Method and System for Defect Detection of Contact Lenses Based on Image Recognition

Abstract

The present invention discloses a method and system for detecting defects of contact lenses based on image recognition, including performing material type and surface grid division on the contact lenses to be tested, calling a trained neural network model for defect detection, obtaining the output result of defect detection, analyzing and processing the acoustic characteristic data and the expected detection depth to generate a frequency selection coefficient, calibrating an ultrasonic probe to obtain reference benchmark data, emitting ultrasonic waves in a frequency band selected by the frequency selection coefficient, obtaining its detection intensity value and detection time delay value, analyzing and processing the reference intensity value and the detection intensity value to obtain a first comparison coefficient, analyzing and processing the reference time delay value and the detection time delay value to obtain a second comparison coefficient, and combining the analysis to obtain an output result fine-tuning index. The present invention can significantly improve the accuracy and reliability of contact lens defect detection while ensuring the detection efficiency, and provides an efficient and accurate automated detection solution for the production quality control of contact lenses.

Description

Technical Field

[0001] The present invention relates to the field of communication circuits, and more specifically, to a method and system for detecting defects in contact lenses based on image recognition. Background Art

[0002] As one of the important tools for modern vision correction, with the progress of manufacturing technology, contact lenses have become the first choice for many patients with vision problems such as myopia, hyperopia, and astigmatism. They not only provide a more natural appearance but also give wearers higher convenience and comfort. However, various defects are inevitable in the manufacturing process of contact lenses, including but not limited to scratches, bubbles, foreign object attachments, optical zone distortions, uneven thickness, etc. on the lens surface. If these defects are not detected and properly processed, they will directly threaten the eye health of users. Therefore, accurate and efficient detection of contact lens defects is particularly important. Traditional manual detection methods are limited by factors such as the resolution, fatigue, and subjective judgment of the human eye, and it is difficult to achieve ideal detection accuracy and efficiency. For this reason, detection technologies based on image recognition have emerged. With the rise of deep learning technology, it has become possible to use pre-trained neural network models for image recognition and defect detection. This method can greatly improve the automation degree and accuracy of detection;

[0003] In the prior art, the publication number is CN115239663A, and the name is a method for detecting contact lens defects. The method includes: obtaining an image of a defect-free contact lens; preprocessing the image of the defect-free contact lens; identifying the pattern area in the image of the defect-free contact lens, where the pattern area is the area where defects occur in the contact lens image; generating a plurality of random defect blocks; repeating the operation of randomly placing N random defect blocks into the pattern area of the image of the defect-free contact lens multiple times to train a neural network model for detecting contact lens defects; detecting defects in the contact lens to be tested;

[0004] The existing technologies often lack targeted detection strategies when dealing with contact lenses of different material types, resulting in the accuracy and reliability of detection results being affected. In addition, the existing technologies also have certain limitations in the precise measurement of defect depth and size, mainly due to the lack of effective calibration and fine-tuning mechanisms; the above information disclosed in the background art section is only used to strengthen the understanding of the background of the present disclosure, so it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The present invention aims to overcome at least one defect in the prior art and provides a method and system for detecting contact lens defects based on image recognition to solve the problems raised in the above background art.

[0006] The technical solution adopted by the present invention is a method for detecting defects of contact lenses based on image recognition, and the specific steps include:

[0007] Step S1: Classify the material types of the contact lenses to be tested and divide the surface into grids, where:

[0008] The material type classification is to classify the contact lenses to be tested according to material types to form a material type set Among them, represents the contact lens of the th material type, represents the total number of material types;

[0009] The surface grid division is to divide the surface of the contact lens into multiple grid regions to form a grid region set , where, represents the number of the rd grid region, represents the total number of grid divisions;

[0010] Step S2: Call the pre-trained neural network model to detect defects of the contact lenses to be tested;

[0011] Step S3: Collect the image data of the contact lenses to be tested, input the image data into the pre-trained neural network model for processing, and finally obtain the output result of defect detection. The output result includes the defect depth value and defect size value of the contact lens of the th material type in the th grid region;

[0012] Step S4: Collect the acoustic characteristic data of the contact lens of the th material type to be tested in the th grid region, and the expected detection depth. The acoustic characteristic data includes sound velocity and acoustic impedance, and the expected detection depth is the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects;

[0013] Step S5: Obtain the acoustic characteristic data and the expected detection depth, perform analysis and processing to generate a frequency selection coefficient, which is used to classify and select the frequency band of the ultrasonic probe in the frequency range of 5 MHz to 15 MHz;

[0014] Step S6: Calibrate the selected ultrasonic probe using a defect-free contact lens sample, and use the calibrated ultrasonic probe to scan the defect-free contact lens sample again to obtain reference benchmark data. The reference benchmark data includes the reference intensity value and reference time delay value of the contact lens of the th material type in the th grid region;

[0015] Step S7: Using the calibrated ultrasonic probe, emit ultrasonic waves in the selected frequency band, making the ultrasonic waves penetrate the tested contact lens of the th material type, and focusing the ultrasonic waves on the th grid area. Subsequently, receive the ultrasonic signal reflected from the surface of this area, and analyze the ultrasonic signal reflected from the th grid area to obtain the detection intensity value and detection time delay value of the

[0016] th grid area;

[0017] Step S8: Obtain the reference intensity value and the detection intensity value, and perform analysis and processing to obtain the first comparison coefficient. At the same time, obtain the reference time delay value and the detection time delay value, and perform analysis and processing to obtain the second comparison coefficient. Subsequently, the detection system combines and analyzes the first comparison coefficient and the second comparison coefficient to obtain the output result fine-tuning index, which is used to provide a fine-tuning strategy for the flaw depth value and flaw size value included in the output result.

