A method, system, device and computer readable storage medium for tuning detection parameters of an automated optical defect detection apparatus

Abstract

The application discloses a kind of detection parameter optimization method, system, equipment and computer readable storage medium of automatic optical defect detection equipment, comprising: control equipment to be detected sample detection, obtain multiple sample images of each region of interest;Based on the preset multidimensional evaluation function, the comprehensive score of each sample image is obtained;According to the target detection parameter combination of comprehensive score determination;According to target detection parameter combination, parameter optimization is carried out to equipment.Multiple-dimensional evaluation function includes signal peak value, signal background ratio, texture complexity and non-periodic signal function.The application is through automation traversal and quantitative score, combined with gray feedback closed loop, solves the problem of low efficiency and poor consistency of manual optimization, realizes parameter automatic optimization, improves detection accuracy and consistency.

Description

Technical Field

[0001] This invention relates to the field of optical inspection technology, and specifically to a method, system, device, and computer-readable storage medium for optimizing the detection parameters of an automatic optical defect detection device. Background Technology

[0002] Currently, Automatic Optical Inspection (AOI) equipment is a key device that utilizes optical principles combined with image processing technology to automatically detect surface defects in objects. It is widely used in fields such as semiconductor wafer, printed circuit board, and precision metal surface inspection. The detection performance of AOI equipment is highly dependent on the configuration of optical imaging parameters, including illumination intensity, polarization state combination, and filter combination.

[0003] In related technologies, manual trial and error methods are typically used to configure detection parameters. Operators manually switch different polarizer angles and filters, adjust the light source brightness, and observe the image effect on the monitor with the naked eye, subjectively judging which set of parameters shows the most obvious defects.

[0004] However, traditional manual optimization methods have significant drawbacks: manual trial and error is extremely time-consuming, severely impacting the rapid capacity switching of production lines; different operators have varying understandings of image quality standards, leading to poor consistency in detection results; and relying on visual observation makes it impossible to quantify key indicators such as signal-to-noise ratio and texture complexity, making it difficult to find the globally optimal solution and often resulting in only locally satisfactory solutions. This leads to low efficiency, poor consistency, and a lack of quantitative standards in configuring detection parameters, ultimately limiting the system's detection accuracy and resulting in a high rate of false positives or false negatives. Summary of the Invention

[0005] This invention provides a method, system, device, and computer-readable storage medium for optimizing the detection parameters of an automatic optical defect detection device, which can solve the problems of low efficiency, poor consistency, and lack of quantitative standards caused by relying on manual experience for optimization in the prior art.

[0006] In a first aspect, embodiments of the present invention provide a method for optimizing the detection parameters of an automatic optical defect detection device, comprising: The automatic optical defect detection equipment is controlled to detect the acquired sample to be tested, and multiple sample images of each region of interest to be detected in the sample to be tested are obtained. Based on a pre-set multidimensional evaluation function and each of the sample images, a comprehensive score for each of the sample images is obtained; Based on the comprehensive score of each sample image, the combination of target detection parameters is determined; The automatic optical defect detection device is optimized by adjusting the parameters according to the target detection parameter combination corresponding to each of the sample images.

[0007] In conjunction with the first aspect, in one implementation, the preset multidimensional evaluation function includes a preset signal peak function, a signal-to-background ratio function, a texture complexity function, and an aperiodic signal function; The process of obtaining a comprehensive score for each sample image based on a preset multidimensional evaluation function and each sample image includes: Based on the signal peak function and each of the sample images, the signal peak value of each of the sample images is calculated; Based on the signal-to-background ratio function and each of the sample images, the signal-to-background ratio value of each of the sample images is calculated; Based on the texture complexity function and each of the sample images, the texture complexity value of each of the sample images is calculated. Based on the aperiodic signal function and each of the sample images, each of the aperiodic signal values ​​is calculated; The comprehensive score of each sample image is obtained based on the signal peak value, the signal-to-background ratio, the texture complexity value, and the aperiodic signal value of each sample image.

