Automatic parameter correction method and device and storage medium

Abstract

The invention provides an automatic parameter correction method and device and a storage medium, and the method comprises the following steps: initializing built-in parameters of a detection equation F, and obtaining an original parameter set Q0; obtaining an input set P converted from a plurality of images corresponding to the over-judgment error; each parameter Aj in the original parameter set Q0 is expanded, and all parameter expansion sets QAj form a parameter expansion total set W; the parameter extraction sets form a parameter extraction total set WX; taking the original parameter set Q0 as a built-in parameter of a detection equation F, and calculating to obtain a first type of detection result set R0; taking each parameter extraction set in the total parameter extraction set WX as a built-in parameter of the detection equation F in sequence, and respectively calculating to obtain X second-class detection result sets R1 ', R2'... RX '; calculating to obtain a difference set E; according to the method, the adjusting speed is increased, and the optimal solution of the parameters can be found out.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic cell testing, and in particular to an automatic parameter correction method for over-judgment during AOI testing of photovoltaic cells. Background Technology

[0002] AOI stands for Automated Optical Inspection. When inspecting photovoltaic cells using AOI, the camera first acquires an image of the cell, which is then transmitted to a computer for image processing. In the computer, the image of the cell is converted into an input matrix P. i (Corresponding to a single image), and the detection result r=F(Q, P) is obtained after calculation using the designed detection equation F. i ), where Q is the parameter set, P i The input to the detection equation F is used; then the detection result r is classified according to certain business rules, thus completing the detection of photovoltaic cells; the detection result r can be the brightness of the calculated image, or the area of ​​a certain region, etc.

[0003] Over-judgment occurs when a photovoltaic cell that was originally qualified is mistakenly judged as unqualified. In this case, it is necessary to adjust the built-in parameters of the detection equation F, that is, to adjust the parameter set Q, so as to correct the detection result and improve the detection accuracy. Usually, when over-judgment errors occur, the parameters need to be adjusted manually based on experience. This is highly dependent on human experience, and the adjustment effect may be poor or the adjustment process may be very slow. Summary of the Invention

[0004] To address at least one technical problem in the prior art, embodiments of the present invention provide an automatic parameter correction method, device, and storage medium, capable of automating the adjustment of parameters in a detection equation, thereby accelerating the adjustment process and identifying the optimal parameter solution under current conditions. To achieve the above technical objectives, the technical solution adopted by embodiments of the present invention is as follows: In a first aspect, embodiments of the present invention provide an automatic parameter correction method, comprising the following steps: Initialize the built-in parameters of the detection equation F to obtain the original parameter set Q0; the number of parameters in the original parameter set Q0 is n; obtain the input set P converted from multiple images corresponding to over-judgment errors; For each parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Expand the set Q of each parameter Aj The total set of compositional parameters is expanded to W; The parameter set Q is expanded sequentially from the total parameter set W to each parameter set Q. Aj A parameter is extracted from each parameter to form a parameter extraction set, and all parameter extraction sets are combined to form the total parameter extraction set W.X X is the total set of parameters extracted, W. X Number of elements in the middle; Each element P in the input set P i As input to the detection equation F, the original parameter set Q0 is used as the built-in parameter of the detection equation F to calculate the first type of detection result and obtain the first type of detection result set R0. Each element P in the input set P i As input to the detection equation F, the parameters are extracted from the total set W. X Each parameter extraction set is used sequentially as a built-in parameter of the detection equation F, and the second-class detection results are calculated respectively, resulting in X sets of second-class detection results R1′, R2′...R X ′; Calculate the sets of each type II detection result: R1′, R2′, ..., R X The difference between the result set R0 and the result set R0 of the first type of detection is used to obtain the difference set E; Calculate the mode e in the set of differences E mod Then calculate the sum of each element in the difference set E and the mode e. mod The absolute value of the difference is taken, and the parameter extraction set corresponding to the maximum value of the absolute value of the difference is taken as the parameter correction result.

[0005] Further, obtaining the input set P, which is transformed from multiple images corresponding to over-judgment errors, includes: The multiple images corresponding to over-judgment errors are each converted into a matrix; The matrices corresponding to each image with an over-judgment error are used to form the input set P.

[0006] Furthermore, for each parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Specifically, it includes: For parameter A j With S j Let m be the step size, and expand downwards. j Next, we get m j One extended parameter; For parameter A j With S j Let m be the step size, and expand upwards. j Next, we get m j One extended parameter; 2m is obtained by expanding upwards and downwards. j The extended parameters form the parameter extension set Q. Aj ;Parameter extension set Q Aj Represented as: QAj ={(A j -m j S j )..., (A j -2S j (A) j -S j (A) j +S j (A) j +2S j ), ... (A j +m j S j )}; The parameter extraction set W X The number of elements in the middle is X= .