[0017] By dividing the contact lenses of different material types and their surfaces, images of different material types and different regions are obtained, making it more targeted; input the collected image data into the model for processing to obtain the flaw detection results, including the area where the flaw is located, the flaw depth value and the flaw size value, obtain the acoustic characteristic data of the area where the flaw is located and the expected detection depth, perform analysis and processing, and generate a frequency selection coefficient for adjusting the frequency of the ultrasonic probe; calibrate the ultrasonic probe to obtain the reference benchmark data of the flawless contact lens; emit the ultrasonic waves with the selected frequency, penetrate the flaw area of the contact lens, and obtain the detection intensity value and the detection time delay value; analyze and process the reference benchmark data, the detection intensity value and the detection time delay value to obtain the first comparison coefficient and the second comparison coefficient, perform combined analysis, and obtain the output result fine-tuning index, which is used to provide a fine-tuning strategy for the flaw depth value and flaw size value included in the output result.

[0018] Furthermore, step S3 includes preprocessing the image of the tested contact lens to make it meet the model input requirements; inputting the preprocessed image into the trained neural network model; the model outputs the predicted flaw information, including the flaw depth value and the flaw size value; mark the flaw depth value and the flaw size value of the contact lens of the th material type in the th grid area as , respectively.

[0019] Furthermore, in step S4, in the th material type of contact lens in the In a grid region, the sound velocity and acoustic impedance included in the acoustic characteristic data are sequentially marked as , , respectively. At the same time, the expected detection depth is marked as , and The calculation formula is as follows:

[0020]

[0021] Among them, is the judgment threshold of the flaw depth value, and it is ensured that .

[0022] Furthermore, the region where the flaw is located is obtained, that is, the acoustic characteristic data and the expected detection depth of the contact lens of the rd material type in the th grid region are analyzed and processed to generate a frequency selection coefficient, and the frequency selection coefficient is defined as , and the calculation formula is as follows:

[0023]

[0024] Among them, a, b, and c are all positive weight coefficients used to adjust the influence degree of each parameter on the frequency selection coefficient;

[0025]

[0026] The ultrasonic frequency selection formula is set as:

[0027]

[0028] Among them, is the ultrasonic frequency selected in the th grid region for the contact lens of the th material type, and the value range of is between 5 MHz and 15 MHz; is the frequency selection coefficient;

[0029] It is set that the value range is (0, 1). When is closer to 0, the selected frequency is closer to 5 MHz; when is closer to 1, the selected frequency is closer to 15 MHz.

[0030] Furthermore, the calibration of the ultrasonic probe in step S6 includes preparing a flawless contact lens sample and performing distance-amplitude calibration , performing sensitivity adjustment and recording the reference intensity value and the reference time delay value;

[0031] The recorded reference intensity value and reference time delay value are specifically obtained by using the adjusted ultrasonic probe to scan the flawless sample again;

[0032] Record the reference intensity value of each grid area and the reference time delay value , The calculation formula is as follows;

[0033]

[0034] Wherein, is the reference intensity value of the th type of material type contact lens in the th grid area; is the amplitude value of the th echo signal; is the total number of echo signals;

[0035] The calculation formula is as follows;

[0036]

[0037] Wherein, is the reference time delay value of the th type of material type contact lens in the th grid area, is the time delay value of the th echo signal.

[0038] Furthermore, step S7 includes ultrasonic transmission, ultrasonic reception, signal analysis, calculation of the detected intensity value and the detected time delay value, and data storage;

[0039] The signal analysis is specifically to perform a fast Fourier transform on the received ultrasonic signal , to extract the frequency components and intensity of the signal. The fast Fourier transform formula is as follows;

[0040]

[0041] Wherein, is the Fourier transform result of the received signal; is the th sampling point of the received signal; is the total number of sampling points; p is the imaginary unit; is the complex exponential function, used to implement the transformation from the time domain to the frequency domain; represents the frequency component.

[0042] Furthermore, the calculation of the detected intensity value and the detected time delay value is specifically based on As a result, the detection intensity value and the detection time delay value of the th grid region are calculated;

[0043]

[0044] Among them, is the detection intensity value of the th type of material - type contact lens in the th grid region;

[0045] is the maximum amplitude of the result;

[0046]

[0047] Among them, is the detection time delay value of the th type of material - type contact lens in the th grid region;

[0048] is the frequency corresponding to the maximum amplitude of the result.

[0049] Furthermore, the reference intensity value and the reference time delay value of the th type of material - type contact lens in the th grid region are sequentially marked as , ;

[0050] The detection intensity value and the detection time delay value of the th type of material - type contact lens in the th grid region are sequentially marked as , ;

[0051] Define the first comparison coefficient as , and the calculation formula is as follows:

[0052]

[0053] is the detection intensity value of the th type of material - type contact lens in the th grid region, is the reference intensity value of the th type of material - type contact lens in the th grid region;

[0054] The value range is (0, 1). When is closer to 0, it indicates that the amplitude of the detection intensity lower than the reference intensity is greater, and the probability of defects is higher; when is closer to 1, it indicates that the detection intensity is closer to the reference intensity, and the probability of defects is lower;

[0055] Define the second comparison coefficient as , and the calculation formula is as follows:

[0056]

[0057] is the detection time delay value of the th type of material type contact lens in the th grid area, is the reference time delay value of the th type of material type contact lens in the th grid area;

[0058] The value range of is (0, 1). When is closer to 0, it indicates that the amplitude of the detection time delay lower than the reference time delay is greater, and the probability of defects is higher; when is closer to 1, it indicates that the detection time delay is closer to the reference time delay, and the probability of defects is lower.

