Light source adjustment method, defect detection method, device and system, storage medium

Abstract

Embodiments of the present application provide a light source adjusting method, a defect detection method, a device and a system. The method comprises: acquiring images of a target object under corresponding detection light sources collected by each scanning probe in a previous scanning period, and determining image parameters of each image; determining whether the image parameters meet the requirements; for each scanning probe, if the corresponding image parameters do not meet the requirements, adjusting the light source parameters of the detection light sources in the current scanning period based on the differences between the image parameters and target values, and determining the true values of the image parameters of the images obtained in the current scanning period and the fitting values of the image parameters of the images based on a fitting function, and returning to the step of determining whether the image parameters meet the requirements; if the corresponding image parameters meet the requirements, determining the target light source parameters of the corresponding detection light sources based on the true values of the image parameters of the images obtained in the previous scanning period and the fitting values of the image parameters of the images.

Description

Technical Field

[0001] This application relates to the field of product testing technology, and in particular to a light source adjustment method, a defect detection method, a defect detection device, a computer-readable storage medium, and a defect detection system for defect detection. Background Technology

[0002] Currently, in the field of panel inspection, the main purpose of panel inspection is to detect various defects present during the manufacturing process, such as particles, dirt, scratches, bumps, dents, short circuits, and open circuits. The primary method used is to scan the panel using multiple scanning probes. Based on the captured images, image processing algorithms are used to analyze the various defects on the panel. At this point, the quality of the captured images directly affects the defect detection capability. Key aspects of image quality include brightness and clarity. However, the brightness and clarity of images captured by scanning probes under the same acquisition conditions vary depending on the stage of the panel product's manufacturing process. Furthermore, hardware differences between multiple scanning probes will also result in variations in the brightness and clarity of images captured by different probes under the same acquisition conditions, thus affecting the accuracy of the panel defect detection results.

[0003] Therefore, it is crucial to improve the quality of images acquired by scanning probes during panel inspection and to ensure the consistency of defect detection capabilities among multiple scanning probes. Summary of the Invention

[0004] To address the existing technical problems, embodiments of this application provide a light source adjustment method, a defect detection method, a defect detection device, a computer-readable storage medium, and a defect detection system for defect detection that can improve the image quality acquired during the detection process and enhance the consistency of defect detection capabilities among multiple scanning probes.

[0005] To achieve the above objectives, the technical solution of this application embodiment is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a light source adjustment method for defect detection, comprising:

[0007] Acquire images of the target object under the corresponding detection light source collected by each scanning probe in the previous scanning cycle, and determine the image parameters of each image;

[0008] Determine whether the image parameters meet the requirements;

[0009] For each scanning probe, if the corresponding image parameters do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters and the target value. The true value of the image parameters of the image obtained in the current scanning cycle and the fitted value of the image parameters of the image are determined based on the fitting function. Then, the step of determining whether the image parameters meet the requirements is returned.

[0010] If the corresponding image parameters meet the requirements, the target light source parameters corresponding to the detection light source are determined based on the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained in the previous scanning cycle.

[0011] Secondly, embodiments of this application provide a defect detection method, including:

[0012] Acquire images of the target object under the corresponding detection light source collected by each scanning probe in the previous scanning cycle, and determine the image parameters of each image;

[0013] Determine whether the image parameters meet the requirements;

[0014] For each scanning probe, if the corresponding image parameters do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters and the target value. The true value of the image parameters of the image obtained in the current scanning cycle and the fitted value of the image parameters of the image are determined based on the fitting function. Then, the step of determining whether the image parameters meet the requirements is returned.

[0015] If the corresponding image parameters meet the requirements, the target light source parameters corresponding to the detection light source are determined based on the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained in the previous scanning cycle.

[0016] Maintaining the target light source parameters of each of the detection light sources, performing image acquisition of the target object in the next scanning cycle, and performing defect analysis on the target object based on the acquired image.

[0017] Thirdly, embodiments of this application provide a defect detection device, including a processor, a memory connected to the processor, and a computer program stored in the memory and executable by the processor. When the computer program is executed by the processor, it implements the steps of the defect detection method described in any embodiment of this application.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the light source adjustment method for defect detection described in any embodiment of this application, or implements the defect detection method described in any embodiment of this application.

[0019] Fifthly, embodiments of this application provide a defect detection system, including a support platform, a plurality of scanning probes spaced apart along a first direction of the support platform, a detection light source corresponding to the scanning probes, and the defect detection equipment described in embodiments of this application;

[0020] The support platform is used to drive the target object to reciprocate along the second direction so as to pass through the scanning probe; or, the support platform is used to drive the scanning probe to reciprocate along the second direction so as to pass through the target object.

[0021] Each of the detection light sources is used to provide illumination when the corresponding scanning probe acquires images;

[0022] Each of the scanning probes is used to acquire images of the target object passing through in each scanning cycle and send the images to the defect detection device.

[0023] The light source adjustment method provided in the above embodiments acquires images of the target object under corresponding detection light sources collected by each scanning probe in the previous scanning cycle, determines the image parameters of each image, and determines whether the image parameters meet the requirements. For each scanning probe, if the corresponding image parameters do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters and the target value. The true value of the image parameters of the image obtained in the current scanning cycle and the fitted value of the image parameters based on the fitting function are determined. The method then returns to the step of determining whether the image parameters meet the requirements. If the corresponding image parameters meet the requirements, the method adjusts the light source parameters according to the previous scanning cycle. By using the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained during the period, the target light source parameters of the corresponding detection light source are determined. Thus, by establishing a fitting function, the detection light sources corresponding to each of the multiple scanning probes are automatically adjusted under the first scanning cycle. This allows for the automatic adjustment of the detection light sources used by each scanning probe to the optimal matching state for various practical application scenarios of defect detection, such as different types of products or the same type of product at different process stages. This ensures that the quality of the acquired images meets the requirements for defect detection analysis, improves the accuracy of product defect detection, and enhances the consistency of defect detection capabilities among multiple scanning probes.

[0024] In the above embodiments, the defect detection method, defect detection equipment, computer-readable storage medium, and defect detection system each include corresponding light source adjustment method embodiments, thereby having the same technical effects as the corresponding light source adjustment method embodiments, which will not be described again here. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating possible application scenarios for light source adjustment methods.

[0026] Figure 2 This is a schematic diagram illustrating another possible application scenario for the light source adjustment method.

[0027] Figure 3 This is a schematic diagram illustrating another possible application scenario for light source adjustment methods.

