Spatial resolution test board of electroluminescent defect detector
By designing an electroluminescent defect detector test board including positioning area, inclination angle detection area, rectangular line-to-region and wedge line-to-region, combined with neural network model and HYRes black line detection algorithm, the problem of large measurement error in the prior art is solved, and high-precision spatial resolution measurement is achieved.
Patent Information
- Application Number
- CN202421912371.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2025-07-29
- Estimated Expiration
- 2034-08-08
AI Technical Summary
The spatial resolution test boards of existing electroluminescent defect detectors are easily affected by sticking operation errors during the measurement process, making it difficult to accurately measure the spatial resolution in all directions, and have low measurement accuracy.
A spatial resolution test board for electroluminescent defect detector is designed, including positioning area, inclination angle detection area, rectangular line pair area and wedge line pair area. By setting rectangular line pair groups and wedge line pair groups in different directions, combining neural network model and HYRes black line detection algorithm, calibration is automatically carried out and measurement accuracy is improved.
It realizes high-precision spatial resolution measurement of electroluminescent defect detectors in different directions, reduces operating errors, improves measurement accuracy and automated discrimination capabilities.
Smart Images
Figure CN223168299U_ABST
Abstract
Description
Technical Field
[0001] The utility model relates to the technical field of electro - luminescence defect detector testing, in particular to a spatial resolution test board for an electro - luminescence defect detector. Background Art
[0002] The national metrological technical specification "JJF 2063 - 2023 Calibration Specification for Electro - luminescence Defect Detectors for Photovoltaic Modules" puts forward the calibration method for electro - luminescence defect detectors for photovoltaic modules (hereinafter referred to as EL devices). As Figure 1 shown, the basic principle of an electro - luminescence defect detector is as follows: A forward current is passed through a photovoltaic module, causing the photovoltaic module to emit near - infrared light. Then, a CCD camera is used to capture the light emitted by the photovoltaic module and form an image. Since the light intensity emitted by the defective part of the photovoltaic module is different from that of the non - defective part, defects that cannot be detected by the human eye in the photovoltaic module can be identified based on the image.
[0003] For the imaging system of an electro - luminescence defect detector, a spatial resolution test board is used for calibration in the national metrological technical specification. The existing spatial resolution test board is as Figure 2 shown, which is a stainless - steel thin sheet with multiple groups of hollowed - out patterns etched on it. One group of hollowed - out patterns consists of 6 equal - width rectangles arranged at equal intervals (the interval is the same as the width). One group of stripes (i.e., one rectangle line - pair group) corresponds to one spatial resolution (spatial resolution = 0.5 / line - pair width). During calibration, this device is pasted on the surface of the photovoltaic module to be measured. After the photovoltaic module is powered on by the electro - luminescence defect detector, the electro - luminescence of the photovoltaic module passes through the hollowed - out rectangles in the spatial resolution test board to form a line - pair group of light and dark. By visually discriminating the line - pair group (from sparse to dense) by the human eye, if 5 black lines can be clearly distinguished, then this line - pair group passes the discrimination. Keep looking until a line - pair group that cannot pass the discrimination is found. Then, the spatial resolution corresponding to the previous line - pair group of this line - pair group is the calibration result of the spatial resolution of the imaging system of the electro - luminescence defect detector at the corresponding position.
[0004] When using the existing spatial test board with rectangular line - pairs to detect the spatial resolution of an electro - luminescence defect detector, on the one hand, it is easily affected by the pasting operation, and the angle deviates, resulting in an error between the judged spatial resolution in the vertical or horizontal direction and the actual value. On the other hand, it is also impossible to accurately measure the resolution between two adjacent line - pair groups, with low measurement accuracy, and it is also impossible to measure the spatial resolution of the electro - luminescence defect detector in other directions simultaneously. Summary of the Utility Model
[0005] The technical problem to be solved by the present utility model is to provide a spatial resolution test board for an electroluminescent defect detector, which can improve the measurement accuracy and at the same time can measure the spatial resolution of the electroluminescent defect detector in different directions.