[0007] Furthermore, for each parameter A in the original parameter set Q0 j m j The value range is 3 to n.

[0008] Furthermore, the m j With parameter A j It shows a negative correlation.

[0009] Furthermore, the step size S j Configure it as the corresponding parameter A j 1 / 8 to 1 / 20 of.

[0010] Furthermore, the calculation of each set of second-type detection results R1′, R2′...R X The difference between R1′ and the first type of detection result set R0 is specifically calculated by calculating the differences between each second type of detection result set R1′, R2′, ..., R0. X ′ and the Euclidean distance of the first type of detection result set R0.

[0011] Furthermore, when there are multiple maximum values ​​of the absolute value of the difference, the variance of the corresponding multiple second-type detection result sets is calculated, and the parameter extraction set corresponding to the second-type detection result set with the smallest variance is selected as the parameter correction result.

[0012] In a second aspect, embodiments of the present invention provide an electronic device, comprising: Memory, which stores computer programs; A processor for running the computer program, which, when running, performs the steps of the parameter automatic correction method as described above.

[0013] Thirdly, embodiments of the present invention provide a storage medium storing a computer program configured to execute the steps of the parameter automatic correction method described above during runtime.

[0014] The beneficial effects of the technical solution provided by the embodiments of the present invention are as follows: 1) It does not require human experience. Even ordinary technicians can use computers to adjust the built-in parameters of the detection equation and obtain the optimal solution.

[0015] 2) The built-in parameters of the detection equation can be adjusted automatically, which greatly speeds up the adjustment process. Attached Figure Description

[0016] Figure 1 This is a flowchart of the automatic parameter correction method in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of an electronic device in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of the embodiments of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present 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. Therefore, they should not be construed as limitations on the present invention. In addition, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0020] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" 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 mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0021] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0022] This invention proposes an automatic parameter correction method for over-judgment during AOI detection of photovoltaic cells, comprising the following steps: Step S10: Initialize the built-in parameters of the detection equation F to obtain the original parameter set Q0, which is represented as: Q0={A1, A2, ..., A j ,…,A n}, j=1……n; n is the number of parameters in the original parameter set Q0; Obtain the input set P, which is the result of transforming multiple images corresponding to over-judgment errors; the input set P is represented as: P = {P1, P2, ..., P} i ...P K}, i=1……K; K is the number of elements in the input set P; Specifically, the input set P obtained from the multiple images corresponding to the over-judgment errors includes: The multiple images corresponding to over-judgment errors are each converted into a matrix; The matrices corresponding to each image with an over-judgment error are used to form the input set P; Step S20, for each parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Expand the set Q of each parameter Aj The total set of compositional parameters is expanded to W; The parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Specifically, it includes: For parameter A j With S j Let m be the step size, and expand downwards. j Next, we get m j One extended parameter; For parameter A j With S j Let m be the step size, and expand upwards. j Next, we get m j One extended parameter; 2m is obtained by expanding upwards and downwards. j The extended parameters form the parameter extension set Q. Aj ;Parameter extension set Q Aj Represented as: Q Aj ={(A j -m j S j )..., (A j -2S j (A) j -S j (A) j +S j (A) j +2S j ), ... (A j +m j S j )} In this embodiment, for each parameter A in the original parameter set Q0 j m j All are configured to be the same (i.e., m1 to m n (All are the same), m j The value range of m is 3 to n; j The preferred value range for m is n, which is the number of parameters in the original parameter set Q0; j The value can be set based on experience, for example, parameter A. j When it is small, the corresponding m j Larger, generally speaking, the m j With parameter A j It shows a negative correlation; In other embodiments, m j They can also be different; Step size S j Configure it as the corresponding parameter A j 1 / 8 to 1 / 20, for example 1 / 8, 1 / 9, 1 / 10, 1 / 12, 1 / 14, 1 / 16, 1 / 18, 1 / 20; The total set of parameter augmentation, W, is represented as: W={Q A1 Q A2 …, Q Aj Q An} Step S30: Sequentially expand each parameter expansion set Q from the total parameter expansion set W. Aj A parameter extraction set is formed by extracting one parameter from each of the following, and the total number of parameters is X = There are several parameter extraction methods, and the various parameter extraction sets are combined to form the total parameter extraction set W. X X is the total set of parameters extracted, W. X The number of elements in the middle, where X = ;Parameter extraction total set W X Represented as: WX= {Q1′,Q2′,…,Q x ′} Where X = ; Q1′,Q2′,…,Q X ' represents the set of parameters extracted; Step S40, take each element P in the input set P i As input to the detection equation F, the original parameter set Q0 is used as the built-in parameter of the detection equation F to calculate the first type of detection result and obtain the first type of detection result set R0. The first type of test result is represented as follows: r i =F(Q0,P i ) The first type of detection result set R0 is represented as: R0 = {r1, r2, ..., r} i , ..., r K} Step S50, take each element P in the input set P i As input to the detection equation F, the parameters are extracted from the total set W. X Each parameter extraction set is used sequentially as a built-in parameter of the detection equation F, and the second-class detection results are calculated respectively, resulting in X sets of second-class detection results R1′, R2′...R X ′; Extract the total set W using parameters X Taking the first parameter extraction set Q1′ as an example, the parameter extraction set Q1′ is used as the built-in parameter of the detection equation F to calculate the second type of detection result r. i =F(Q1′,P i The corresponding set of second-type detection results is R1′={r1′, r2′…, r…} i ′…,r K ′}; and so on, to obtain the remaining sets of second-type detection results R2′……R x ′; Step S60: Calculate the sets of each type II detection result R1′, R2′...R X The difference between the result set R0 and the result set R0 of the first type of detection is used to obtain the difference set E; In this embodiment, the calculation of each set of second-type detection results R1′, R2′...R X The difference between R1′ and the first type of detection result set R0 is specifically calculated by calculating the differences between each second type of detection result set R1′, R2′, ..., R0. X The Euclidean distance between ′ and the first type of detection result set R0; The set of differences E is represented as follows: E={e1′,e2′,…,e x ′} e1′, e2′, ..., e x R' represents the sets of various second-type detection results R1', R2', ..., R'. X The Euclidean distance between ′ and the first type of detection result set R0; For example, e1′=R1′-R0=