[0028] Figure 4 This is a flowchart of a light source adjustment method in one embodiment;

[0029] Figure 5 This is a schematic diagram of multiple sets of data samples in one embodiment;

[0030] Figure 6 This is a schematic diagram of the product under test arranged on a carrier panel in one embodiment;

[0031] Figure 7 This is a schematic diagram of an embodiment where one scan cycle is used;

[0032] Figure 8 This is a schematic diagram illustrating a scanning cycle based on different scanning stages within a single scan in another embodiment;

[0033] Figure 9 This is a flowchart of a defect detection method in one embodiment;

[0034] Figure 10 A flowchart of a defect detection method provided as an optional specific example;

[0035] Figure 11 This is a schematic diagram of the defect detection device in one embodiment;

[0036] Figure 12 This is a schematic diagram of the structure of a defect detection system in one embodiment. Detailed Implementation

[0037] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to limit the ways in which the invention may be implemented. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0039] In the description of this invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0040] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0041] Please see Figure 1This diagram illustrates an optional application scenario of a light source adjustment method provided in this application embodiment. The light source adjustment method is applied to a defect detection system for detecting product defects. The defect detection system includes a support platform 11, a scanning device 12 mounted on the support platform 11, and a defect detection device 13. The scanning device 12 includes a scanning axis arranged along the width direction of the support platform 11, multiple scanning probes 121 spaced apart on the scanning axis, and multiple detection light sources 122 corresponding to each scanning probe 121. The products to be inspected are arranged in rows on a carrier panel, and the support platform 11 drives the carrier panel to move along its length. When the carrier panel carrying the products passes the position of the scanning probes 121, each detection light source 122 provides illumination to each row of products on the carrier panel. Each scanning probe 121 acquires an image of its corresponding product and sends it to the defect detection device 13. The defect detection device 13 analyzes and detects the presence of various defects in the products based on the product images using image processing algorithms. The image processing algorithm can be an AI-based algorithm, which trains an AI model based on sample images containing specified defect categories. The trained AI model then identifies whether the product images acquired by each scanning probe 121 contain corresponding defects. Alternatively, the image processing algorithm can employ known traditional image processing algorithms, such as binary image segmentation algorithms, to identify whether the product images acquired by each scanning probe 121 contain corresponding defects. Before the defect detection device 13 analyzes and detects the presence of various defects in the product based on the product images, it executes the light source adjustment method provided in this embodiment to adjust the light source parameters of the detection light source 122 corresponding to each scanning probe 121 until the image parameters of the product images acquired by each scanning probe 121 meet the requirements. This prevents the accuracy of the product defect detection results from being affected by substandard image quality of the product images captured by different scanning probes 121 under the same illumination conditions, even under uniform acquisition conditions.

[0042] The defect detection device 13 can be an independent device physically separate from the support platform 11 and the scanning device 12. For example, it can be a smart electronic device such as a mobile terminal or personal computer loaded with a computer program implementing the light source adjustment method in this embodiment or a computer program implementing the defect detection method in this embodiment. Optionally, the defect detection device 13 can also be a device integrated with the support platform 11, such as... Figure 2 As shown, the defect detection device 13 can also be a console integrated on the support platform 11, providing a user-interactive and visual display interface. The console is loaded with a computer program implementing the light source adjustment method in the embodiments of this application or a computer program implementing the defect detection method in the embodiments of this application; such as Figure 3 As shown, the defect detection device 13 can also be a device integrated with the scanning device 12. The scanning device 12 is a computer program capable of executing the light source adjustment method in the embodiments of this application, or a computer program capable of executing the defect detection method in the embodiments of this application.

[0043] Please see Figure 4 This application provides a light source adjustment method for defect detection, which can be applied to... Figures 1 to 3 The defect detection equipment described herein includes the following steps:

[0044] S101, acquire the images of the target object collected by each scanning probe in the previous scanning cycle under the corresponding detection light source, and determine the image parameters of each image.

[0045] The target object can be any preset product to be inspected, such as a panel (Glass). A scanning cycle can be a preset fixed duration, or the duration required for each scanning probe to complete one or more image acquisition operations on the target object. Acquiring the images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source can be done after each scanning cycle ends and before the start of the next scanning cycle, acquiring the images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source. The image parameters can be specified types of parameters characterizing image quality, such as brightness, grayscale, and saturation. In this embodiment, the image parameter refers to grayscale.

[0046] S102, determine whether the image parameters meet the requirements.

[0047] Whether the image parameters meet the requirements can indicate whether the quality of the corresponding image meets the needs of defect detection and analysis based on the image. Taking grayscale as an example, the grayscale value of the target object image acquired by each scanning probe reaches the expected grayscale value as a condition for judging whether the image parameters meet the requirements. In this embodiment, when the grayscale of the image reaches the expected grayscale value, it can be considered that the brightness, clarity, etc. of the corresponding image can meet the needs of subsequent defect detection and analysis.

[0048] S103, for each scanning probe, if the corresponding image parameters do not meet the requirements, adjust the light source parameters of the detection light source in the current scanning cycle according to the difference between the image parameters and the target value, and determine the true value of the image parameters of the image obtained in the current scanning cycle and the image parameter fitting value of the image based on the fitting function, and return to the step of determining whether the image parameters meet the requirements.

[0049] Here, the target value refers to a preset image parameter value or a range of image parameter values. Optionally, determining whether the image parameters of each image meet the requirements can be done by comparing the image parameters of each image with the target value; if the image parameters of the corresponding image are lower than the target value, the image parameters of the corresponding image can be considered as not meeting the requirements. Optionally, a threshold image parameter value smaller than the target value can be determined based on the target value, and the image parameters of each image can be compared with the threshold to determine whether the image parameters of the corresponding image meet the requirements. Each scanning probe acquires images of the corresponding target object under the illumination conditions provided by the corresponding detection light source. For each scanning probe, the image acquired by the scanning probe is judged. When the image parameters of the image do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters of the corresponding image and the target value. Here, the light source parameters of the detection light source refer to the light source characteristics that can change the illumination conditions provided by the detection light source to the target object, thereby changing the specified type of image parameters of the image acquired under the corresponding illumination conditions. For example, if the image parameter is grayscale, the light source parameter of the detection light source refers to light intensity. The light source parameters of the detection light source corresponding to the image are adjusted so that the detection light source can provide improved illumination conditions, enabling the scanning probe to acquire an image of the target object whose image parameters meet the requirements under the improved illumination conditions provided by the corresponding detection light source.

[0050] After adjusting the light source parameters of the detection light source, each scanning probe performs image acquisition in the current scanning cycle, obtaining images of the target object under the corresponding detection light source acquired by each scanning probe in the current scanning cycle. Returning to step S102, it is determined whether the image parameters of the images acquired by each scanning probe in the current scanning cycle meet the requirements. Simultaneously, the image parameters of each image obtained in the current scanning cycle are determined. For ease of description and differentiation, the image parameters of each image obtained by each scanning probe in the current scanning cycle are referred to as the true values ​​of the image parameters. The fitting function refers to a functional relationship established based on the correspondence between the image parameters and the light source parameters, with the image parameters as the independent variable and the light source parameters as the dependent variable. Based on the true values ​​of the image parameters of each image obtained by each scanning probe in the current scanning cycle, the fitting function is substituted to determine the fitted values ​​of the image parameters for each image.

[0051] Based on the images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source, the image quality is judged to meet the requirements. If the image quality does not meet the requirements, the light source parameters of the corresponding detection light source are adjusted. Then, the images of the target object acquired by each scanning probe in the current scanning cycle are acquired after adjusting the light source parameters of each detection light source. The process returns to S102, determining whether the image parameters meet the requirements. By returning to the step of determining whether the image parameters meet the requirements, for images whose image parameters do not meet the requirements, the light source parameters of the detection light source are adjusted again in the new current scanning cycle based on the difference between the image parameters of the corresponding image and the target value. This allows the detection light source to provide improved illumination conditions, enabling the scanning probe to acquire images of the target object with satisfactory image parameters under the improved illumination conditions provided by the corresponding detection light source.