[0059] Furthermore, define the output result fine-tuning index as , and the calculation formula is as follows:

[0060]

[0061] Among them, is the output result fine-tuning index, which is used for the fine-tuning strategy of the th type of material type contact lens in the th grid area; is the maximum value of the first comparison coefficient in all grid areas; is the maximum value of the second comparison coefficient in all grid areas; u1 and u2 are the positive weight coefficients of the first comparison coefficient and the second comparison coefficient respectively, and u1 + u2 = 1, 0.14 ≤ u1 ≤ 0.78, 0.08 ≤ u2 ≤ 0.89;

[0062] The value range of is (0, 1). When is closer to 0, it indicates that the probability of defects is higher; when is closer to 1, it indicates that the probability of defects is lower;

[0063] Set the output result fine-tuning index The comparison threshold is Q1;

[0064] 1) When the output result fine-tuning exponent is greater than the threshold Q1, the fine-tuning strategy is not to perform fine-tuning;

[0065] Defect depth value and defect size value remain unchanged;

[0066] 2) When the output result fine-tuning exponent is equal to the threshold Q1, the fine-tuning strategy is to slightly adjust the defect depth value and defect size value ;

[0067] 3) When the output result fine-tuning exponent is less than the threshold Q1, the fine-tuning strategy includes:

[0068] 3.1) When the first comparison coefficient is less than 0.3 and the second comparison coefficient is greater than or equal to 0.8:

[0069] If u1≥u2, the fine-tuning strategy is a medium adjustment strategy mainly based on ;

[0070] If u1<u2, the fine-tuning strategy is a slight adjustment strategy mainly based on ;

[0071] 3.2) When the first comparison coefficient is greater than or equal to 0.8 and the second comparison coefficient is less than 0.3:

[0072] If u1≥u2, the fine-tuning strategy is a slight adjustment strategy mainly based on ;

[0073] If u1<u2, the fine-tuning strategy is a medium adjustment strategy mainly based on ;

[0074] Furthermore, the present invention provides an image recognition-based contact lens defect detection system for performing the above-mentioned image recognition-based contact lens defect detection method, including:

[0075] Partition module: used for partitioning the material type and surface grid of the contact lens to be measured;

[0076] Neural network model calling module: used for calling a pre-trained neural network model to detect defects in the contact lens to be measured;

[0077] Output result generation module: It is used to collect the image data of the contact lens to be tested, input the image data into a pre-trained neural network model for processing, and finally obtain the output result of defect detection. The output result includes the defect depth value and defect size value of the contact lens of the th material type in the th grid area;

[0078] Data acquisition module: It is used to collect the acoustic characteristic data of the contact lens of the th material type to be tested in the th grid area, as well as the expected detection depth. The acoustic characteristic data includes the sound velocity and acoustic impedance. The expected detection depth is the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects;

[0079] Frequency band classification selection module: It is used to obtain the acoustic characteristic data and the expected detection depth, and perform analysis and processing to generate a frequency selection coefficient. The frequency selection coefficient is used for frequency band classification selection of the ultrasonic probe within the frequency range of 5 MHz to 15 MHz;

[0080] Reference value generation module: It is used to calibrate the selected ultrasonic probe with a defect-free contact lens sample, and use the calibrated ultrasonic probe to scan the defect-free contact lens sample again to obtain reference benchmark data. The reference benchmark data includes the reference intensity value and reference time delay value of the contact lens of the th material type in the th grid area;

[0081] Detection value generation module: It is used to use the calibrated ultrasonic probe to emit ultrasonic waves in the frequency band selected by the frequency selection coefficient, so that the ultrasonic waves penetrate the contact lens to be tested of the th material type and focus on the th grid area. Subsequently, it receives the ultrasonic wave signal reflected from the surface of this area, and analyzes the ultrasonic wave signal reflected from the th grid area to obtain the detection intensity value and detection time delay value of the th grid area;

[0082] Fine-tuning strategy generation module: It is used to obtain the reference intensity value and the detection intensity value, and perform analysis and processing to obtain the first comparison coefficient. At the same time, it obtains the reference time delay value and the detection time delay value, and performs analysis and processing to obtain the second comparison coefficient. Subsequently, the detection system combines and analyzes the first comparison coefficient and the second comparison coefficient to obtain an output result fine-tuning index. The output result fine-tuning index is used to provide a fine-tuning strategy for the defect depth value and defect size value included in the output result.

[0083] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0084] By performing a fine grid division on the surface of the contact lens, targeted defect detection can be carried out according to different material types, obtaining the area where the defect is located, the defect depth value, and the defect size value, obtaining and processing the acoustic characteristic data and the expected detection depth of the area where the defect is located, generating a frequency selection coefficient, and the frequency selection coefficient is used to classify and select the frequency band of the ultrasonic probe within the frequency range of 5 MHz to 15 MHz; calibrating the ultrasonic probe to obtain reference benchmark data; emitting ultrasonic waves with a selected frequency to obtain a detection intensity value and a detection time delay value; analyzing and processing the reference benchmark data, the detection intensity value, and the detection time delay value to obtain a first comparison coefficient and a second comparison coefficient, and performing combined analysis to obtain an output result fine-tuning index, and the output result fine-tuning index is used to provide a fine-tuning strategy for the defect depth value and the defect size value included in the output result. The present invention can significantly improve the accuracy and reliability of contact lens defect detection while ensuring the detection efficiency, providing an efficient and precise automated detection solution for the production quality control of contact lenses. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The drawings of the present invention are only for illustrative purposes and cannot be construed as a limitation of the present invention. In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0086] Figure 1 It is a schematic diagram of the overall method flow of the present invention.

[0087] Figure 2 It is a block diagram of the system module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0088] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0089] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those of ordinary skill in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0090] Embodiment 1

[0091] As Figure 1 shown, a method and system for detecting defects of contact lenses based on image recognition provided in this embodiment specifically include the following steps:

[0092] Step S1: The detection system classifies the material type and divides the surface grid of the contact lens to be tested, including:

[0093] Step S11: The material type classification is to classify the contact lens to be tested according to the material type to form a material type set , where represents the contact lens of the th material type, and represents the total number of material types. The operation steps are as follows:

[0094] Step S111: Use a spectral analyzer to perform spectral scanning on the contact lens to be tested to obtain its spectral characteristics;

[0095] Step S112: Compare the obtained spectral characteristics with a pre-established standard material spectral database, and use the following formula to determine the material type:

[0096]

[0097] where represents the value of the spectral characteristic of the contact lens to be tested at wavelength l; represents the standard spectral value of the kth material at wavelength l, and L represents the total number of wavelengths of the spectrum; represents finding the k value that makes the function value the smallest in the function;

[0098] Step S113: Determine the material type of the contact lens to be tested according to the comparison result 。

[0099] Step S12: Surface mesh generation divides the surface of the contact lens into multiple mesh regions, forming a set of mesh regions , where represents the th mesh region number, represents the total number of mesh generations. The operation steps are as follows:

[0100] Step S121: Use a high-resolution camera to capture an image of the contact lens to be measured, obtaining its surface image;

[0101] Step S122: Preprocess the surface image, including denoising and contrast enhancement operations, to improve the accuracy of subsequent mesh generation;

[0102] Step S123: Use an image processing algorithm to perform mesh generation on the preprocessed image. Specifically, divide the image into equal-area rectangular mesh regions, and the number of each mesh region ranges from 1 to .