[0023] Since the current condition is an over-judgment error, if e1′, e2′, ..., e x The smaller the difference in e1′, the closer the second-class detection result set is to the first-class detection result set, meaning the higher the probability of misclassification. Therefore, the optimization direction is to find e1′, e2′, ..., e x The outlier with the largest difference in the set of differences E is the one that is furthest from the set of differences E. Step S70: Calculate the mode e in the difference set E. mod Then calculate the sum of each element in the difference set E and the mode e. mod The absolute value of the difference between ′ is taken as the parameter extraction set corresponding to the maximum value of the absolute value of the difference, and the parameter correction result is obtained. The mode is the number that appears most frequently in the combination of differences E; Calculate the sum of each element in the difference set E and the mode e. mod The absolute value of the difference between ′ is expressed as: |e1′-e mod ′|,|e2′-e mod ′|,…,|e x ′-e mod ′| The parameter extraction set corresponding to the second type of detection result set corresponding to the maximum absolute value of the difference is the parameter correction result; More preferably, when there are multiple maximum values ​​of the absolute value of the difference, the stability of the detection needs to be considered, that is, the fluctuation of the detection result after calculation of K images is minimized; the following steps can be performed; Step S80: When there are multiple maximum values ​​of the absolute value of the difference, calculate the variance of the corresponding multiple second-type detection result sets, and select the parameter extraction set corresponding to the second-type detection result set with the smallest variance as the parameter correction result. like Figure 2 As shown, this embodiment of the invention also proposes an automatic parameter correction device, comprising: The initialization and input acquisition module is used to initialize the built-in parameters of the detection equation F to obtain the original parameter set Q0; and to obtain the input set P converted from multiple images corresponding to over-judgment errors. The parameter expansion module is used to expand each parameter A in the original parameter set Q0. j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Expand the set Q of each parameter Aj The total set of compositional parameters is expanded to W; The parameter extraction module is used to sequentially expand each parameter expansion set Q from the overall parameter expansion set W. Aj A parameter is extracted from each parameter to form a parameter extraction set, and all parameter extraction sets are combined to form the total parameter extraction set W. X ; The first calculation module is used to process each element P in the input set P. i As input to the detection equation F, the original parameter set Q0 is used as the built-in parameter of the detection equation F to calculate the first type of detection result, thereby obtaining the first type of detection result set R0; The second calculation module is used to process each element P in the input set P. i As input to the detection equation F, the parameters are extracted from the total set W. X Each parameter extraction set is used sequentially as a built-in parameter of the detection equation F, and the second-type detection results are calculated respectively, thus obtaining X sets of second-type detection results R1′, R2′...R X ′; The difference calculation module is used to calculate the values ​​of each type II detection result set R1′, R2′, ..., R. X The difference between the result set R0 and the result set R0 of the first type of detection is used to obtain the difference set E; The correction module is used to calculate the mode e in the difference set E. mod Then calculate the sum of each element in the difference set E and the mode e. mod The absolute value of the difference is taken, and the parameter extraction set corresponding to the maximum absolute value of the difference is taken as the parameter correction result; when there are multiple maximum absolute values ​​of the difference, the variance of the multiple second-type detection result sets is calculated, and the parameter extraction set corresponding to the second-type detection result set with the smallest variance is selected as the parameter correction result.