[0052] S104, if the corresponding image parameters meet the requirements, determine the target light source parameters corresponding to the detection light source based on the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained in the previous scanning cycle.

[0053] For each scanning cycle, the true values ​​of the image parameters of each image obtained within the corresponding scanning cycle can be determined, and the fitted values ​​of the image parameters of each image can be determined based on the fitting function and the true values ​​of the image parameters.

[0054] In each scanning cycle, images acquired by each scanning probe are obtained. Steps S102 to S104 are then executed to adjust each detection light source. This adjustment may require multiple scanning cycles. The real-time values ​​of the light source parameters obtained after adjustment in each scanning cycle are used as reference values ​​for the light source parameters in the next scanning cycle, until the images of the target object acquired by each scanning probe in a given scanning cycle under the corresponding detection light source meet the requirements. At this point, the target light source parameters corresponding to the detection light source are determined based on the true values ​​and fitted values ​​of the image parameters obtained in the preceding scanning cycles. The preceding scanning cycles typically refer to the two or more preceding scanning cycles determined with reference to the scan cycle in which the images of the target object acquired by each scanning probe in a given scanning cycle under the corresponding detection light source meet the requirements.

[0055] For ease of understanding, taking a scanning probe including probe 1, probe 2, and probe 3, and a corresponding detection light source including light source 1, light source 2, and light source 3 as an example, light source 1, light source 2, and light source 3 provide illumination to target object a, target object b, and target object c, respectively. Probe 1, probe 2, and probe 3 acquire images a1, image b1, and image c1 corresponding to target object a, target object b, and target object c, respectively. If the image parameters of image a1 do not meet the requirements, the light source parameters of light source 1 are adjusted according to the difference between the image parameters of image a1 and the target value. At the same time, if the image parameters of image b1 meet the requirements, there is no need to adjust the light source parameters of light source 2; however, if the image parameters of image b1 do not meet the requirements, the light source parameters of light source 2 are adjusted according to the difference between the image parameters of image b1 and the target value. Similarly, if the image parameters of image c1 meet the requirements, there is no need to adjust the light source parameters of light source 3; however, if the image parameters of image c1 do not meet the requirements, the light source parameters of light source 3 are adjusted according to the difference between the image parameters of image c1 and the target value.

[0056] Taking the images a1, a2, and a3 acquired by probes 1, 2, and 3 respectively in the first scanning cycle as an example, where the image parameters of image a1 do not meet the requirements, while the image parameters of images b1 and c1 both meet the requirements, after adjusting the light source parameters of light source 1 based on the difference between the image parameters of image a1 and the target value, and considering that the image parameters of images b2 and c2 meet the requirements, there is no need to adjust the light source parameters of light sources 2 and 3. Entering the second scanning cycle, images a2, b2, and c2 acquired by probes 1, 2, and 3 in the second scanning cycle are acquired again, and it is determined again whether the image parameters of each image a2, b2, and c2 meet the requirements. If the image parameters of image a2 still do not meet the requirements, then the light source parameters of light source 1 are adjusted again based on the difference between the image parameters of image a2 and the target value; similarly, since the image parameters of images b2 and c2 meet the requirements, there is no need to adjust the light source parameters of light sources 2 and 3. In the third scanning cycle, images a3, b3, and c3 acquired by probes 1, 2, and 3 during the second scanning cycle are acquired again. It is then determined whether the image parameters of each image a3, b3, and c3 meet the requirements. If the image parameters of images a3, b3, and c3 all meet the requirements, the target light source parameters for each light source are determined based on the actual and fitted values ​​of the image parameters obtained in the first scanning cycle, and the actual and fitted values ​​of the image parameters obtained in the second scanning cycle. For example, for light source 1 whose light source parameters have been adjusted in the first and second scanning cycles, a first conversion coefficient is obtained based on the ratio of the fitted value to the actual value of the image parameters in image a1 during the first scanning cycle; a second conversion coefficient is obtained based on the ratio of the fitted value to the actual value of the image parameters in image a2 during the second scanning cycle; and the target fitted value of the image parameters for image a1 is calculated based on the first and second conversion coefficients and the target value of the image parameters. Based on the correspondence between image parameters and light source parameters, the target light source parameters for light source 1 can be determined based on the target fitted value of the image parameters.

[0057] Optionally, the target light source parameters corresponding to the light source 1 can be determined based on the target fitting value of the image parameters according to the correspondence between the image parameters and the light source parameters. Alternatively, the target light source parameters can be determined by inverting the fitting function based on the target fitting value of the image parameters and the fitting function.

[0058] In the above embodiments, by acquiring images of the target object under the corresponding detection light source collected by each scanning probe in the previous scanning cycle, determining the image parameters of each image, and judging whether the image parameters meet the requirements, for each scanning probe, if the corresponding image parameters do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters and the target value, and the true value of the image parameters of the image obtained in the current scanning cycle and the fitted value of the image parameters of the image determined based on the fitting function are determined, and the step of judging whether the image parameters meet the requirements is returned; if the corresponding image parameters meet the requirements, according to the prior... By combining the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained within the scanning cycle, the target light source parameters of the corresponding detection light source are determined. Thus, by establishing a fitting function, the detection light sources corresponding to each of the multiple scanning probes are automatically adjusted during the scanning cycle. This allows for optimal matching of the detection light sources used by each scanning probe when acquiring corresponding images in various practical application scenarios, such as different types of products or different process stages of the same type of product. This ensures that the acquired image quality meets the requirements for defect detection analysis, improving the accuracy of product defect detection and the consistency of defect detection capabilities among multiple scanning probes.

[0059] In some embodiments, before determining the image parameter fitting values ​​of the image based on the fitting function, the light source adjustment method further includes:

[0060] Establish a polynomial function with the light source parameters of the detected light source as independent variables and the image parameters as dependent variables;

[0061] For each of the scanning probes, a preset number of data samples are collected; each data sample represents the correspondence between the light source parameter values ​​and the image parameter values ​​of the detection light source.

[0062] Based on the data sample, the polynomial coefficients of the polynomial function are calculated, and the fitting function is obtained based on the calculation results.

[0063] By collecting sufficient data samples, a polynomial function is fitted using multiple sets of data samples, and the polynomial coefficients of the polynomial function are calculated to obtain the fitted function. Data sample collection can involve setting initial values ​​and single adjustment step sizes for the light source parameters of each detection light source, and then sequentially acquiring images of the light source parameters at the initial values, and then at each light source parameter value with progressively increasing initial values ​​and single adjustment step sizes, thus obtaining the image parameter values ​​corresponding to each light source parameter value. Please refer to [link / reference]. Figure 5The light source parameter is light intensity, and the image parameter is grayscale. An initial value for the light intensity and an adjustment step size are set, and the grayscale feedback corresponding to each light intensity is obtained sequentially, establishing a light intensity-grayscale correspondence table. In the correspondence table, each light intensity value and its corresponding grayscale value form a data sample.