[0103] Step S124: Calculate the center coordinates of each mesh region using the following formula:

[0104]

[0105] where and respectively represent the relative horizontal position and vertical position of the center coordinates of the th mesh region in the image; represents the number of rows and columns of the mesh generation, represents the modulo operation, and in the formula calculates divided by and the remainder of this result is then divided by to obtain a value in the range [0,1).

[0106] Step S125: Determine the boundaries of each mesh region and store them as reference data for subsequent detection.

[0107] Step S2: Invoke a pre-trained neural network model for defect detection of the contact lens to be measured. The construction of the neural network model specifically includes the following:

[0108] Step S21: Obtain images of defect-free contact lenses;

[0109] Step S22: Preprocess the images of defect-free contact lenses;

[0110] Step S23: Identify the pattern area in the flawless contact lens image, where the pattern area is the area where flaws occur in the contact lens image;

[0111] Step S24: Generate multiple random flaw blocks;

[0112] Step S25: Repeat the operation of randomly placing the N random flaw blocks into the pattern area of the flawless contact lens image multiple times to obtain multiple flawed contact lens images, where N≥1 and N is a random integer;

[0113] Step S26: Train a neural network model for contact lens flaw detection based on multiple flawed contact lens images, specifically including:

[0114] Step S261: Data segmentation: Divide the dataset into a training set, a validation set, and a test set;

[0115] Step S262: Model construction: Use CNN for feature extraction, construct a neural network architecture, and then connect a fully connected layer to predict the flaw depth value and flaw size value;

[0116] Step S263: Model training: Use the training set to train the model and adjust the network weights so that the model can accurately predict the flaw depth value and flaw size value;

[0117] Step S264: Model validation: Use the validation set to evaluate the performance of the model and adjust the hyperparameters and model structure;

[0118] Step S265: Model testing: Use the test set to evaluate the final model and verify the generalization ability of the model;

[0119] Step S3: The detection system collects the image data of the contact lens to be measured, inputs the image data into the pre-trained neural network model for processing, and finally obtains the output result of flaw detection. The output result includes the flaw depth value and flaw size value of the contact lens of the th material type in the th grid area, specifically including the following content:

[0120] Based on the trained neural network model, perform flaw detection on the contact lens to be measured,

[0121] Step S31: Preprocess the image of the contact lens to be measured to make it meet the model input requirements,

[0122] Step S32: Input the preprocessed image into the trained neural network model;

[0123] Step S33: The model outputs the predicted defect information, including the defect depth value and the defect size value, and marks the defect depth value and the defect size value of the th type of material type contact lens in the th grid area as , .

[0124] Step S4: Collect the acoustic characteristic data and the expected detection depth of the th type of material type contact lens to be tested in the th grid area. The acoustic characteristic data includes the sound velocity and the acoustic impedance, and the expected detection depth is the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects. Specifically, it includes the following content:

[0125] In the th type of material type contact lens to be tested in the th grid area, mark the sound velocity and the acoustic impedance included in the acoustic characteristic data as , respectively. At the same time, mark the expected detection depth as , and the calculation formula is as follows;

[0126]

[0127] Wherein, is the judgment threshold of the defect depth value, is determined according to the experimental data by the expert group to ensure .

[0128] Step S5: Obtain the acoustic characteristic data and the expected detection depth, and perform analysis and processing to generate a frequency selection coefficient. The frequency selection coefficient is used for frequency band classification selection of the ultrasonic probe in the frequency range of 5 MHz to 15 MHz. Specifically, it includes the following content:

[0129] Step S51: Define the frequency selection coefficient as , and the calculation formula is as follows:

[0130]

[0131] Where a, b, and c are all positive weight coefficients, and a, b, and c are determined by experimental data and statistical methods and are used to adjust the influence degree of each parameter on the frequency selection coefficient.

[0132]

[0133] Step S52: Set the frequency band selection formula as:

[0134]

[0135] Among them, is the ultrasonic frequency (unit: MHz) selected by the th type of material type contact lens in the th grid area; The value range of is between 5 MHz and 15 MHz;

[0136] Set the value range of to be (0, 1). When is closer to 0, the selected frequency

[0137] When is closer to 1, the selected frequency is closer to 15 MHz.

[0138] Divide the frequency range from 5 MHz to 15 MHz into 10 equally spaced frequency segments, each with a width of 1 MHz, and sequentially number each interval. These frequency segments are represented as follows:

[0139] Number 1 represents 5 MHz - 6 MHz;

[0140] Number 2 represents 6 MHz - 7 MHz;

[0141] Number 3 represents 7 MHz - 8 MHz;

[0142] Number 4 represents 8 MHz - 9 MHz;

[0143] Number 5 represents 9 MHz - 10 MHz;

[0144] Number 6 represents 10 MHz - 11 MHz;

[0145] Number 7 represents 11 MHz - 12 MHz;

[0146] Number 8 represents 12 MHz - 13 MHz;

[0147] Number 9 represents 13 MHz - 14 MHz;

[0148] Number 10 represents 14 MHz - 15 MHz;

[0149] Step S53: And define the frequency segment number selection function as , and the calculation formula is as follows:

[0150]

[0151] Among them, represents the ceiling function;

[0152] When is close to 0, the selected frequency is close to 5 MHz, the rounded-up value of FB is 1, and the value "1" represents 5 MHz - 6 MHz indicated by number 1;

[0153] When is close to 1, the selected frequency is close to 15 MHz, the rounded-up value of FB is 10, and the value "10" represents 14 MHz - 15 MHz indicated by number 10.