[0024] like Figure 2As shown in the figure, this invention also proposes an electronic device, including: a processor and a memory; the processor and the memory communicate with each other, for example, by being connected and communicating with each other via a bus; the memory stores a computer program; the processor is used to run the computer program, and the computer program executes the steps of the parameter automatic correction method described above when it runs; the processor may be a CPU, or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.; the memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory may also include combinations of the above types of memory.

[0025] This invention also proposes a storage medium storing a computer program configured to execute the steps of the parameter automatic correction method described above during runtime. The storage medium includes a disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), and any combination thereof.

[0026] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for automatic parameter correction, characterized in that, Includes the following steps: Initialize the built-in parameters of the detection equation F to obtain the original parameter set Q0; the number of parameters in the original parameter set Q0 is n; obtain the input set P converted from multiple images corresponding to over-judgment errors; For each parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Expand the set Q of each parameter Aj The total set of compositional parameters is expanded to W; The parameter set Q is expanded sequentially from the total parameter set W to each parameter set Q. Aj A parameter is extracted from each parameter to form a parameter extraction set, and all parameter extraction sets are combined to form the total parameter extraction set W. X X is the total set of parameters extracted, W. X Number of elements in the middle; Each element P in the input set P i As input to the detection equation F, the original parameter set Q0 is used as the built-in parameter of the detection equation F to calculate the first type of detection result and obtain the first type of detection result set R0. Each element P in the input set P i As input to the detection equation F, the parameters are extracted from the total set W. X Each parameter extraction set is used sequentially as a built-in parameter of the detection equation F, and the second-class detection results are calculated respectively, resulting in X sets of second-class detection results R1′, R2′...R X ′; Calculate the sets of each type II detection result: R1′, R2′, ..., R X The difference between the result set R0 and the result set R0 of the first type of detection is used to obtain the difference set E; Calculate the mode e in the set of differences E mod Then calculate the sum of each element in the difference set E and the mode e. mod The absolute value of the difference is taken, and the parameter extraction set corresponding to the maximum value of the absolute value of the difference is taken as the parameter correction result.

2. The automatic parameter correction method as described in claim 1, characterized in that, The step of obtaining the input set P, which is formed by converting multiple images corresponding to over-judgment errors, includes: The multiple images corresponding to over-judgment errors are each converted into a matrix; The matrices corresponding to each image with an over-judgment error are used to form the input set P.

3. The automatic parameter correction method as described in claim 1, characterized in that, The parameter A in the original parameter set Q0 j By expanding the set of parameters, we obtain the expanded parameter set Q corresponding to each parameter. Aj Specifically, it includes: For parameter A j With S j Let m be the step size, and expand downwards. j Next, we get m j One extended parameter; For parameter A j With S j Let m be the step size, and expand upwards. j Next, we get m j One extended parameter; 2m is obtained by expanding upwards and downwards. j The extended parameters form the parameter extension set Q. Aj ;Parameter extension set Q Aj Represented as: Q Aj ={(A j -m j S j )……,(A j -2S j ),(A j -S j ),(A j +S j ),(A j +2S j ),……(A j +m j S j )}; The parameter extraction set W X The number of elements in the middle is X= .

4. The automatic parameter correction method as described in claim 3, characterized in that, For each parameter A in the original parameter set Q0 j m j The value range is 3 to n.

5. The automatic parameter correction method as described in claim 4, characterized in that, The m j With parameter A j It shows a negative correlation.

6. The automatic parameter correction method as described in claim 3, characterized in that, The step size S j Configure it as the corresponding parameter A j 1 / 8 to 1 / 20 of.

7. The automatic parameter correction method as described in claim 1, characterized in that, The calculation of each second-type detection result set R1′, R2′...R X The difference between R1′ and the first type of detection result set R0 is specifically calculated by calculating the differences between each second type of detection result set R1′, R2′, ..., R0. X ′ and the Euclidean distance of the first type of detection result set R0.

8. The automatic parameter correction method as described in claim 1, characterized in that, When there are multiple maximum values ​​of the absolute value of the difference, the variance of the corresponding multiple second-type detection result sets is calculated, and the parameter extraction set corresponding to the second-type detection result set with the smallest variance is selected as the parameter correction result.

9. An electronic device, characterized in that, include: Memory, which stores computer programs; A processor for running the computer program, wherein the computer program, when running, performs the steps of the automatic parameter correction method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium stores a computer program configured to execute the steps of the parameter automatic correction method as described in any one of claims 1 to 8 when running.

Images