[0064] The polynomial function with the light source parameters as independent variables and the image parameters as dependent variables can be represented by the following formula:

[0065]

[0066] Where M is the order of the polynomial function, ω0, ω1, ω2…ωM are the coefficients of the polynomial function, denoted as W. The independent variable x can represent the light source parameters, and the dependent variable y can represent the image parameters. The polynomial function y(x, W) is established as a nonlinear function of x, and also a linear function of the polynomial coefficients W. By iteratively fitting the polynomial function using multiple sets of collected data samples, the coefficients of the polynomial function are obtained, thus obtaining the curve corresponding to the light intensity (actual light output of the light source) of the light source and the grayscale of the image collected under the corresponding lighting conditions, that is, obtaining the fitting function.

[0067] In the above embodiments, based on the characteristic that there is a non-linear relationship between the light intensity level of a light source and the actual light output of the light source under normal circumstances, by collecting data samples and fitting them, the non-linear relationship between the light source parameters and the image parameters of the image acquired under the corresponding lighting conditions is determined. A fitting function is established to characterize the correspondence between the light source parameters and the image parameters. The fitted value of the image is calculated through the fitting function to calculate the final target light source parameters, which can improve the convergence speed of adjusting the light source parameters.

[0068] In some embodiments, before determining the image parameter fitting values ​​of the image based on the fitting function, the light source adjustment method further includes:

[0069] Establish a linear function with the light source parameters of the detected light source as the independent variable and the image parameters as the dependent variable;

[0070] For each of the scanning probes, a preset number of data samples are collected; each data sample represents the correspondence between the light source parameter values ​​and the image parameter values ​​of the detection light source.

[0071] Based on the data sample, the proportional coefficient of the linear function is calculated, and the fitting function is obtained based on the calculation result.

[0072] Linear functions are simpler than polynomial functions, allowing for faster determination of scaling factors based on a smaller amount of data. Data collection can be achieved by setting initial values ​​and single-step adjustment increments for the light source parameters of each detection light source, and then sequentially acquiring images of each light source parameter at its initial value, and then at each parameter value that is progressively increased with the initial value and the single-step adjustment increment. This yields the image parameter values ​​corresponding to each light source parameter value. For example, using light intensity as the light source parameter and grayscale as the image parameter, each light intensity value and its corresponding grayscale value constitutes a data sample.

[0073] In the above embodiments, by establishing a linear function and substituting multiple sets of collected data samples into the linear function, the parameters of each item in the linear function are determined, and a fitting function that can be used to characterize the correspondence between light source parameters and image parameters is obtained. The fitting value of the image is calculated through the fitting function to calculate the final target light source parameters, which can improve the convergence speed of adjusting the light source parameters.

[0074] In some embodiments, obtaining the fitting function based on the calculation result includes:

[0075] Based on the data sample, the deviation between the fitted value obtained by the fitting function and the estimated value is calculated based on a preset error function.

[0076] The fitting function is evaluated based on the deviation.

[0077] If the evaluation fails, return to the step of collecting a preset number of data samples for each of the scanning probes;

[0078] If the evaluation passes, the fitted function is determined to be the final fitted function.

[0079] In the process of designing a fitting function to adjust the light source parameters of the detection light source, an optimal light source parameter value (estimated value) is usually predicted to predict the fitted value obtained by the fitting function. However, an error inevitably exists between the estimated value and the fitted value. Introducing an error function allows for the calculation of the deviation between the fitted value and the estimated value, thereby evaluating whether the fitted value obtained by the fitting function meets the conditions and avoiding overfitting or underfitting. Evaluating the fitted polynomial function by designing an error function can achieve a reasonable balance between the accuracy and efficiency of the fitting result. In an optional embodiment, the error function uses the mean-square error (MSE), as shown in Formula 2 below:

[0080]

[0081] Where N refers to the number of data samples, tn is the estimated value, and the mean squared error E(W) is a measure of the difference between the estimated value (the fitted value obtained according to the fitting function) and the estimated value.

[0082] In the above embodiments, an error function is introduced to evaluate the fitting function, and the polynomial coefficients in the fitting function are iterated based on whether the evaluation passes, so as to ensure that the result of the polynomial fitting can be closer to the actual situation.

[0083] In some embodiments, step S103 involves adjusting the light source parameters corresponding to the detection light source within the current scanning cycle, and determining the true values ​​of the image parameters of the image obtained within the current scanning cycle, as well as determining the fitted values ​​of the image parameters based on the fitting function, including:

[0084] Based on the difference between the image parameters and the target value, the light source parameters of the detection light source are adjusted according to the current light source parameters and parameter adjustment step size of the detection light source.

[0085] The images of the target object acquired by each of the scanning probes in the current scanning cycle under the adjusted detection light source are obtained. The true values ​​of the image parameters of the images obtained in the current scanning cycle are determined, and the fitted values ​​of the image parameters of the images are determined based on the fitting function.

[0086] The parameter adjustment step size is the magnitude of one adjustment to the detection light source. When the image quality acquired by each scanning probe fails to meet the requirements, the current light source parameters of the detection light source are adjusted based on the difference between the image parameters and the target value, using the parameter adjustment step size. After the light source parameters of the detection light source are adjusted, a new scanning cycle begins, and the image acquired by the scanning probe is acquired again. The quality of the re-acquired image is then judged. This process is repeated for multiple scanning cycles until the image quality acquired in the new scanning cycle meets the requirements. For each preceding scanning cycle, the true values ​​of the image parameters acquired in each scanning cycle are determined, and the fitted values ​​of the image parameters corresponding to the image are determined based on the fitting function. Optionally, for each preceding scanning cycle, before acquiring the images of the target object under the corresponding detection light source in the first scanning cycle, the following steps are included: initializing the light source parameters of the detection light source to set the light source parameters of each detection light source as initial values. In each of the first scanning cycles, by uniformly setting the light source parameters of the detection light sources to initial values ​​and then uniformly adjusting them according to the parameter adjustment step size based on the initial values, the selection of reasonable initial values ​​facilitates a faster achievement of the adjustment target.

[0087] In the above embodiments, during each scanning cycle, based on the deviation between the image quality acquired in the previous scanning cycle and the expected quality, the light source parameters of the detection light source are adjusted. Each adjustment increment or decrement is made by a preset parameter adjustment step. Taking light intensity as the light source parameter and grayscale as the image parameter, based on the linear relationship between light intensity and grayscale feedback, the amount of grayscale change caused by a single light intensity adjustment can be estimated to set the light intensity adjustment step, achieving the effect of gradually approaching the target value within fewer scanning cycles. After successful adjustment, the target adjustment slope is calculated for subsequent adjustments to the light source parameters of the detection light source, which can improve the convergence speed of subsequent adjustments and enhance the accuracy of the adjustment results.

[0088] Optionally, adjusting the light source parameters of the detection light source based on the difference between the image parameters and the target value, and based on the current light source parameters and parameter adjustment step size of the detection light source, includes:

[0089] If the image parameters are less than the target value, adjust the step size according to the current light source parameters and parameters of the corresponding detection light source, and increase the light source parameters of the detection light source;

[0090] If the image parameters are greater than the target value, the step size is adjusted according to the current light source parameters and parameters of the corresponding detection light source to reduce the light source parameters of the detection light source.

