Diode conductivity testing method based on machine vision positioning

Abstract

The invention relates to the technical field of diode detection, in particular to a diode conductivity testing method based on machine vision localization, which comprises the following steps: acquiring a diode group image through an image detection module, and determining an arrangement distance and direction; calculating an actual arrangement deviation degree, and comparing the actual arrangement deviation degree with a standard threshold value to determine whether conductivity detection is carried out or not; if qualified, a parallel volt-ampere characteristic curve is detected, the actual segmentation slope and the number of inflection points are determined after segmentation, and the diode damage type and the corresponding number are determined based on the actual segmentation slope and the number of inflection points. According to the invention, the actual arrangement deviation degree of the diode group is determined based on the arrangement spacing and the arrangement direction so as to determine whether the diode conductivity detection is carried out, and the diode damage type and the corresponding damage number are determined based on the actual segmentation slope and the actual inflection point number, so that the diode conductivity detection accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of diode testing technology, and in particular to a method for testing the conductivity of diodes based on machine vision positioning. Background Technology

[0002] A diode is an electronic device made of semiconductor materials such as silicon, selenium, and germanium. A diode has two electrodes: a positive terminal, also called the anode, and a negative terminal, also called the cathode. When a forward voltage is applied between the two terminals, the diode conducts; when a reverse voltage is applied, the diode is cut off. The conduction and cutoff of a diode are equivalent to the switching on and off of a switch. Diodes have unidirectional conductivity; when conducting, the current flows from the anode through the diode to the cathode.

[0003] Chinese Patent Publication No. CN117214650A discloses a method and system for intelligent comprehensive performance testing of diodes, comprising: determining the performance testing parameters of the diodes; acquiring learning data; acquiring standard performance data and standard deviation rate of the performance data of the diodes; determining the maximum permissible deviation rate of the performance data of the diodes; determining the optimal number of diodes to be tested in the same batch; dividing the batches according to the optimal number; obtaining the diode testing components for each batch; obtaining the total performance testing data of the diode testing components for each batch; and determining whether the diode testing components are qualified. The advantages of this invention are: through high-precision detection algorithms and intelligent processing methods, it can quickly and accurately identify defective products, significantly shorten diode testing time, improve production efficiency, and reduce diode production costs.

[0004] Therefore, the current technology has the following problems: when testing the conductivity of diode groups, there is a lack of methods to determine the actual arrangement deviation of the diode group based on the spacing and orientation to determine whether to perform diode conductivity testing, and there is a lack of methods to determine the type and number of diode damages based on the actual segment slope and the actual number of inflection points, thereby improving the accuracy of diode conductivity testing. Summary of the Invention

[0005] To address these issues, the present invention provides a diode conductivity performance testing method based on machine vision positioning. This method overcomes the problems in existing technologies where, when testing the conductivity performance of diode groups, the actual arrangement deviation of the diode group based on the arrangement spacing and direction is not determined to determine whether diode conductivity performance testing should be performed, and the type and number of diode damages are determined based on the actual segment slope and the actual number of inflection points, thereby improving the accuracy of diode conductivity performance testing.

[0006] To achieve the above objectives, the present invention provides a method for testing the conductivity of diodes based on machine vision positioning, comprising: The diode group image is obtained by acquiring image data of the diode group based on the image detection module. The arrangement spacing of the diode group is determined based on the pin distance of any diode in the diode group image. The arrangement direction of the diode group is determined based on the pin markings of the diodes. The actual arrangement deviation of the diode group is determined based on the arrangement spacing and arrangement direction. Compare the actual arrangement deviation with the standard arrangement deviation threshold to obtain the arrangement deviation comparison result, and determine whether to perform diode conductivity testing based on the arrangement deviation comparison result. If the arrangement deviation comparison results are qualified, the conductivity of the parallel diode group is tested based on the preset voltage to obtain the parallel current value. Based on the parallel current value and the parallel voltage value, a parallel volt-ampere characteristic curve is obtained. The parallel volt-ampere characteristic curve is segmented to obtain a segmented volt-ampere characteristic curve. The actual segment slope is determined based on the segmented volt-ampere characteristic curve. The actual number of inflection points is determined based on the actual segmented slope and the parallel volt-ampere characteristic curve. The diode damage type and corresponding damage quantity are determined based on the actual segment slope and the actual number of inflection points. Wherein, the pin distance is the average distance from any diode pin to its adjacent diode pin.

[0007] Furthermore, the process of determining the arrangement spacing of the diode group based on the pin distance of any diode in the diode group image includes, Select any diode and determine its pin position to obtain the actual pin position. Based on the actual pin position, determine the pin distance between it and the pins of surrounding diodes. Based on the pin distance and the pin distance threshold, determine the arrangement spacing.

[0008] Furthermore, the process of determining the arrangement direction of the diode group based on the diode pin markings includes, The actual pin identifier is obtained by determining the pin identifier on any side of any diode based on the diode group image, and the arrangement direction is determined based on the actual pin identifier and the standard pin identifier.