[0154] Step S6: The detection system calibrates the selected ultrasonic probe using a flawless contact lens sample. The calibration content includes distance amplitude calibration (DAC) and sensitivity adjustment, and uses the calibrated ultrasonic probe to scan the flawless contact lens sample again to obtain reference benchmark data. The reference benchmark data includes the reference intensity value and reference time delay value of the contact lens of the th material type in the th grid area; among them, for the contact lens of the th material type, a flawless contact lens sample of the same material type is selected, and the flawless contact lens sample is divided into the same m areas. Use the calibrated ultrasonic probe to scan the flawless contact lens sample again. The intensity value measured by the flawless contact lens in the th area is the reference intensity value of the contact lens of the th material type in the th grid area, and the measured time delay value is the reference time delay value of the contact lens of the th material type in the th grid area. Specifically, it includes the following content:

[0155] Calibrating the ultrasonic probe

[0156] Step S61: Prepare a flawless contact lens sample, that is, select a flawless sample with the same material and thickness as the contact lens to be measured, and ensure that the sample surface is clean without any impurities or damage;

[0157] Step S62: Perform distance amplitude calibration ( ):

[0158] Distance amplitude calibration ( ) is a technique used to compensate for the amplitude attenuation caused by the increase in the propagation distance of sound waves in materials. In ultrasonic testing, sound waves gradually attenuate during propagation due to factors such as absorption and scattering of the material, which means that the amplitude of the reflected signal decreases as the detection distance increases. By , different gains can be set for different distances to correct the natural attenuation during sound wave propagation and ensure consistent signal strength at various distances;

[0159] Use an ultrasonic probe to scan a flawless sample and record the echo signal strength at different depth positions. Based on the recorded data, plot a distance-amplitude curve, which is the curve, and the curve formula is as follows:

[0160]

[0161] where is the normalized amplitude value at depth ; is the actual amplitude value at depth ; is the maximum amplitude value.

[0162] Step S63: Perform sensitivity adjustment:

[0163] According to the curve, adjust the sensitivity of the ultrasonic probe so that the echo signal strengths at different depths are consistent. The sensitivity adjustment formula is as follows:

[0164]

[0165] where is the adjusted sensitivity; is the sensitivity before adjustment; is the reference value; is the actually measured value;

[0166] Step S64: Record the reference intensity value and the reference time delay value:

[0167] Use the adjusted ultrasonic probe to scan the flawless sample again;

[0168] Record the reference intensity value and the reference time delay value for each grid area, and the calculation formula is as follows;

[0169]

[0170] where is the reference intensity value of the contact lens of the th material type in the th grid area; is the amplitude value of the th echo signal; is the total number of echo signals;

[0171] The calculation formula is as follows;

[0172]

[0173] in, It is Type of material contact lenses The reference time delay value in the grid area, It is The time delay value of the echo signal.

[0174] Step S65: Save reference benchmark data:

[0175] The baseline intensity value to be recorded and the reference time delay value Save as reference data Type of material contact lenses The detection intensity values ​​and detection time delay values ​​in the grid areas are marked as , .

[0176] Step S7: Use the calibrated ultrasonic probe to transmit ultrasonic waves of a frequency band selected by the frequency selection coefficient, so that the ultrasonic waves penetrate the first The contact lens of the material type to be tested and the ultrasound is focused on the The ultrasonic signal reflected from the surface of the grid area is then received and analyzed. The ultrasonic signal reflected from the grid area is obtained The detection intensity value and detection time delay value of each grid area are as follows:

[0177] Step S71: Ultrasonic emission:

[0178] Using a calibrated ultrasonic probe, determine the frequency selectivity according to the procedure , select the appropriate frequency The ultrasonic wave is emitted and then focused on the grid area;

[0179] Step S72: Ultrasonic wave reception:

[0180] Ultrasonic waves penetrate the contact lens to be tested and The surface of each grid area is reflected back;

[0181] Receive the reflected ultrasonic signal and use a high-sensitivity receiver to collect the signal;

[0182]

[0183] Among them, is the received signal strength; is the transmitted signal strength; is the reflection coefficient; is the attenuation coefficient; is the distance that the ultrasonic wave propagates;

[0184] Step S73: Signal analysis:

[0185] Perform a fast Fourier transform on the received ultrasonic signal , to extract the frequency components and intensity of the signal. The fast Fourier transform formula is as follows;

[0186]

[0187] Among them, is the Fourier transform result of the received signal;

[0188] is the th sampling point of the received signal; is the total number of sampling points; is the imaginary unit; is the complex exponential function, used to implement the transformation from the time domain to the frequency domain; represents the frequency component;

[0189] Step S74: Detection intensity value and detection time delay value calculation:

[0190] According to the result, calculate the detection intensity value and the detection time delay value of the th grid area;

[0191]

[0192] Among them, is the detection intensity value of the th type of material type contact lens in the th grid area, is the maximum amplitude of the result;

[0193]

[0194] Among them, is the detection time delay value of the th type of material type contact lens in the th grid area, is the frequency corresponding to the maximum amplitude of the result;

[0195] When performing signal analysis using the Fourier transform ( ), the time delay is inferred through phase information, and the relationship between phase, frequency, and time delay is expressed by the formula where is the frequency corresponding phase, is the time delay. In this embodiment, represents , represents ;

[0196] By finding the frequency corresponding to the maximum amplitude and using the phase of this frequency component to calculate the time delay, the calculation formula is as follows:

[0197]

[0198] Step S75: Data saving:

[0199] Save the calculated detection intensity value and the detection time delay value for subsequent defect detection and analysis. It should be noted that the detection value is limited to be less than or equal to the reference value. If there are defects in the contact lens, the detection intensity value or the detection time delay value will decrease, resulting in a ratio less than 1. If there are no defects in the contact lens, the detection value will be close to the reference value, resulting in a ratio close to 1.