[0091] In this embodiment, an adjustment strategy for a single adjustment of the detection light source can be preset. By presetting the parameter adjustment step size of the light source parameters, when the image parameter is less than the target value, the parameter adjustment step size is increased based on the current light source parameters of the detection light source, and the light source parameter value increases, so that the image parameter value of the image acquired after adjustment can increase; when the image parameter is greater than the target value, the parameter adjustment step size is decreased based on the current light source parameters of the detection light source, and the light source parameter value decreases, so that the image parameter value of the image acquired after adjustment can decrease.

[0092] Taking grayscale as the image parameter and light intensity as the light source parameter as an example, if the grayscale value of the image is less than the target grayscale value, the current light intensity of the corresponding detection light source is increased by the light intensity adjustment step. By increasing the light intensity, the grayscale value of the re-acquired image can be closer to the target grayscale value. If the grayscale value of the image is greater than the target grayscale value, the current light intensity of the corresponding detection light source is decreased by the light intensity adjustment step. By decreasing the light intensity, the grayscale value of the re-acquired image can be closer to the target grayscale value.

[0093] In the above embodiments, by changing the light source parameters, the illumination conditions provided by the detection light source to the target object are correspondingly altered, as are the image parameters of the target object image acquired under the corresponding illumination conditions. A judgment is made based on whether the images acquired by each scanning probe in the previous scanning cycle meet the requirements. Based on the judgment result, the light source parameters of the detection light source are adjusted so that the image parameter values ​​of the target object image acquired under the illumination conditions provided by the detection light source are closer to the target values. Thus, after adjustment through one or more scanning cycles, the scanning probes, the detection light source, and the target object can automatically reach an optimal matching state, and the quality of the images acquired by each scanning probe for the target object under the illumination conditions provided by its corresponding detection light source can meet expectations.

[0094] In some embodiments, in step S105, determining the target light source parameters corresponding to the detection light source based on the true values ​​of the image parameters and the fitted values ​​of the image parameters obtained in the previous scanning cycle includes:

[0095] For each scanning cycle, a conversion coefficient is determined based on the ratio of the fitted image parameter value to the image parameter value of the image obtained within the scanning cycle;

[0096] Calculate the target conversion coefficient based on the conversion coefficients corresponding to the two preceding scan cycles;

[0097] The target fitting value of the image parameters is determined by converting the target conversion coefficient, and the target light source parameters corresponding to the detection light source are calculated based on the target fitting value and the fitting function.

[0098] During the light source adjustment process, conversion coefficients are determined for each scanning cycle based on the images acquired by each scanning probe in the previous few scanning cycles. The conversion coefficient is an intermediate parameter used to calculate the target light source parameters for each detection light source. For each detection light source, a conversion coefficient is calculated for each scanning cycle based on the ratio of the fitted value of the image parameters obtained from the fitting function to the true value. The target conversion coefficient is a predicted target value of the conversion coefficient calculated based on the conversion coefficients of the previous multiple scanning cycles. In an optional specific embodiment, the image parameter refers to grayscale, and the light source parameter refers to light intensity. The light intensity adjustment step ΔI and the target grayscale value acceptance range for each detection light source are predetermined. First, the light intensity of each detection light source is set as an initial value. During the first scanning cycle, images are acquired by each scanning probe under the illumination condition of the corresponding detection light source at the initial value I0. Based on the comparison results of the grayscale of each image with the target grayscale value, and referring to Table 1: if the grayscale value G0 of image 1 acquired by probe 1 under the illumination condition of light source 1 at the initial value I0 is less than the target grayscale value, then the light intensity adjustment step ΔI is added to light source 1 based on the initial value I0 as the first adjustment value. After the first adjustment of the light intensity of light source 1, the light intensity of light source 1 is the first adjustment value. During the second scanning cycle, images are acquired by each scanning probe under the illumination conditions provided by the corresponding detection light source. Based on the comparison results of the grayscale of each image with the target grayscale value, for example: if the grayscale G1 of image 1 acquired by probe 1 under the illumination condition of light source 1 at the first adjustment value is still less than the target grayscale value, then the light intensity adjustment step ΔI is added to light source 1 based on the first adjustment value as the second adjustment value. After the light intensity value of light source 1 is adjusted for the second time, the light intensity of light source 1 is the second adjustment value. Similarly, in the third scanning cycle, the images acquired by each scanning probe under the illumination conditions provided by the corresponding detection light source are acquired again. Based on the comparison results of the grayscale of each image with the target grayscale value, it is determined whether to adjust the light intensity of each detection light source. At this time, the image parameter values ​​of the target object acquired by each scanning probe under the corresponding illumination conditions of each detection light source approach the target value, that is, the image quality of the target object acquired at this time can meet the requirements. Based on the ratio of the image parameter fitting value G0Y of image 1 to the true image parameter value G0 in the first scanning cycle, the first conversion coefficient Ratio1 is calculated; based on the ratio of the image parameter fitting value G1Y of image 1 to the true image parameter value G1 in the second scanning cycle, the second conversion coefficient Ratio2 is calculated; based on the product of the average of the first conversion coefficient and the second conversion coefficient and the target value, the target fitting value GTargetY is obtained; based on the fitting function established based on the correspondence between light intensity and grayscale, the target light source parameters of light source 1 are determined based on the target fitting value.

[0099] Table 1:

[0100]

[0101] It should be noted that in the above embodiment, after the light source parameters of light source 1 are adjusted twice, in the third scanning cycle, the difference between the gray value of the image collected by probe 1 under the illumination of light source 1 and the target gray value is within the target gray value acceptance range, and the adjustment of light source 1 is successful; conversely, if the difference between the gray value of the image collected by probe 1 under the illumination of light source 1 and the target gray value still exceeds the target gray value acceptance range, then light source 1 needs to be adjusted according to the light intensity adjustment step until it meets the requirements.

[0102] In the above embodiments, after adjustment through one or more scanning cycles, the target light source parameters of each detection light source are determined, and each scanning probe, detection light source and target object can automatically reach the best matching state. Under the illumination conditions provided by the corresponding detection light source, the quality of the images acquired by each scanning probe for the target object can meet expectations.

[0103] In some embodiments, determining whether the image parameters meet the requirements includes:

[0104] If the image parameters are within the target acceptance range, then the image parameters meet the requirements;

[0105] If the image parameters are outside the target acceptance range, then the image parameters do not meet the requirements.

[0106] The target acceptance interval refers to the range within which the values ​​of the corresponding image parameters are allowed to vary. Optionally, different target acceptance intervals can be set for different detection light sources. For example, multiple scanning probes can be used to acquire images of the product to be inspected. During the process of detecting and analyzing defects of a specified type on the product based on these images, the target values ​​of the image parameters can be determined according to the image quality requirements required to improve the accuracy of defect detection. The target acceptance intervals of the corresponding image parameters are then set based on these target values. It should be noted that the product to be inspected can be a different category of product corresponding to different scanning probes, a product of the same category but at a different process stage, or a product of the same category from the same batch. For different categories of products, or products of the same category but at different process stages, the image quality requirements relied upon during defect detection and analysis based on the images of these products are often different. Thus, the target values ​​and target acceptance intervals of the image parameters for different images can be different.