[0008] In conjunction with the first aspect, in one implementation, calculating the signal peak value of each of the sample images based on the signal peak function and each of the sample images includes: Extract the gray level with the highest frequency of occurrence in the region of interest to be detected in the sample image, and calculate the signal peak value using a signal peak function. The calculation formula is:

[0009] in, It is a grayscale level, and the grayscale level range is... , For each gray level The number of pixels.

[0010] In conjunction with the first aspect, in one implementation, calculating the signal-to-background ratio of each of the sample images based on the signal-to-background ratio function and each of the sample images includes: The average gray value of the region of interest to be detected is calculated as the signal intensity, and the average gray value of the background region is calculated as the background intensity. The signal-to-background ratio (SBR) is then calculated using the signal-to-background ratio function. The formula for calculating the SBR is as follows:

[0011] in, This represents the average grayscale value of all pixels in the signal region. This represents the average grayscale value of all pixels in the background area.

[0012] In conjunction with the first aspect, in one implementation, obtaining the comprehensive score of each sample image based on the signal peak value, the signal-to-background ratio, the texture complexity value, and the aperiodic signal value of each sample image includes: Set the priorities for the signal peak function, the signal-to-background ratio function, the texture complexity function, and the aperiodic signal function; The weighting coefficients corresponding to each of the multidimensional evaluation functions are determined according to the priority. The comprehensive score is obtained by weighting and summing the signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value of each sample image using the weighting coefficients.

[0013] In conjunction with the first aspect, in one embodiment, the step of optimizing the parameters of the automatic optical defect detection device based on the target detection parameter combinations corresponding to each of the sample images includes: The target detection parameters are combined and configured into an automated optical defect detection device; Obtain the preset target grayscale value; Acquire a verification image after applying the target detection parameter combination, and obtain the current grayscale value of the verification image; Calculate the deviation between the current grayscale value and the target grayscale value to perform parameter tuning.

[0014] In conjunction with the first aspect, in one embodiment, after calculating the deviation between the current grayscale value and the target grayscale value, the method further includes: If the deviation value is greater than the preset deviation threshold, the illumination intensity is adjusted, and the process returns to the steps of acquiring and verifying the image, obtaining the current grayscale value, and calculating the deviation. If the deviation value is less than or equal to the preset deviation threshold, then the parameter tuning is considered complete.

[0015] Secondly, embodiments of the present invention provide a detection parameter optimization system for an automatic optical defect detection device, the detection parameter optimization system for the automatic optical defect detection device comprising: The image acquisition module is used to control the automatic optical defect detection equipment to detect the acquired sample to be tested and acquire multiple sample images of each region of interest to be detected in the sample to be tested. The scoring acquisition module is used to obtain the comprehensive score of each sample image based on a preset multidimensional evaluation function and each sample image. The target determination module is used to determine the combination of target detection parameters based on the comprehensive score of each sample image; The parameter tuning module is used to tune the parameters of the automatic optical defect detection equipment based on the target detection parameter combinations corresponding to each sample image.

[0016] Thirdly, embodiments of the present invention provide an automatic optical defect detection device, including an optical component, a processor, and a memory, wherein the processor is communicatively connected to the optical component, and the processor executes the steps of the detection parameter optimization method of the automatic optical defect detection device as described in any one of claims 1 to 7.

[0017] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a detection parameter optimization program for an automatic optical defect detection device, wherein when the detection parameter optimization program for the automatic optical defect detection device is executed by a processor, it implements the steps of the detection parameter optimization method for the automatic optical defect detection device as described in any one of claims 1 to 7.