[0200] Step S8: The detection system obtains the reference intensity value and the detection intensity value, and performs analysis and processing to obtain the first comparison coefficient. At the same time, it obtains the reference time delay value and the detection time delay value, and performs analysis and processing to obtain the second comparison coefficient. Subsequently, the detection system combines and analyzes the first comparison coefficient and the second comparison coefficient to obtain the output result fine-tuning index, and the output result fine-tuning index is used to provide a fine-tuning strategy for the defect depth value and the defect size value included in the output result. The specific content is as follows:

[0201] Step S81: Define the first comparison coefficient as , and the calculation formula is as follows:

[0202]

[0203] is the detection intensity value of the contact lens of the th material type in the th grid area, is the The reference intensity value of the contact lens of the th material type in the

[0204] value range is (0, 1). When is closer to 0, it indicates that the amplitude of the detected intensity lower than the reference intensity is larger, and the probability of having defects is higher. When is closer to 1, it indicates that the detected intensity is closer to the reference intensity, and the probability of having defects is lower.

[0205] Step S82: Define the second comparison coefficient as , and the calculation formula is as follows:

[0206]

[0207] is the detection time delay value of the contact lens of the th material type in the th grid area, is the reference time delay value of the contact lens of the th material type in the th grid area;

[0208] value range is (0, 1). When is closer to 0, it indicates that the amplitude of the detected time delay lower than the reference time delay is larger, and the probability of having defects is higher. When is closer to 1, it indicates that the detected time delay is closer to the reference time delay, and the probability of having defects is lower.

[0209] Step S83: Define the output result fine-tuning index as , and the calculation formula is as follows:

[0210]

[0211] Among them, is the output result fine-tuning index, used for the fine-tuning strategy of the contact lens of the th material type in the th grid area; is the maximum value of the first comparison coefficient in all grid areas, used for normalization processing; is the maximum value of the second comparison coefficient in all grid areas, used for normalization processing; and are the positive weight coefficients of the first comparison coefficient and the second comparison coefficient respectively, and + = 1, 0.14 ≤ ≤ 0.78, 0.08 ≤ ≤ 0.89, and Determined by experimental data and statistical methods.

[0212] The value range of is (0, 1). When is closer to 0, it indicates a higher probability of defects; when is closer to 1, it indicates a lower probability of defects.

[0213] Step S84: Set the comparison threshold Q1 for the output result fine-tuning index of

[0214] 1) When the output result fine-tuning index is greater than the threshold Q1, the fine-tuning strategy is not to perform fine-tuning, that is, the defect depth value and the defect size value remain unchanged.

[0215] 2) When the output result fine-tuning index is equal to the threshold Q1, 0.47 ≤ Q1 ≤ 0.78, and the specific value of Q1 is determined by the expert group through experimental data and will not be elaborated. The fine-tuning strategy is to slightly adjust the defect depth value and the defect size value The specific adjustments include:[[]]

[0216] 2.1) When and are both greater than or equal to 0.8, the defect depth value increases by 3%, and the defect size value increases by 2%;

[0217] 2.2) When any one of and is less than 0.8, the defect depth value decreases by 2%, and the defect size value decreases by 4%.

[0218] 3) When the output result fine-tuning index is less than the threshold Q1, the fine-tuning strategy includes:[[]]

[0219] 3.1) When the first comparison coefficient is less than 0.3 and the second comparison coefficient is greater than or equal to 0.8:[[]]

[0220] If ≥ at this time, the fine-tuning strategy is a medium adjustment strategy mainly based on The specific adjustment is: the defect depth value increases by 7.5%, and the defect size value increases by 5.5%;

[0221] If < , the fine-tuning strategy is a minor adjustment strategy mainly based on . The specific adjustment is as follows: the defect depth value increases by 3.6%, and the defect size value increases by 4.8%.

[0222] 3.2) When the first comparison coefficient is greater than or equal to 0.8 and the second comparison coefficient is less than 0.3:

[0223] If ≥ , the fine-tuning strategy is a minor adjustment strategy mainly based on . The specific adjustment is as follows: the defect depth value increases by 4.6%, and the defect size value decreases by 3.5%;

[0224] If < , the fine-tuning strategy is a medium adjustment strategy mainly based on . The defect depth value decreases by 5.4%, and the defect size value decreases by 7.6%.

[0225] Embodiment 2

[0226] Please refer to Figure 2 . A contact lens defect detection system based on image recognition. The system is used to execute the contact lens defect detection method based on image recognition described in Embodiment 1, and includes:

[0227] Partitioning module: used to perform material type partitioning and surface grid partitioning on the contact lens to be tested, where:

[0228] The material type partitioning classifies the contact lens to be tested according to the material type. The material type set is represented as , where represents the contact lens of the th material type, and represents the total number of material types;

[0229] The surface grid partitioning divides the surface of the contact lens into multiple grid regions. The grid region set is represented as , where represents the number of the th grid region, and represents the total number of grid partitions;

[0230] Neural network model calling module: used to call a pre-trained neural network model for defect detection of the contact lens to be tested;

[0231] Output result generation module: used to collect the image data of the contact lens to be tested, input the image data into the pre-trained neural network model for processing, and finally obtain the output result of defect detection. The output result includes the defect depth value and defect size value of the contact lens of the th material type in the th grid area;

[0232] Data acquisition module: used to collect the acoustic characteristic data of the contact lens of the th material type to be tested in the th grid area, and the expected detection depth. The acoustic characteristic data includes sound velocity and acoustic impedance, and the expected detection depth is the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects;

[0233] Frequency band classification and selection module: used to obtain the acoustic characteristic data and the expected detection depth, and perform analysis and processing to generate a frequency selection coefficient, which is used to classify and select the frequency band of the ultrasonic probe in the frequency range of 5 MHz to 15 MHz;