[0107] In this embodiment, the image parameter refers to the grayscale value of the image. By acquiring the product image collected by each scanning probe under the illumination conditions provided by the corresponding detection light source, it is determined whether the grayscale value of the product image is within the target grayscale value acceptance range, thereby determining whether the quality of the product image meets the image quality requirements required for defect detection and analysis of the product. Accordingly, the light source parameters of each detection light source are adjusted so that the grayscale value of the product image collected by each scanning probe under the illumination intensity provided by the corresponding detection light source can be close to the target grayscale value.

[0108] In some embodiments, acquiring the images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source includes:

[0109] When each scanning probe completes scanning the target object from the first side to the second side along the first scanning direction, it acquires the image of the target object under the corresponding detection light source, which was collected by each scanning probe along the first scanning direction.

[0110] After determining whether the image parameters meet the requirements, for each scanning probe, if the corresponding image parameters do not meet the requirements, the method includes:

[0111] When each scanning probe completes scanning the target object from the second side to the first side along the second scanning direction, it acquires images of the target object under the corresponding detection light source, collected by each scanning probe along the second scanning direction.

[0112] One scan cycle refers to the time required for the target object to pass through the position of the scanning probe once along the first scanning direction in one scan cycle. Please refer to [link / reference]. Figure 6 The target objects include multiple electronic device screens, such as mobile phone screens and tablet screens, arranged in rows and columns on a panel carrier. Please refer to [link / reference]. Figure 7The scanning probes are mounted above the panel carrier, with multiple probes arranged in pairs along the width of the panel carrier. The interval (trips) between two adjacent probes is equal to the width of the mounting positions of multiple rows of products on the panel carrier. The panel carrier moves from right to left. As the scanning probes sequentially pass the positions of the rows of products directly opposite them, they scan and acquire product images. When the panel carrier moves from the first position below the scanning probe on the right to the second position below the scanning probe on the left, it indicates that the target object has passed the position of the scanning probe once along the first scanning direction, thus completing one scanning cycle. At this time, the light source adjustment method provided in this application is used to acquire the images acquired by each scanning probe in the previous scanning cycle. The quality of each image is judged to meet the requirements, and the light source parameters of the detection light source are adjusted accordingly before entering the next scanning cycle. The panel carrier moves from left to right. When the panel carrier moves from the second position below the scanning probe on the left to the first position below the scanning probe on the right, it indicates that the target object has passed the position of the scanning probe once along the second scanning direction, thus completing another scanning cycle. At this time, the light source adjustment method provided in this application provides to acquire images collected by each scanning probe, determine whether the quality of each image meets the requirements, and adjust the light source parameters of the detection light source accordingly before entering the next scanning cycle; or, if the quality of each image meets the requirements, the light source parameters of each detection light source are kept unchanged before entering the next scanning cycle.

[0113] In this defect detection system, the light source adjustment method provided in the embodiments of this application addresses the challenge of determining a suitable light intensity value for a scanning probe to capture a sufficiently clear image when a brand-new panel product first enters the machine. This is because the reflectivity of the panel product under the current optical environment of the machine is unknown. Furthermore, the involvement of multiple scanning probes and their inherent hardware differences—such as variations in light source intensity, fiber optic transmission, performance differences in optical lens groups, and CCD camera chip responsivity—all contribute to differences in grayscale imaging. Therefore, based on these factors, it is impossible to manually specify a suitable light source intensity to ensure compatibility with multiple scanning probes.

[0114] In this embodiment, each scanning probe scans the target object from one side to the other along the same scanning direction, and the image acquisition is considered as one scanning cycle. The interval between two scanning cycles is used as an opportunity to adjust the light source parameters of the detection light source. This not only makes full use of the idle time in the defect detection system and is compatible with the original defect detection process of completing defect detection by acquiring product images, but also intelligently adjusts each detection light source at the end of each scanning cycle to improve the quality of the product images acquired in the next scanning cycle, thereby gradually improving and maintaining the defect detection system's ability to detect product defects.

[0115] In some embodiments, acquiring the images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source includes:

[0116] When each scanning probe scans the target object from the first side to the second side along the first scanning direction, after completing a preset scanning stage, the images of the target object collected by each scanning probe during the preset stage under the corresponding detection light source are acquired.

[0117] After determining whether the image parameters meet the requirements, for each scanning probe, if the corresponding image parameters do not meet the requirements, the method includes:

[0118] If the previous scanning cycle is not the last scanning stage in the first scanning direction, then when each scanning probe completes the next preset stage of scanning during the scanning process of the target object from the first side to the second side along the first scanning direction, the images of the target object collected by each scanning probe in the preset stage under the corresponding detection light source are acquired.

[0119] If the previous scanning cycle is the last scanning stage in the first scanning direction, then when each scanning probe completes a preset scanning stage during the scanning process of the target object from the second side to the first side along the second scanning direction, the images of the target object collected by each scanning probe in the preset stage under the corresponding detection light source are acquired.

[0120] The scanning path of a target object moving along the same scanning direction and passing through the location of the scanning probe in one scan cycle is divided into multiple stages. One scan cycle refers to the time required for the target object to move along the same scanning direction and pass through one stage. Target objects include multiple electronic device screens, such as mobile phone screens and tablet screens, arranged in rows and columns on a panel carrier. Please refer to... Figure 8The scanning probes are mounted on top of the panel carrier. Multiple scanning probes are arranged at intervals along the width of the panel carrier. The interval between two adjacent scanning probes is equal to the width of the mounting positions of multiple rows of products on the panel carrier. The panel carrier moves from the first position below the scanning probe on the right to the second position below the scanning probe on the left as one round of scanning, or from the second position below the scanning probe on the left to the first position below the scanning probe on the right as one round of scanning. The running path of one round of scanning is divided into four stages. In each round of scanning, the scanning probe sequentially scans the row of products directly opposite it and collects the image of the corresponding row of products. Each time the target object moves through one stage, one scanning cycle is completed.

[0121] exist Figure 8 Taking scanning probe 1 as an example, the panel carrier drives the target object along the first scanning direction through the first stage. Products on the path in the first stage pass sequentially through the position of the scanning probe, thus completing one scanning cycle. At this time, the light source adjustment method provided in this application provides to acquire images collected by each scanning probe. Based on whether the quality of each image meets the requirements, the light source parameters of the detection light source are adjusted accordingly, and then the next scanning cycle begins. The panel carrier drives the target object along the first scanning direction through the second stage. Products on the path in the second stage pass sequentially through the position of the scanning probe, completing another scanning cycle. At this time, the light source adjustment method provided in this application provides to acquire images collected by each scanning probe. Based on whether the quality of each image meets the requirements, the light source parameters of the detection light source are adjusted accordingly, and then the next scanning cycle begins; or, if the quality of each image meets the requirements, the light source parameters of each detection light source remain unchanged, and then the next scanning cycle begins.

[0122] In the first scanning cycle, assuming the initial light intensity of each detection light source is I0, the first stage of scanning is completed. Taking scanning probe 1 as an example, based on the image acquired by scanning probe 1 under the illumination of light source 1 in the first stage, grayscale feedback G0 is obtained. Based on the difference between grayscale feedback G0 and the target grayscale value, the light intensity value of light source 1 is adjusted for the first time to I1, and the second scanning cycle is entered to complete the second stage of scanning. Based on the image acquired by scanning probe 1 in the second stage, grayscale feedback G1 is obtained. Similarly, based on the difference between grayscale feedback G1 and the target grayscale value, the light intensity value of light source 1 is adjusted for the second time to I2, and the third scanning cycle is entered to complete the third stage of scanning, and so on.