[0018] The beneficial effects of the technical solution provided by the embodiments of the present invention include: obtaining the detection interest area by acquiring and analyzing the process formula and layout information of the sample to be tested; controlling the optical components to switch sequentially to each combination of detection parameters to acquire sample images; calculating the score of each sample image based on the multidimensional evaluation function to determine the target detection parameter combination and applying it to the equipment. This solves the technical problems in related technologies, such as low optimization efficiency caused by relying on manual experience to try different polarization states, filters and light intensity combinations one by one, which seriously affects the rapid switching of production capacity; and poor consistency of detection results caused by relying on visual observation of image effects and subjective judgment, which lacks quantitative standards and makes it difficult to find the global optimal solution. The present invention adopts an automated traversal and multidimensional quantitative evaluation mechanism to improve the parameter optimization efficiency, avoid subjective bias caused by reliance on manual experience, and improve detection consistency and detection accuracy. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating an embodiment of the method for optimizing the detection parameters of the automatic optical defect detection equipment of the present invention; Figure 2 A schematic diagram of the functional modules of the detection parameter optimization system of the automatic optical defect detection equipment provided in this embodiment of the invention; Figure 3 This is a diagram showing the combination configuration of three polarization states on the illumination side and three polarization states on the imaging side in an embodiment of the automatic optical defect detection equipment optimization method of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0022] Automatic optical defect detection equipment: This is a device that uses optical imaging principles combined with image processing technology to automatically detect defects on the surface of an object. It includes optical components, a processor, and a memory.

[0023] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0024] In a first aspect, embodiments of the present invention provide a method for optimizing the detection parameters of an automatic optical defect detection device.

[0025] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for optimizing the detection parameters of the automatic optical defect detection equipment of the present invention. Figure 1 As shown, the method for optimizing the detection parameters of an automated optical defect detection device includes: The automatic optical defect detection equipment is controlled to detect the acquired sample to be tested, and multiple sample images of each region of interest to be detected in the sample are obtained. As an example, the region of interest to be inspected is determined by analyzing the process formula and layout information. The optical components of the automated optical defect inspection equipment are then controlled to sequentially switch to various combinations of detection parameters to inspect the sample. The combinations of detection parameters include polarization state combinations, filter combinations, and light intensity parameters. The switching of optical components is achieved by a motor driving the polarizer wheel and filter wheel to rotate. After each switch, mechanical stabilization is waited before the camera exposure is triggered. The automated optical defect inspection equipment system traverses each combination of detection parameters, controls the camera to acquire the original image containing the region of interest to be inspected, and extracts the pixel data located in the region of interest to be inspected from the original image as a sample image for subsequent calculations, eliminating interference from irrelevant background data.

[0026] Based on a pre-defined multidimensional evaluation function and each sample image, a comprehensive score for each sample image is obtained. The exemplary pre-defined multidimensional evaluation functions include a pre-defined signal peak value function, a signal-to-background ratio function, a texture complexity function, and an aperiodic signal function. Based on the multidimensional evaluation functions, the sample images are quantitatively evaluated by calculating the signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value for each sample image, and obtaining a comprehensive score based on the signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value for each sample image.

[0027] Based on the comprehensive score of each sample image, the combination of target detection parameters is determined; As an example, the priorities of the signal peak function, signal-to-background ratio function, texture complexity function, and aperiodic signal function are set. The weighting coefficients corresponding to each multidimensional evaluation function are determined according to the priorities. The signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value of each sample image are weighted and summed using the weighting coefficients to obtain a comprehensive score. The detection parameter combination with the highest comprehensive score is determined as the target detection parameter combination.

[0028] The parameters of the automatic optical defect detection equipment are optimized based on the target detection parameter combinations corresponding to each sample image. As an example, the target detection parameter combination is configured into the automatic optical defect detection equipment, a preset target grayscale value is acquired, a verification image after applying the target detection parameter combination is acquired, the current grayscale value of the verification image is acquired, and the deviation value between the current grayscale value and the target grayscale value is calculated for parameter optimization. The preset target grayscale value ranges from 0 to 255, with 200 being a commonly used value. If the deviation value is greater than the preset deviation threshold, the illumination intensity is adjusted, and the process returns to the steps of acquiring the verification image, acquiring the current grayscale value, and calculating the deviation. If the deviation value is less than or equal to the preset deviation threshold, the parameter optimization is considered complete.