[0006] The present utility model is implemented as follows: A spatial resolution test board for an electroluminescent defect detector, on which a positioning area, an inclination angle detection area, a rectangular line pair area and a wedge line pair area are provided. A positioning circle is provided on the positioning area, and an inclination angle detection strip is vertically provided on the inclination angle detection area. The positioning circle is located at the upper left position of the inclination angle detection strip;
[0007] In the rectangular line pair area, rectangular line pair groups in different directions are provided. A plurality of rectangular line pair groups corresponding to different resolutions are provided in the same direction, and a wedge line pair group is correspondingly provided for the rectangular line pair groups in the same direction. The wedge line pair group is located in the wedge line pair area.
[0008] Further, both the rectangular line pair group and the wedge line pair group include horizontal direction, vertical direction, 45° inclination direction and 135° inclination direction.
[0009] Further, the positioning circle, the rectangular line pair group, the wedge line pair group and the inclination angle detection strip are all provided with hollow-outs.
[0010] The present utility model has the following advantages: By setting the positioning area in combination with the inclination angle detection area, the inclination angle error during operation can be corrected. By setting rectangular line pair groups in different directions, the resolution in each direction can be measured. A wedge line pair group is correspondingly provided for each direction to further accurately measure the measurement result of the rectangular line pair group and improve the measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The following further describes the present utility model with reference to the accompanying drawings in conjunction with embodiments.
[0012] Figure 1 It is a schematic diagram of the test principle of the electroluminescent defect detector.
[0013] Figure 2 It is a schematic diagram of the structure of the spatial resolution test board adopted by the existing electroluminescent defect detector.
[0014] Figure 3 It is a schematic diagram of the structure of a spatial resolution test board for an electroluminescent defect detector provided by the present utility model. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0015] Please refer to Figure 3, the present utility model provides a spatial resolution test board for an electroluminescence defect detector. A positioning area 1, an inclination angle detection area 2, a rectangular line pair area 3, and a wedge-shaped line pair area 4 are provided on the test board. A positioning circle is provided on the positioning area, and an inclination angle detection strip is vertically provided on the inclination angle detection area. The positioning circle is located at the upper left position of the inclination angle detection strip;
[0016] In the rectangular line pair area, rectangular line pair groups in different directions are provided. A plurality of rectangular line pair groups corresponding to different resolutions are provided in the same direction, and a wedge-shaped line pair group is correspondingly provided for the rectangular line pair groups in the same direction. The wedge-shaped line pair group is located within the wedge-shaped line pair area. The position division and positioning of different areas facilitate subsequent automatic image recognition and interception.
[0017] In a specific embodiment, both the rectangular line pair group and the wedge-shaped line pair group include horizontal direction, vertical direction, 45° inclination direction, and 135° inclination direction.
[0018] In a specific embodiment, the positioning circle, the rectangular line pair group, the wedge-shaped line pair group, and the inclination angle detection strip are all provided with hollow openings.
[0019] The process of using the test board of the present utility model for testing is as follows:
[0020] Step S1: Manufacture the test board of the present utility model;
[0021] Step S2: Paste the spatial resolution test board on the surface of the photovoltaic module to be tested. After powering on the photovoltaic module through the electroluminescence defect detector, the electroluminescence of the photovoltaic module passes through the positioning circle, the inclination angle detection strip, the rectangular line pair group, and the wedge-shaped line pair group in the spatial resolution test board to obtain corresponding image information.