[0123] In this defect detection system, the light source adjustment method provided in the embodiments of this application addresses the challenge of determining a suitable light intensity value for a scanning probe to capture a sufficiently clear image when a brand-new panel product first enters the machine. This is because the reflectivity of the panel product under the current optical environment of the machine is unknown. Furthermore, the involvement of multiple scanning probes and their inherent hardware differences—such as variations in light source intensity, fiber optic transmission, performance differences in optical lens groups, and CCD camera chip responsivity—all contribute to differences in grayscale imaging. Therefore, based on these factors, it is impossible to manually specify a suitable light source intensity to ensure compatibility with multiple scanning probes.

[0124] In this embodiment, each scanning probe performs a full scan of the target object from one side to the other along the same scanning direction, divided into multiple stages. Each stage is considered a scanning cycle. The interval between two scanning cycles serves as an opportunity to adjust the light source parameters of the detection light source. This not only intelligently adjusts each detection light source at the end of each scanning cycle, improving the quality of the product image acquired in the next scanning cycle and gradually enhancing and maintaining the defect detection system's ability to detect product defects, but also accelerates the adjustment process by dividing a full scan into multiple stages and using each stage to differentiate the scanning cycle. For example, if the light source parameters of a certain light source require two adjustments to achieve optimal matching with the corresponding scanning probe, the desired result can be achieved in the third stage of a full scan.

[0125] In another aspect, this application provides a defect detection method, which, based on the light source adjustment method provided in any embodiment of this application, further performs defect detection on the target object based on the images acquired by each scanning probe.

[0126] Please see Figure 9 The defect detection method, after adjusting the light source parameters of each detection light source through a light source adjustment method until the image quality acquired by each scanning probe meets the requirements, further includes the following steps:

[0127] S105, maintain the target light source parameters of each of the detection light sources, perform image acquisition of the target object in the next scanning cycle, and perform defect analysis on the target object based on the acquired image.

[0128] Each detection light source provides illumination for each scanning probe to acquire images. Each scanning probe acquires an image of its corresponding product and sends it to the defect detection equipment. The defect detection equipment then uses image processing algorithms to analyze and detect whether the product contains various defects based on the product images. The image processing algorithm can be a known traditional image processing algorithm, such as a binary image segmentation algorithm, to identify whether the product images acquired by each scanning probe contain the corresponding defects.

[0129] In the above embodiments, before the defect detection equipment analyzes and detects whether the product has various defects based on the acquired images, it adjusts the light source parameters of the detection light source corresponding to each scanning probe by executing the light source adjustment method provided in this application embodiment until the image parameters of the product images acquired by each scanning probe meet the requirements. This is to avoid the image quality of the product images captured by different scanning probes under the same acquisition conditions, such as when each detection light source is under the same illumination conditions, failing to meet the requirements, thereby improving the accuracy of the product defect detection results.

[0130] Please see Figure 10 To provide a more comprehensive understanding of the defect detection method described in the embodiments of this application, the following example uses grayscale as the image parameter, light intensity as the light source parameter, and a panel as the target object. The defect detection method includes:

[0131] S11, Set the initial light intensity value of each detection light source;

[0132] S12, perform one round of scanning; in one round of scanning, each scanning probe acquires panel images under the illumination conditions provided by the corresponding detection light source;

[0133] S13, perform noise filtering on the panel image;

[0134] S14, determine the grayscale value of each panel image, and compare the grayscale value with the target grayscale value; for each panel image, if the grayscale value of the panel image is less than the target grayscale value, execute S15; if the grayscale value of the panel image is greater than the target grayscale value, execute S16; if the difference between the grayscale value of the panel image and the target grayscale value meets the acceptable range, execute S17.

[0135] S15, increase the light intensity value of the corresponding detection light source, return to step S12, calculate the light intensity fitting value of the image within the corresponding scanning period by combining the fitting function, and calculate the corresponding conversion coefficient based on the actual light intensity value and the light intensity fitting value; wherein, the increase in light intensity value can be based on increasing the light intensity adjustment step size on the basis of the current light intensity value;

[0136] S16, reduce the light intensity value of the corresponding detection light source, return to step S12, calculate the light intensity fitting value of the image within the corresponding scanning period by combining the fitting function, and calculate the corresponding conversion coefficient based on the actual light intensity value and the light intensity fitting value; wherein, the reduction of the light intensity value can be based on the current light intensity value by reducing the light intensity adjustment step.

[0137] S17, calculate the target fitting value based on the conversion coefficient obtained from the previous scanning cycle, calculate the target light intensity value by combining the fitting function, maintain the target light intensity value of each detection light source, and complete the scanning of the current batch of panel products.

[0138] S18, Based on the acquired panel image, perform defect analysis on the panel.

[0139] In the above embodiments, when the panel product undergoes its first defect detection, the light source intensity of multiple scanning probes can be set to the same initial value. In subsequent rounds of testing, the light source intensity of each scanning probe is dynamically adjusted based on the grayscale results of the panel image in each round, until the grayscale of the images of the panel product acquired by all scanning probes reaches the desired grayscale value. This overcomes the problem that different scanning probes may have different hardware (such as cameras, light sources, optical fibers, etc.), causing the same light source intensity to be unsuitable for multiple scanning probes, thus affecting the defect detection capability of the acquired product image quality. Similarly, the problem that different panel product types or different process stages may also cause the same light source intensity to be unsuitable for multiple scanning probes, thus affecting the defect detection capability of the acquired product image quality. Through automatic adjustment of the light source intensity, a suitable light source intensity value can be quickly and accurately found for each scanning probe, ensuring that the image quality of the acquired panel product meets the requirements for defect detection analysis.

[0140] In another aspect of the embodiments of this application, please refer to Figure 11 This is a schematic diagram of an optional hardware structure of a defect detection device provided in an embodiment of this application. The defect detection device includes a processor 111 and a memory 112 connected to the processor 111. The memory 112 is used to store various types of data to support the operation of the defect detection device, and stores a computer program that can be executed by the processor. When the computer program is executed by the processor 111, it implements the steps of the defect detection method described in any embodiment of this application and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0141] In another aspect of the embodiments of this application, please refer to Figure 12A defect detection system is provided, including a support platform 11, a plurality of scanning probes 121 spaced apart along a first direction of the support platform 11, detection light sources 122 corresponding to the scanning probes 121, and the defect detection device described in the embodiments of this application. The support platform 11 is used to drive the target object to reciprocate along a second direction to pass through the scanning probes 121. Optionally, the support platform 11 is used to drive the scanning probes 121 to reciprocate along the second direction to pass through the target object. Each detection light source 122 is used to provide illumination for the corresponding scanning probe 121 to acquire images. Each scanning probe 121 is used to acquire images of the target object passing through in each scanning cycle and send the images to the defect detection device.

[0142] Optionally, the carrier platform 11 is provided with a carrier panel 113, and the target objects are arranged in rows and columns on the carrier panel 113; the interval between two adjacent scanning probes 121 is equal to the row spacing of the target object row on the carrier panel 113; or, the interval between two adjacent scanning probes 121 is equal to a multiple of the row spacing of the target object row on the carrier panel 113.