[0029] In this embodiment, sample images are acquired by control equipment, and the target parameter combination is determined by calculating a comprehensive score based on a multi-dimensional evaluation function. The grayscale value feedback is then used for optimization. This solves the technical problems in related technologies, such as low optimization efficiency caused by relying on manual experience to try different polarization states, filters, and light intensity combinations one by one, which seriously affects the rapid switching of production capacity, and poor consistency of detection results caused by relying on visual observation of image effects and subjective judgment, which lacks quantitative standards and makes it difficult to find the global optimal solution. An automated traversal and multi-dimensional quantitative evaluation mechanism are adopted to improve the parameter optimization efficiency, avoid subjective bias caused by reliance on manual experience, and improve detection consistency and accuracy.

[0030] Furthermore, in one embodiment, based on a preset multidimensional evaluation function and each sample image, a comprehensive score for each sample image is obtained, including: Based on the signal peak function and each sample image, the signal peak value of each sample image is calculated; Based on the signal-to-background ratio function and each sample image, the signal-to-background ratio of each sample image is calculated. Based on the texture complexity function and each sample image, the texture complexity value of each sample image is calculated. Based on the non-periodic signal function and each sample image, the values ​​of each non-periodic signal are calculated. A comprehensive score for each sample image is obtained based on its signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value.

[0031] As an example, a mathematical model is introduced for quantitative evaluation. The signal peak function is used to ensure that the defect signal is not overwhelmed. The gray level with the highest frequency of occurrence in the region of interest to be detected in the sample image is extracted, and the signal peak is calculated by the signal peak function. The signal-to-background ratio function is the ratio of the average gray value of the pixels in the signal region to the average gray value of the pixels in the background region. The texture complexity function is defined as the average value of the gradient magnitude map, where the gradient is calculated using the Sobel algorithm. The non-periodic signal function is obtained by performing frequency domain analysis on the image. If there are unexpected energy peaks in the high-frequency region, they are judged as noise. This index is used to suppress parameter combinations with high noise.

[0032] Further, in one embodiment, the signal peak value of each sample image is calculated based on the signal peak function and each sample image, including: Extract the gray level with the highest frequency of occurrence in the region of interest to be detected in the sample image, and calculate the signal peak value using the signal peak function. Let the gray image... The size is The grayscale range is (generally ), statistic for each gray level Number of pixels Then the formula for calculating the peak value P of the signal is:

[0033] Further, in one embodiment, the signal-to-background ratio of each sample image is calculated based on the signal-to-background ratio function and each sample image, including: The average gray value of the region of interest to be detected is calculated as the signal intensity, and the average gray value of the background region is calculated as the background intensity. The signal-to-background ratio (SBR) is then calculated using the signal-to-background ratio function. The formula for calculating the SBR is as follows:

[0034] in, This represents the average grayscale value of all pixels in the signal region. This represents the average grayscale value of all pixels in the background area.

[0035] As an example, signal-to-background segmentation can be achieved using a global thresholding method. This method employs a maximum inter-class variance algorithm to automatically segment pixels within the target region into foreground and background classes, with the foreground corresponding to the signal region. Let the segmentation threshold be... ,but:

[0036] in, For the first grayscale value of each pixel. , These represent the number of pixels in the foreground and background, respectively.

[0037] Furthermore, in one embodiment, the texture complexity value of each sample image is calculated based on the texture complexity function and each sample image, including: The Sobel operator is used to calculate the gradients in the horizontal and vertical directions of the image, and the average value of the gradient magnitude map is used as the texture complexity value.

[0038] Exemplary, for grayscale images The horizontal gradient was calculated using the Sobel operator. and vertical gradient The formula for calculating the texture complexity function is as follows:

[0039] in, As a discrete difference operator, the image is approximated by the central difference. First-order partial derivative in the direction, As a discrete difference operator, the image is approximated by the central difference. The first-order partial derivatives in the two directions mentioned above can be used to calculate the gradient magnitude of the image. for:

[0040] Texture complexity Defined as the average value of the gradient magnitude plot: in, These represent the height and width of the image, respectively.

[0041] Furthermore, in one embodiment, the aperiodic signal values ​​are calculated based on the aperiodic signal function and each sample image, including: By performing frequency domain analysis on the image, the energy proportion of the high-frequency region is calculated to obtain the aperiodic signal value.