[0009] Furthermore, the process of determining the actual arrangement deviation of the diode group based on the arrangement spacing and arrangement direction includes, Regional diode group images are obtained by performing region segmentation based on diode group images. The regional diode group images are compared with standard diode group images to obtain feature points. The actual arrangement deviation of the diode group is determined based on the number of feature points. The feature points include points with abnormal spacing and points with abnormal pin markings.

[0010] Furthermore, the process of determining whether to perform diode conductivity testing based on the arrangement deviation comparison results includes: If the actual arrangement deviation is less than the standard arrangement deviation threshold, the conductivity of the diode is tested. If the actual arrangement deviation is greater than or equal to the standard arrangement deviation threshold, the conductivity performance of the diode will not be tested.

[0011] Furthermore, the process of verifying the arrangement direction based on the actual segmented slope includes, Determine the forward and reverse voltage ranges of the parallel current-voltage characteristic curve; Compare the actual segmented slope of the forward voltage range with the standard segmented slope range of forward conduction, and compare the actual segmented slope of the reverse voltage range with the standard segmented slope range of reverse cutoff. If the actual segmented slope of the forward voltage range is within the range of the standard segmented slope of the forward conduction and the actual segmented slope of the reverse voltage range is within the range of the standard segmented slope of the reverse cutoff, then the arrangement direction is consistent with the arrangement direction determined based on the pin identifier. If the actual segmented slope of the forward voltage range is within the range of the reverse cutoff standard segmented slope and the actual segmented slope of the reverse voltage range is within the range of the forward conduction standard segmented slope, then the arrangement direction is opposite to the arrangement direction determined based on the pin identifier, and the output arrangement direction verification result is abnormal.

[0012] Furthermore, the process of determining the actual number of inflection points based on the actual piecewise slope and the parallel volt-ampere characteristic curve includes, Select any actual segment slope as the actual segment slope of this segment, determine the actual segment slopes of its adjacent upper and lower segments based on the actual segment slope of this segment, determine the absolute value of the first slope difference based on the actual segment slope of this segment and the actual segment slope of the upper segment, determine the absolute value of the second slope difference based on the actual segment slope of this segment and the actual segment slope of the lower segment, and determine the actual number of inflection points based on the absolute value of the difference and the absolute value threshold of the difference.

[0013] Furthermore, the process of determining the actual number of inflection points based on the absolute value of the difference and the threshold value of the difference includes, If the absolute value of the difference is less than the absolute value threshold of the difference, then the actual segment slope of this segment does not change significantly from the actual segment slope of the adjacent segment, and there is no inflection point between this segment and the adjacent segment. If the absolute value of the difference is greater than or equal to the threshold of the absolute value of the difference, then there is a significant change between the actual segment slope of this segment and the actual segment slope of the adjacent segment. There is an inflection point between this segment and the adjacent segment, and this inflection point is included in the actual number of inflection points.

[0014] Furthermore, the process of determining the diode damage type and corresponding damage quantity based on the actual segmented slope and the actual number of inflection points includes, If the actual number of inflection points is 0, and there exists any actual segment slope that is greater than the maximum value of the standard slope range of the forward conduction area, then it is determined that there is a short-circuited damaged diode. If the actual number of inflection points is greater than the standard number of inflection points, and there is at least one segment where the actual slope is less than the minimum value of the standard slope range of the forward cutoff region, then it is determined that there is an open-circuit damaged diode. If the actual number of inflection points is equal to the standard number of inflection points, and the actual segmented slope is not within its corresponding slope range, then the diode is determined to have performance degradation. The standard slope range of the forward conduction region is the range when the corresponding voltage is higher than the conduction threshold; the standard slope range of the forward cutoff region is the range when the corresponding voltage is lower than the conduction threshold; the conduction threshold voltage is determined based on the material of the test diode.

[0015] Furthermore, the process of determining the diode damage type and corresponding damage quantity based on the actual segment slope and the actual number of inflection points also includes: Based on the diodes with short-circuit damage, the short-circuit slope increment is determined based on the slope contribution value of a single short-circuit diode under forward voltage. Based on the slope difference between the actual segmented slope and the maximum value of the standard slope range in the forward conduction region, the number of short-circuit damaged diodes is determined based on the short-circuit slope increment and the slope difference. Based on the open-circuit damaged diodes, obtain the standard current value of normal parallel diodes under the same voltage, determine the current difference based on the actual current value and the standard current value, use the conduction current of a single normal diode as the preset single-tube current, and determine the number of open-circuit damaged diodes based on the current difference and the preset single-tube current. Based on diodes with performance degradation, a slope deviation threshold for a single performance degradation diode is set. The total deviation is determined based on the actual segmented slope and the standard slope. The number of performance degradation diodes is determined based on the total deviation and the slope deviation threshold.