[0143] by Figure 12 Taking the orientation shown as an example, the right side is the front end of the support platform 11, and the left side is the rear end of the support platform 11. The target object is supported on the carrier panel 11 and can be sent in by the robot arm from the front end of the support platform 11 and placed on the support platform 11. The support platform 11 can drive the carrier panel 113 to move the target object. The target object reciprocates along the front-back direction (second direction) of the support platform 11. The direction of travel of the target object is defined as the length direction of the support platform 11. The direction orthogonal to the length direction of the support platform 11 on the horizontal plane is defined as the width direction (first direction) of the support platform 11. The scanning axis 114, which is spaced apart from the support platform 11 on the vertical plane, carries n scanning probes 121. The number and spacing of the scanning probes 121 are equal, and they are evenly distributed on the scanning axis 114. The spacing is related to the field of view of each scanning probe 121 and the number of target objects, so as to ensure that the scanning probes 121 can be responsible for scanning the entire area of ​​the target object on the carrier panel 113 through a set number of scanning cycles, collecting images of the target object for defect identification, and finally identifying and determining all defects of the target object.

[0144] In each scanning cycle, the scanning probe 121 acquires an image of the target object that passes through its location. This can be achieved by the platform 11 driving the target object to reciprocate along the second direction so that the target object passes through the location of the scanning probe 121; or by the platform 11 driving the scanning probe 121 to reciprocate along the second direction so that the target object passes through the location of the scanning probe 121.

[0145] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described image processing method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0146] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0147] 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) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0148] As will be understood by those skilled in the art, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments of this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0149] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for adjusting a light source for defect detection, characterized in that, include: The images of the target object acquired by each scanning probe in the previous scanning cycle under the corresponding detection light source are obtained, and the image parameters of each image are determined; wherein, the image parameters are specified type parameters characterizing image quality; Determine whether the image parameters meet the requirements; For each scanning probe, if the corresponding image parameters do not meet the requirements, the light source parameters of the corresponding detection light source in the current scanning cycle are adjusted according to the difference between the image parameters and the target value. The true value of the image parameters of the image obtained in the current scanning cycle and the fitted value of the image parameters of the image are determined based on the fitting function. Then, the step of determining whether the image parameters meet the requirements is returned. If the corresponding image parameters meet the requirements, the target light source parameters corresponding to the detection light source are determined based on the true values ​​and fitted values ​​of the image parameters obtained in the previous scanning cycle. Determining the target light source parameters based on the true values ​​and fitted values ​​of the image parameters obtained in the previous scanning cycle includes: for each scanning cycle, determining a conversion coefficient based on the ratio of the fitted value to the image parameter value of the image obtained in the scanning cycle; calculating a target conversion coefficient based on the conversion coefficients corresponding to the two previous scanning cycles; converting and determining a target fitted value of the image parameters based on the target conversion coefficient; and calculating the target light source parameters corresponding to the detection light source based on the target fitted value and the fitting function.

2. The light source adjustment method as described in claim 1, characterized in that, Before determining the image parameter fitting values ​​of the image based on the fitting function, the method further includes: Establish a polynomial function with the light source parameters of the detected light source as independent variables and the image parameters as dependent variables; For each of the scanning probes, a preset number of data samples are collected; each data sample represents the correspondence between the light source parameter values ​​and the image parameter values ​​of the detection light source. Based on the data sample, the polynomial coefficients of the polynomial function are calculated, and the fitting function is obtained based on the calculation results.

3. The light source adjustment method as described in claim 1, characterized in that, Before determining the image parameter fitting values ​​of the image based on the fitting function, the method further includes: Establish a linear function with the light source parameters of the detected light source as the independent variable and the image parameters as the dependent variable; For each of the scanning probes, a preset number of data samples are collected; each data sample represents the correspondence between the light source parameter values ​​and the image parameter values ​​of the detection light source. Based on the data sample, the proportional coefficient of the linear function is calculated, and the fitting function is obtained based on the calculation result.

4. The light source adjustment method as described in claim 2 or 3, characterized in that, After obtaining the fitting function based on the calculation results, the process includes: Based on the data sample, the deviation between the fitted value obtained by the fitting function and the estimated value is calculated based on a preset error function. The fitting function is evaluated based on the deviation. If the evaluation fails, return to the step of collecting a preset number of data samples for each of the scanning probes; If the evaluation passes, the fitted function is determined to be the final fitted function.

5. The light source adjustment method as described in claim 1, characterized in that, The steps of adjusting the light source parameters corresponding to the detection light source within the current scanning cycle, determining the true values ​​of the image parameters of the image obtained within the current scanning cycle, and determining the fitted values ​​of the image parameters of the image based on the fitting function include: Based on the difference between the image parameters and the target value, the light source parameters of the detection light source are adjusted according to the current light source parameters and parameter adjustment step size of the detection light source. The images of the target object acquired by each of the scanning probes in the current scanning cycle under the adjusted detection light source are obtained. The true values ​​of the image parameters of the images obtained in the current scanning cycle are determined, and the fitted values ​​of the image parameters of the images are determined based on the fitting function.

6. The light source adjustment method as described in claim 5, characterized in that, The step of adjusting the light source parameters of the detection light source based on the difference between the image parameters and the target value, and based on the current light source parameters and parameter adjustment step size of the detection light source, includes: If the image parameters are less than the target value, adjust the step size according to the current light source parameters and parameters of the corresponding detection light source, and increase the light source parameters of the detection light source; If the image parameters are greater than the target value, the step size is adjusted according to the current light source parameters and parameters of the corresponding detection light source to reduce the light source parameters of the detection light source.

7. A defect detection method based on the light source adjustment method as described in any one of claims 1 to 6, characterized in that, Also includes: Maintaining the target light source parameters of each of the detection light sources, performing image acquisition of the target object in the next scanning cycle, and performing defect analysis on the target object based on the acquired image.

8. A defect detection device, characterized in that, It includes a processor, a memory connected to the processor, and a computer program stored in the memory and executable by the processor. When the computer program is executed by the processor, it implements the steps of the defect detection method as described in claim 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the light source adjustment method as described in any one of claims 1 to 6, or the steps of the defect detection method as described in claim 8.

10. A defect detection system, characterized in that, The device includes a support platform, a plurality of scanning probes spaced apart along a first direction of the support platform, a detection light source corresponding to the scanning probes, and a defect detection device as described in claim 8. The support platform is used to drive the target object to reciprocate along the second direction so as to pass through the scanning probe; or, the support platform is used to drive the scanning probe to reciprocate along the second direction so as to pass through the target object. Each of the detection light sources is used to provide illumination when the corresponding scanning probe acquires images; Each of the scanning probes is used to acquire images of the target object passing through in each scanning cycle and send the images to the defect detection device.

11. The defect detection system as described in claim 10, characterized in that, The support platform is provided with a carrier panel, and the target objects are arranged in rows and columns on the carrier panel; The interval between two adjacent scanning probes is equal to the row spacing of the target object row on the carrier panel; or, the interval between two adjacent scanning probes is equal to a multiple of the row spacing of the target object row on the carrier panel.

Images