[0042] As an example, by performing frequency domain analysis on the image, calculating the energy proportion of high-frequency regions to obtain the number of aperiodic signals, and then performing a two-dimensional Fast Fourier Transform (FFT) on the image and centering it, the formula is as follows:

[0043] in, Image in pixel coordinates grayscale value at that location The horizontal frequency in the frequency domain. The frequency in the vertical direction is the frequency. These are the height and width of the image, respectively; Amplitude spectrum Let the center of the spectrum be Preset radius threshold ( (If the value can be 0.1~0.2), then:

[0044] in, The energy of the periodic low-frequency component, This represents the total energy of the spectrum.

[0045] Furthermore, in one embodiment, a comprehensive score for each sample image is obtained based on its signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value, including: Since different evaluation functions have different dimensions, normalization is required to unify the evaluation criteria. The weighting coefficients can be adjusted according to the type of detection task. For example, for the detection of minute scratches, the weight of texture complexity can be increased; for the detection of foreign objects, the weight of signal peak value can be increased. The formula for calculating the comprehensive score is as follows:

[0046] in, For the evaluation function index, For the first The weighting coefficients of each evaluation function. For the first The normalized values ​​of the evaluation functions To determine the total number of evaluation functions to be selected, the combination of detection parameters with the highest comprehensive score is determined as the target detection parameter combination. If multiple combinations have the same score, the combination with the lower light intensity parameter is selected first.

[0047] Furthermore, in one embodiment, the automatic optical defect detection device is optimized based on the target detection parameter combinations corresponding to each sample image, including: Configure the target detection parameters into the automated optical defect detection equipment; Obtain the preset target grayscale value; Acquire a verification image after combining the target detection parameters, and obtain the current grayscale value of the verification image; Calculate the deviation between the current grayscale value and the target grayscale value to optimize the parameters.

[0048] In the exemplary automatic tuning mode, the system uses feedback control to adjust the light source intensity so that the average gray value of the image reaches the preset target value. If, during the tuning process, all combined scores are found to be lower than the preset threshold, an alarm signal is generated to prompt the system to check the status of the optical components.

[0049] Furthermore, in one embodiment, after calculating the deviation between the current grayscale value and the target grayscale value, the method further includes: If the deviation value is greater than the preset deviation threshold, the illumination intensity is adjusted, and the process returns to the steps of acquiring and verifying the image, obtaining the current grayscale value, and calculating the deviation. If the deviation value is less than or equal to the preset deviation threshold, then the parameter tuning is considered complete.

[0050] As an example, closed-loop feedback adjustment ensures that the final image brightness remains stable within the target range, eliminating the impact of light source fluctuations on the detection results.

[0051] In this embodiment, through the above steps, the judgment based on human experience is transformed into a multi-dimensional weighted function calculation and ranking, thereby realizing the automated optimization of detection parameters. The best imaging mode can be obtained without relying on human experience, which improves debugging efficiency and detection consistency.

[0052] Secondly, embodiments of the present invention also provide a system for optimizing the detection parameters of an automatic optical defect detection device.

[0053] In one embodiment, reference is made to Figure 2 , Figure 2 This is a functional module diagram of an embodiment of the detection parameter optimization system of the automatic optical defect detection equipment of the present invention, as shown below. Figure 2 As shown, the parameter optimization system for the automated optical defect detection equipment includes: The image acquisition module is used to control the automatic optical defect detection equipment to detect the acquired sample to be tested and acquire multiple sample images of each region of interest to be detected in the sample to be tested. The scoring acquisition module is used to obtain the comprehensive score of each sample image based on a preset multidimensional evaluation function and each sample image. The target determination module is used to determine the combination of target detection parameters based on the comprehensive score of each sample image; The parameter tuning module is used to tune the parameters of the automatic optical defect detection equipment based on the target detection parameter combinations corresponding to each sample image.

[0054] Furthermore, in one embodiment, the scoring acquisition module includes a peak value calculation unit, a background ratio calculation unit, a texture complexity calculation unit, and an aperiodic signal calculation unit, which are used to calculate the signal peak value, the signal-to-background ratio, the texture complexity value, and the aperiodic signal value, respectively.