[0234] Reference value generation module: used to calibrate the selected ultrasonic probe using a defect-free contact lens sample. The calibration content includes distance amplitude calibration and sensitivity adjustment, and use the calibrated ultrasonic probe to scan the defect-free contact lens sample again to obtain reference data. The reference data includes the reference intensity value and reference time delay value of the contact lens of the th material type in the th grid area;

[0235] Detection value generation module: used to use the calibrated ultrasonic probe to emit ultrasonic waves in the frequency band selected by the frequency selection coefficient, so that the ultrasonic waves penetrate the contact lens to be tested of the th material type and focus on the th grid area, then receive the ultrasonic signal reflected from the surface of this area, and analyze the ultrasonic signal reflected from the th grid area to obtain the detection intensity value and detection time delay value of the th grid area;

[0236] Fine-tuning strategy generation module: used to obtain a reference intensity value and a detection intensity value, and perform analysis and processing to obtain a first comparison coefficient. At the same time, obtain a reference time delay value and a detection time delay value, and perform analysis and processing to obtain a second comparison coefficient. Subsequently, the detection system combines and analyzes the first comparison coefficient and the second comparison coefficient to obtain an output result fine-tuning index, which is used to provide a fine-tuning strategy for the defect depth value and defect size value included in the output result.

[0237] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula that is closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0238] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0239] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, and may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0240] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor limit the invention to only the specific implementation manners. Obviously, according to the content of this specification, many modifications and variations can be made. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the present invention, so that those skilled in the relevant technical fields can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A contact lens defect detection method based on image recognition, characterized in that: The specific steps include: Step S1: Classify the material type and surface mesh of the contact lens to be tested, wherein: The material type classification is to classify the contact lenses to be tested according to the material type to form a material type set. in, Indicates contact lenses of various materials, Indicates the total number of material types; The surface meshing is to divide the surface of the contact lens into a plurality of mesh regions to form a mesh region set. ,in, Indicates The grid area number, Represents the total number of grid divisions; Step S2: calling a pre-trained neural network model to perform defect detection on the contact lens to be tested; Step S3: Collect image data of the contact lens to be tested, and input the image data into the pre-trained neural network model for processing, and finally obtain the output result of defect detection, which includes the first The material type of contact lenses is Defect depth value and defect size value in each grid area; Step S4: Collect the The material type to be tested is contact lenses. Acoustic characteristic data in a grid area, and an expected detection depth, wherein the acoustic characteristic data includes sound velocity and acoustic impedance, and the expected detection depth is the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects; Step S5: Acquire the acoustic characteristic data and the expected detection depth, perform analysis and processing, and generate a frequency selection coefficient, where the frequency selection coefficient is used to classify and select the frequency band of the ultrasonic probe within the frequency range of 5 MHz to 15 MHz; Step S6: Use the flawless contact lens sample to calibrate the selected ultrasonic probe, and use the calibrated ultrasonic probe to scan the flawless contact lens sample again to obtain reference benchmark data, which includes the first The material type of contact lenses is The reference intensity value and reference time delay value in each grid area; Step S7: Using the calibrated ultrasonic probe, transmit ultrasonic waves of the frequency band selected by the frequency selection coefficient, so that the ultrasonic waves penetrate the first The contact lens of the material type to be tested and the ultrasound is focused on the The ultrasonic signal reflected from the surface of the grid area is then received and analyzed. The ultrasonic signal reflected from the grid area is obtained The detection intensity value and detection time delay value of each grid area; Step S8: Obtain a reference intensity value and a detection intensity value, and perform analysis and processing to obtain a first comparison coefficient. At the same time, obtain a reference time delay value and a detection time delay value, and perform analysis and processing to obtain a second comparison coefficient. Subsequently, the detection system combines the first comparison coefficient and the second comparison coefficient for analysis to obtain an output result fine-tuning index. The output result fine-tuning index is used to provide a fine-tuning strategy for the defect depth value and defect size value included in the output result.

2. The method for detecting contact lens defects based on image recognition according to claim 1, characterized in that: The step S3 includes preprocessing the contact lens image to be tested to make it meet the model input requirements; inputting the preprocessed image into the trained neural network model; the model outputs predicted defect information, including defect depth value and defect size value; The material type of contact lenses is The defect depth values ​​and defect size values ​​in the grid areas are marked as , .

3. The method for detecting contact lens defects based on image recognition according to claim 2, characterized in that: In step S4, Type of contact lens In each grid area, the acoustic characteristic data including the sound velocity and acoustic impedance are marked as , , and the expected detection depth is marked as ,and The calculation formula is as follows: in, is the threshold for judging the defect depth value, and ensures .

4. The method for detecting contact lens defects based on image recognition according to claim 3, characterized in that: The frequency selection coefficient is defined in step S5 as: , the calculation formula is as follows: Where a, b, and c are all positive weight coefficients used to adjust the influence of each parameter on the frequency selection coefficient; The formula for setting the frequency band selection is: in, It is The material type of contact lenses is The ultrasonic frequency is selected in each grid area. The value range is between 5MHz and 15MHz; is the frequency selection coefficient; set up The value range is (0,1). The closer it is to 0, the frequency of selection The closer to 5MHz; when The closer it is to 1, the frequency of the selection The closer to 15MHz.

5. The method for detecting contact lens defects based on image recognition according to claim 4, characterized in that: The step S6 of calibrating the ultrasonic probe includes preparing a flawless contact lens sample and performing distance amplitude calibration. , perform sensitivity adjustment and record the reference intensity value and reference time delay value; The recording of the reference intensity value and the reference time delay value is specifically performed by scanning the flawless sample again using the adjusted ultrasonic probe; Record the baseline intensity value for each grid area and the reference time delay value , The calculation formula is as follows; in, It is The material type of contact lenses is The baseline intensity value in each grid area; It is The amplitude value of the echo signal; is the total number of echo signals; The calculation formula is as follows; in, It is The material type of contact lenses is The reference time delay value in the grid area, It is The time delay value of the echo signal.