[0022] Step S3: Establish a Cartesian coordinate system with the geometric center of the image as the zero point, the horizontal right direction as the positive x-axis direction, and the vertical upward direction as the positive y-axis direction. The obtained image is subjected to the following operations: grayscale conversion - binarization - edge detection - Hough detection to obtain the center coordinates (x1, y1) of the positioning circle, the coordinates of the two end points (x2, y2), (x3, y3) of the inclination angle detection strip, and its included angle θ with the y-axis. Rotate the image by -θ to keep the inclination angle detection vertical. Then, continuously perform the following operations on the image: rotate the image by 90° around the image center, or flip the image along the x-axis or y-axis until the following two conditions are simultaneously met: (1) x1 < x2 and x1 < x3; (2) y1 > 0.5×(y2 + y3);
[0023] Step S4: According to the distance relationship between each pair of lines and the center (x1, y1) of the positioning circle, the computer automatically intercepts all rectangular line pair regions and wedge-shaped line pair regions. For each line pair image, according to its inclination angle, the rectangular line pair images with horizontal, 45° inclination, and 135° inclination are rotated so that the long sides of the rectangles are finally parallel to the vertical direction;
[0024] Step S5: Construct a neural network model to judge the image. First, extract features of the image through multiple feature modules (each feature module has the same structure but different parameters). Each feature extraction module includes a convolutional layer, a ReLU activation function layer, and a max pooling layer. The convolutional layer is used to extract features of each dimension of the image, the ReLU activation function layer is used to introduce non-linearity to improve the expression ability of the model, and the max pooling layer is used to reduce the dimensionality of the image size. After the last feature extraction module is executed, a spatial pyramid pooling layer is used to unify the sizes of image features of different sizes and then output. Then, connect a classifier composed of a certain number of fully connected layers to output the judgment value J1 of the rectangular line pair image, where 0 <= J1 <= 1. For the neural network model, the training set is prepared in the following way: Paste the spatial resolution test board on different types of photovoltaic modules (different numbers of grid lines, different technical types (PERC, topcon, HJT, etc.)), use EL devices produced by different manufacturers, and use different shooting parameter settings (parameters include: exposure time, ISO, energized current, etc.) to shoot images of the spatial resolution test board. Separate all the rectangular line pair images from them and rotate them to the vertical direction. Copy all the line pair images N times (an odd number) and shuffle the order. Then select M (an odd number) discriminators. Each discriminator discriminates all the line pair images. Each person discriminates each image N times. For each discriminator, the discrimination result of each image is based on the majority of the N results. For each image, the final discrimination result is based on the majority of the M persons. Finally, obtain a certain number of line pair images with labels of "discrimination passed" or "discrimination failed" and use them as the training set of the neural network to train the neural network model.
[0025] Step S6: Obtain the rectangular line pair image in Step S4, calculate the average value of the image grayscale along the long side direction of the black and white line pair, and obtain a grayscale curve with the image width (pixel points) as the independent variable and the grayscale average value as the dependent variable. Use the SR02 black line detection algorithm in the HYRes algorithm to obtain the number of black lines in the grayscale curve. If the number of black lines is not 5, output the determination value J2 as 0; otherwise, output the determination value J2 as 1. Specifically, the execution process of the SR02 black line detection algorithm in the HYRes algorithm is as follows: Find the number of minimum values in each grayscale curve to distinguish the number of black lines in the image. Set a threshold. Only when the increase in the local minimum value exceeds the threshold is it determined that there is a black line in the image, and as the line pair density increases, this threshold gradually decreases to adapt to the decrease in the grayscale amplitude.
[0026] Step S7: Obtain the rectangular line pair image in Step S3 and magnify it. The magnification ratio is set according to actual needs. Judge the rectangular line pair image by the human eye. When the number of black lines can be recognized as 5, obtain the determination value J3 as 1; otherwise, J3 is 0. For a black line in a line pair image, if the ratio of the recognized length of the black line to the total length of the line pair image exceeds 50%, then it is defined that the black line can be recognized.
[0027] Step S8: Calculate the final discrimination value J. If J is greater than or equal to 0.5, the discrimination result of the rectangular line pair image is discriminable; otherwise, it is non - discriminable.