[0055] Furthermore, in one embodiment, the parameter optimization module includes a configuration unit, a grayscale acquisition unit, and a deviation calculation unit. The configuration unit is used to configure the target detection parameter combination to the device, the grayscale acquisition unit is used to acquire the preset target grayscale value and the current grayscale value of the verification image, and the deviation calculation unit is used to calculate the deviation value and control the adjustment of the illumination intensity.

[0056] Furthermore, in one embodiment, the image acquisition module is configured to drive a motor to rotate the polarizer wheel and filter wheel to switch the combination of detection parameters and control the camera to acquire the original image.

[0057] Furthermore, in one embodiment, the scoring acquisition module is also configured to use the maximum inter-class variance algorithm for foreground and background segmentation, and to use the Sobel algorithm to calculate the gradient magnitude.

[0058] The functions of each module in the aforementioned automatic optical defect detection equipment's parameter tuning system correspond to the steps in the aforementioned automatic optical defect detection equipment's parameter tuning method embodiment, and their functions and implementation processes will not be elaborated here. The functions of each module are implemented by the processor executing programs stored in memory.

[0059] Thirdly, embodiments of the present invention provide an automatic optical defect detection device, which includes optical components (such as light source, lens, polarizer wheel, filter wheel, etc.) and control circuit. The processor and the optical components are connected in a communication connection method commonly used in the art (such as through GPIO interface, serial port or motion control card). The processor drives the optical components to switch parameters by sending control commands. The device uses its built-in optical components to perform image acquisition and uses the processor to execute the above-mentioned optimization logic, thereby realizing the automatic or semi-automatic optimization of detection parameters.

[0060] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0061] The processor can be a general-purpose processor, which can call the detection parameter tuning program of the automatic optical defect detection device stored in the memory and execute the detection parameter tuning method of the automatic optical defect detection device provided in the embodiments of the present invention. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the detection parameter tuning program of the automatic optical defect detection device is called can refer to the various embodiments of the detection parameter tuning method of the automatic optical defect detection device of the present invention, which will not be repeated here. The processor may also include a graphics processing unit (GPU) for accelerating the parallel calculation of the image evaluation function.

[0062] Those skilled in the art will understand that the hardware structure shown in the figures does not constitute a limitation of the present invention, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0063] Fourthly, embodiments of the present invention also provide a computer-readable storage medium.

[0064] The present invention stores a detection parameter optimization program for an automatic optical defect detection device on a computer-readable storage medium. When the detection parameter optimization program for the automatic optical defect detection device is executed by a processor, it implements the steps of the detection parameter optimization method for the automatic optical defect detection device as described above.

[0065] The method implemented when the detection parameter optimization program of the automatic optical defect detection equipment is executed can be referred to in various embodiments of the detection parameter optimization method of the automatic optical defect detection equipment of the present invention, and will not be repeated here.

[0066] It should be noted that the sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0068] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0069] In the description of the embodiments of the present invention, terms such as "exemplary," "for example," or "for instance" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design that is described as "exemplary," "for example," or "for instance" in the embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a specific manner.

[0070] In the description of the embodiments of the present invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" means two or more.

[0071] In some processes described in the embodiments of the present invention, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of the present invention, or may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0072] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of the present invention.

[0073] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for optimizing the detection parameters of an automatic optical defect detection device, characterized in that, include: The automatic optical defect detection equipment is controlled to detect the acquired sample to be tested, and multiple sample images of each region of interest to be detected in the sample to be tested are obtained. Based on a pre-set multidimensional evaluation function and each of the sample images, a comprehensive score for each of the sample images is obtained; Based on the comprehensive score of each sample image, the combination of target detection parameters is determined; The automatic optical defect detection device is optimized by adjusting the parameters according to the target detection parameter combination corresponding to each of the sample images.

2. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 1, characterized in that, The preset multidimensional evaluation functions include a preset signal peak value function, a signal-to-background ratio function, a texture complexity function, and a non-periodic signal function; The process of obtaining a comprehensive score for each sample image based on a preset multidimensional evaluation function and each sample image includes: Based on the signal peak function and each of the sample images, the signal peak value of each of the sample images is calculated; Based on the signal-to-background ratio function and each of the sample images, the signal-to-background ratio value of each of the sample images is calculated; Based on the texture complexity function and each of the sample images, the texture complexity value of each of the sample images is calculated. Based on the aperiodic signal function and each of the sample images, each of the aperiodic signal values ​​is calculated; The comprehensive score of each sample image is obtained based on the signal peak value, the signal-to-background ratio, the texture complexity value, and the aperiodic signal value of each sample image.

3. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 2, characterized in that, The step of calculating the signal peak value of each sample image based on the signal peak function and each sample image includes: Extract the gray level with the highest frequency of occurrence in the region of interest to be detected in the sample image, and calculate the signal peak value using a signal peak function. The calculation formula is: in, It is a grayscale level, and the grayscale level range is... , For each gray level The number of pixels.

4. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 2, characterized in that, The step of calculating the signal-to-background ratio of each sample image based on the signal-to-background ratio function and each sample image includes: The average gray value of the region of interest to be detected is calculated as the signal intensity, and the average gray value of the background region is calculated as the background intensity. The signal-to-background ratio (SBR) is then calculated using the signal-to-background ratio function. The formula for calculating the SBR is as follows: in, This represents the average grayscale value of all pixels in the signal region. This represents the average grayscale value of all pixels in the background area.

5. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 2, characterized in that, The step of obtaining the comprehensive score for each sample image based on the signal peak value, the signal-to-background ratio, the texture complexity value, and the aperiodic signal value of each sample image includes: Set the priorities for the signal peak function, the signal-to-background ratio function, the texture complexity function, and the aperiodic signal function; The weighting coefficients corresponding to each of the multidimensional evaluation functions are determined according to the priority. The comprehensive score is obtained by weighting and summing the signal peak value, signal-to-background ratio, texture complexity value, and aperiodic signal value of each sample image using the weighting coefficients.

6. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 1, characterized in that, The step of optimizing the parameters of the automatic optical defect detection device based on the target detection parameter combinations corresponding to each of the sample images includes: The target detection parameters are combined and configured into an automated optical defect detection device; Obtain the preset target grayscale value; Acquire a verification image after applying the target detection parameter combination, and obtain the current grayscale value of the verification image; Calculate the deviation between the current grayscale value and the target grayscale value to perform parameter tuning.

7. The method for optimizing the detection parameters of the automatic optical defect detection equipment according to claim 6, characterized in that, After calculating the deviation between the current grayscale value and the target grayscale value, the method further includes: If the deviation value is greater than the preset deviation threshold, the illumination intensity is adjusted, and the process returns to the steps of acquiring and verifying the image, obtaining the current grayscale value, and calculating the deviation. If the deviation value is less than or equal to the preset deviation threshold, then the parameter tuning is considered complete.

8. A parameter optimization system for an automatic optical defect detection device, characterized in that, include: The image acquisition module is used to control the automatic optical defect detection equipment to detect the acquired sample to be tested and acquire multiple sample images of each region of interest to be detected in the sample to be tested. The scoring acquisition module is used to obtain the comprehensive score of each sample image based on a preset multidimensional evaluation function and each sample image. The target determination module is used to determine the combination of target detection parameters based on the comprehensive score of each sample image; The parameter tuning module is used to tune the parameters of the automatic optical defect detection equipment based on the target detection parameter combinations corresponding to each sample image.

9. An automatic optical defect detection device, comprising an optical component, a processor, and a memory, wherein the processor is communicatively connected to the optical component, characterized in that, The processor executes the steps of the method for optimizing the detection parameters of the automatic optical defect detection device according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a detection parameter optimization program for an automatic optical defect detection device, wherein when the detection parameter optimization program for the automatic optical defect detection device is executed by a processor, it implements the steps of the detection parameter optimization method for the automatic optical defect detection device as described in any one of claims 1 to 7.

Images