_optimal; Based on the received light signal, the lighting angle and intensity are monitored and adjusted in real time; Using the real-time monitoring data, the position and angle of the optical elements in the reflection control module are dynamically adjusted. The new angle adjustment ΔA_new satisfies: ΔA_new=A_optimal-A_current, where A_current is the current actual angle. The image data captured under the adjusted lighting conditions is passed to the image capture module, and the working status of the reflection control module is continuously optimized through a real-time monitoring feedback loop.

4. A high-pressure compressor impeller visual inspection system according to claim 3, characterized in that: The image capture module is also used to perform the steps: receiving a light signal under the lighting conditions optimized by the reflection control module; Based on the received light signal, adjust the camera parameters in the image capture module, including the exposure time T and the gain value G. The exposure time T satisfies: T=T_base*(I_target / I_received), G=G_base*(I_target / I_received), where T_base and G_base are initial values, I_target is the target light intensity, and I_received is the actual light intensity received; Using the adjusted camera parameters, the image capture module is started to capture the detailed image of the impeller surface; The captured image data is passed to the contrast enhancement module, and the working status of the image capture module is monitored and adjusted in real time through the internal calibration mechanism.

5. A high pressure compressor impeller visual inspection system according to claim 4, characterized in that: The color balance module is also used to perform the steps: receiving image data processed by the exposure correction module; Based on the received image data, analyze the brightness distribution of the red, green and blue color channels in the image, and determine the average brightness values ​​R_avg, G_avg and B_avg of each channel; Using the average brightness value of each channel, adjust the gain coefficient of each channel. The new gain coefficients are Gain_R, Gain_G and Gain_B, which satisfy: Gain_R = (R_target / R_avg), Gain_G = (G_target / G_avg), Gain_B = (B_target / B_avg), where R_target, G_target and B_target are the target brightness values; The adjusted gain factor is applied to the color channel of the image and the color reproduction is continuously optimized through real-time monitoring to ensure accurate and consistent color reproduction under different lighting conditions. The processed image is passed to the defect recognition module.

6. A high-pressure compressor impeller visual inspection system according to claim 5, characterized in that: The defect identification module is also used to perform the steps: Receiving image data processed by the color balance module; Preprocess the received image data, including noise reduction and edge enhancement. The noise level N satisfies: N=Σ|I(x,y)-I_avg| / (M*N), where I(x,y) is the pixel brightness value, I_avg is the average brightness value of the image, and M and N are the width and height of the image respectively; Using the preprocessed image data, the normal area and the potential defect area are distinguished by setting a threshold. The threshold T satisfies: T=μ+kσ, where μ is the mean brightness of the local area of ​​the image, σ is the standard deviation, and k is a constant; The identified potential defect areas are compared with the known defect pattern library, the defect types are located and classified, and the results are passed to the report generation module.

7. A high pressure compressor impeller visual inspection system according to claim 6, characterized in that: The defect identification module information is transmitted to the report generation module for generating a detection report.

8. A high pressure compressor impeller visual inspection system according to claim 7, characterized in that: The report generation module is connected to the user interface module to display the detection results and allow users to interactively query.

Images