[0028] Step S9: Repeat Steps S5 to S8, and discriminate one by one from the rectangular line pair image corresponding to the low - spatial resolution to the rectangular line pair image corresponding to the high - spatial resolution. If the discrimination result of a certain rectangular line pair image is discriminable, continue to discriminate the next rectangular line pair image until the discrimination result of a certain rectangular line pair image is non - discriminable. Record the spatial resolution corresponding to the rectangular line pair image with the previous discriminable result as the initial spatial resolution R of the EL device at this position and in this direction. s1 ;
[0029] Step S10: Find the spatial resolution R corresponding to the last distinguishable line pair in the rectangular line pair s1 and the spatial resolution R corresponding to the first non - distinguishable line pair s2 , and analyze in combination with the wedge - shaped line pair image in the same direction: If the wedge - shaped line pair is in the horizontal direction, the inclined 45° direction, or the inclined 135° direction, then rotate it so that the wedge - shaped line pair image is vertically arranged, that is, the low - spatial resolution area is at the bottom of the screen and the high - spatial resolution area is at the top of the screen. Find the rows corresponding to the spatial resolutions R s1 and R s2 in the wedge - shaped line pair image. Starting from the row corresponding to R s1 and moving towards the row corresponding to R s2The corresponding rows measure the number of black lines row by row using the SR02 black line detection algorithm in the HYRes algorithm until the row where the number of black lines is not 5 is found, and the corresponding spatial resolution is the final spatial resolution Rs of the EL device at this position and in this direction;
[0030] Among them, when measuring the number of black lines row by row using the SR02 black line detection algorithm in the HYRes algorithm, the image gray level is averaged along the long side direction of the black and white line pairs in advance to obtain a gray level curve with the image width as the independent variable and the gray level average value as the dependent variable. Then, the SR02 black line detection algorithm in the HYRes algorithm is used to obtain the number of black lines in the gray level curve. Specifically, for each gray level curve, the number of minimum values in the curve is found to determine the number of black lines in the image. A threshold is set, and only when the increase in the local minimum value exceeds the threshold is it determined that there is a black line in the image. And as the line pair density increases, the threshold gradually decreases to adapt to the decrease in the gray level amplitude.
[0031] The above test method uses the existing common test methods (human eye discrimination and neural network model) for the spatial resolution of the electroluminescence defect detector, and introduces the HYRes black line detection algorithm in the "CPIA DC-003 Measuring Method for the Resolution of Digital Cameras" standard to discriminate the spatial resolution of the electroluminescence defect detector. After weighting the three discrimination methods, the preliminary discrimination result for the rectangular line pair is obtained, and then the wedge line pair is used for further analysis to determine the final resolution, improving the test accuracy.
[0032] Compared with the existing spatial resolution test board that only sets rectangular line pairs, the test board of the present utility model is provided with a detection strip to facilitate image correction after imaging, and uses a positioning circle to flip the reversed image and other operations so that the required image block can be automatically intercepted during the algorithm execution. At the same time, rectangular line pairs and wedge line pairs in different directions are set, and the final measurement result with higher accuracy is obtained through two steps of preliminary measurement and precise measurement. Using the self-made test board of the present utility model can reduce the operation error, provide support for automatic discrimination, and improve the detection accuracy of the spatial resolution of the electroluminescence defect detector.
[0033] Although the specific implementation manners of the present utility model have been described above, those skilled in the art of this technology should understand that the specific embodiments we described are illustrative rather than used to limit the scope of the present utility model. Equivalent modifications and variations made by those skilled in the art in accordance with the spirit of the present utility model should all be covered within the scope protected by the claims of the present utility model.
Claims
1. A spatial resolution test board for an electroluminescence defect detector, characterized in that: The test board is provided with a positioning area, an inclination angle detection area, a rectangular line pair area, and a wedge line pair area. A positioning circle is provided on the positioning area, and an inclination angle detection strip is vertically provided on the inclination angle detection area. The positioning circle is located at the upper left position of the inclination angle detection strip; The rectangular line pair area is provided with rectangular line pair groups in different directions. A plurality of rectangular line pair groups corresponding to different resolutions are arranged in the same direction, and a wedge line pair group is correspondingly arranged for the rectangular line pair groups in the same direction. The wedge line pair group is located within the wedge line pair area.
2. The spatial resolution test board of an electroluminescence defect detector according to claim 1, wherein: Both the rectangular line pair groups and the wedge line pair groups include horizontal direction, vertical direction, 45° inclination direction, and 135° inclination direction.
3. The spatial resolution test board of an electroluminescence defect detector according to claim 1, characterized in that: The positioning circle, the rectangular line pair groups, the wedge line pair groups, and the inclination angle detection strip are all provided with hollow-outs.