6. The method for detecting contact lens defects based on image recognition according to claim 5, characterized in that: The step S7 includes ultrasonic emission, ultrasonic reception, signal analysis, detection intensity value and detection time delay value calculation and data storage; The signal analysis is specifically to perform fast Fourier transform on the received ultrasonic signal , to extract the frequency component and intensity of the signal, the fast Fourier transform formula is as follows; in, is the Fourier transform result of the received signal; The received signal sampling points; is the total number of sampling points; p is the imaginary unit; It is a complex exponential function, which is used to realize the transformation from time domain to frequency domain; Represents the frequency component.

7. The method for detecting contact lens defects based on image recognition according to claim 6, characterized in that: The detection intensity value and the detection time delay value are calculated as follows: As a result, calculate the The detection strength value of the grid area and detection time delay value ; in, It is The material type of contact lenses is The detection intensity value in the grid area, yes The maximum amplitude of the result; in, It is The material type of contact lenses is The detection time delay value in the grid area, yes The frequency corresponding to the maximum magnitude of the result.

8. The method for detecting contact lens defects based on image recognition according to claim 7, characterized in that: The first The material type of contact lenses is The reference intensity values ​​and reference time delay values ​​in the grid areas are marked as , ; The material type of contact lenses is The detection intensity values ​​and detection time delay values ​​in the grid areas are marked as , ; Define the first comparison coefficient as , the calculation formula is as follows: It is The material type of contact lenses is The detection intensity value in the grid area, It is The material type of contact lenses is The baseline intensity value in each grid area; The value range of is (0,1). The closer it is to 0, the greater the magnitude of the detection intensity below the reference intensity, and the greater the probability of a defect. The closer it is to 1, the closer the detection intensity is to the reference intensity, and the smaller the probability of defects. Define the second comparison coefficient as , the calculation formula is as follows: It is The material type of contact lenses is The detection time delay value in the grid area, It is The material type of contact lenses is The reference time delay value in the grid area; The value range of is (0,1). The closer it is to 0, the greater the detection time delay is below the reference time delay, and the greater the probability of a defect. The closer it is to 1, the closer the detection time delay is to the reference time delay, and the smaller the probability of a defect.

9. The method for detecting contact lens defects based on image recognition according to claim 8, characterized in that: Define the output result fine-tuning index as , the calculation formula is as follows: in, Fine-tune the index for the output result, used for The material type of contact lenses is Fine-tuning strategies in grid regions; is the maximum value of the first comparison coefficient in all grid areas; is the maximum value of the second comparison coefficient in all grid areas; u1 and u2 are the positive weight coefficients of the first comparison coefficient and the second comparison coefficient, respectively, and u1+u2=1, 0.14≤u1≤0.78, 0.08≤u2≤0.89; The value range of is (0,1). The closer it is to 0, the higher the probability of defect. The closer it is to 1, the lower the probability of defects; Set the output result fine-tuning index The comparison threshold is Q1; 1) When the output result is fine-tuned index When the defect depth value is greater than the threshold Q1, the fine-tuning strategy is not to perform fine-tuning. and defect size value remain unchanged; 2) When the output result is fine-tuned index When the threshold Q1 is equal, the fine-tuning strategy is to adjust the defect depth value and defect size value Make minor adjustments; 3) When the output result is fine-tuned index When it is less than the threshold Q1, the fine-tuning strategy includes: 3.1) When the first comparison coefficient Less than 0.3, the second comparison coefficient When greater than or equal to 0.8: If u1≥u2, the fine-tuning strategy is A medium adjustment strategy based on If u1 < u2, the fine-tuning strategy is a minor adjustment strategy mainly based on ; 3.2) When the first comparison coefficient Greater than or equal to 0.8, the second comparison coefficient When less than 0.3: If u1≥u2, the fine-tuning strategy is Minor adjustments to the strategy; If u1 < u2, the fine-tuning strategy is a medium adjustment strategy mainly based on .

10. A contact lens defect detection system based on image recognition, characterized in that: The system is used to execute the contact lens defect detection method based on image recognition according to any one of claims 1 to 9, comprising: Division module: used to divide the material type and surface mesh of the contact lens to be tested; Neural network model calling module: used to call the pre-trained neural network model for defect detection of contact lenses to be tested; Output result generation module: used to collect image data of the contact lens to be tested, and input the image data into the pre-trained neural network model for processing, and finally obtain the output result of defect detection. The output result includes the first The material type of contact lenses is Defect depth value and defect size value in each grid area; Data acquisition module: used to collect The material type to be tested is contact lenses. The acoustic characteristic data in each grid area, and the expected detection depth, the acoustic characteristic data including the acoustic velocity and acoustic impedance, the expected detection depth being the maximum depth at which the ultrasonic probe penetrates the contact lens material and detects surface defects; Frequency band classification and selection module: used to obtain acoustic characteristic data and expected detection depth, and perform analysis and processing to generate a frequency selection coefficient, which is used to classify and select the frequency band of the ultrasonic probe within the frequency range of 5 MHz to 15 MHz; Reference value generation module: used to calibrate the selected ultrasonic probe using a flawless contact lens sample, and use the calibrated ultrasonic probe to scan the flawless contact lens sample again to obtain reference reference data. The reference reference data includes the first The material type of contact lenses is The reference intensity value and reference time delay value in each grid area; Detection value generation module: used to use the calibrated ultrasonic probe to transmit ultrasonic waves of the frequency band selected by the frequency selection coefficient, so that the ultrasonic waves penetrate the first of the contact lenses to be tested and focus on The ultrasonic signal reflected from the surface of the grid area is then received and analyzed. The ultrasonic signal reflected from the grid area is obtained The detection intensity value and detection time delay value of each grid area; Fine-tuning strategy generation module: used to obtain the reference intensity value and the detection intensity value, and perform analysis and processing to obtain the first comparison coefficient. At the same time, it obtains the reference time delay value and the detection time delay value, and performs analysis and processing to obtain the second comparison coefficient. Subsequently, the detection system combines the first comparison coefficient and the second comparison coefficient for analysis to obtain the output result fine-tuning index. The output result fine-tuning index is used to provide a fine-tuning strategy for the defect depth value and defect size value included in the output result.

Images