[0016] Compared with existing technologies, the advantages of this invention lie in its use of machine vision positioning technology and image detection modules to accurately acquire the spacing and orientation of diode groups. This achieves automated and high-precision identification of the spatial arrangement information of diode groups, providing a reliable positioning benchmark for subsequent conductivity performance testing and reducing detection errors caused by positional deviations. By comparing the actual arrangement deviation with the standard deviation to determine whether to perform conductivity performance testing, situations that may affect the accuracy of testing or damage the testing equipment due to abnormal arrangement can be screened out in advance, improving the stability and safety of the testing process and avoiding invalid testing operations. By combining segmented analysis of the volt-ampere characteristic curve, slope, and number of inflection points to determine the type and quantity of damage, visual positioning and electrical performance testing are combined to achieve integrated testing from appearance positioning to performance evaluation, ensuring both the comprehensiveness of the testing and the accuracy of fault identification.

[0017] Furthermore, selecting the diode in the middle position as a reference effectively avoids the problem of interference from PCB boundaries on edge devices, providing a more reliable benchmark for subsequent pin positioning and reducing errors caused by boundary effects. Combining mature contour extraction and Hough line detection technology to locate pin edges, and determining the pin's centerline and endpoints through centerline calculation, fully utilizes the high contrast features between the pin and the background, ensuring the accuracy of pin position identification and providing a reliable coordinate basis for distance calculation. By acquiring multiple sets of adjacent pin distances and covering the devices on both sides of the reference diode, the random errors of single distance measurements can be effectively reduced; setting reasonable thresholds to filter out abnormal values ​​can eliminate extreme data caused by pin bending, soldering misalignment, etc., further improving the effectiveness of distance data; averaging the remaining effective distances results in a more representative arrangement spacing that truly reflects the overall layout pattern of the diode group.

[0018] Furthermore, when determining the arrangement direction based on pin markings, identifying characteristic markings such as white color rings and comparing them with the standard direction allows for accurate judgment of the overall consistency of the diode's positive and negative electrode arrangement. This avoids reverse voltage application during subsequent conductivity performance testing due to incorrect orientation, ensuring the fundamental correctness of electrical testing. When determining the actual degree of arrangement deviation, the diode group is divided into independent sub-regions through area segmentation. The spacing and abnormal points are extracted by comparing with the standard image, and the quantified number of abnormalities reflects the degree of deviation. This makes the identification of arrangement problems more systematic and objective, avoiding the ambiguity of subjective judgment and providing a clear basis for subsequent decisions. The comparison between the degree of deviation and a threshold determines whether to perform conductivity performance testing. This ensures the normal progress of the testing process when the deviation is controllable, using fine-tuning to ensure precise probe contact and improve testing efficiency. Conversely, it promptly stops testing when the deviation is too large, avoiding short circuits and incorrect voltage application caused by pin misalignment, protecting the diodes and testing equipment. Simultaneously, alarm-triggered manual review reduces invalid operations and potential losses in mass production.

[0019] Furthermore, by cross-validating the diode's electrical characteristics with the visual positioning results, the accuracy and reliability of the alignment direction determination are significantly improved. This effectively compensates for the potential limitations of machine vision recognition. In image detection, factors such as blurred pin markings and lighting interference can easily lead to misjudgment of direction. The inherent unidirectional conductivity of diodes causes significant differences in the slope characteristics of forward conduction and reverse cutoff. By comparing the actual segmented slope with the standard range, the visual positioning results can be objectively verified, reducing the error of a single recognition method. The alignment direction directly affects the interpretation of the current-voltage characteristic curve interval. Misjudgment of direction can lead to distortion of curve feature analysis, such as mistaking reverse characteristics for forward characteristics. By inferring the correctness of direction through electrical characteristics, it ensures that subsequent slope analysis, inflection point counting, and damage judgment are based on correct electrical logic, improving the rigor and reliability of the overall testing process, reducing invalid detections or misjudgments caused by incorrect direction, and enhancing the practicality of the method in industrial scenarios.

[0020] Furthermore, the continuous increase in voltage can comprehensively cover the entire operating range of the diode from cutoff to conduction, ensuring the capture of current change characteristics under different voltages and avoiding the omission of characteristics due to discrete voltage points. For the overall testing of parallel diode groups, it can directly reflect the comprehensive conductivity characteristics of the assembly, eliminating the need for individual testing and summarization, greatly improving testing efficiency, especially suitable for mass production scenarios. The synchronous acquisition of voltage and current ensures the consistency of data timing, providing continuous and complete raw data for the subsequent plotting of the volt-ampere characteristic curve, ensuring the accuracy of the curve shape. The standardized control of the preset voltage reduces the randomness of human operation, making the test data of different batches and groups comparable, and providing a unified benchmark for subsequent fault analysis. By selecting the actual slope of each segment, calculating the absolute value of the slope difference with the adjacent upper and lower segments, and combining it with the preset threshold to determine whether there is an inflection point, the precise capture of abrupt changes in the volt-ampere characteristic curve is achieved. This segment-by-segment analysis method can systematically cover all intervals of the curve, avoiding the omission of any possible characteristic abrupt changes, and ensuring the comprehensiveness and accuracy of the statistical count of inflection points. By comparing the absolute value of the difference with a threshold as the criterion, the fuzzy concept of significant change is transformed into a quantifiable standard, reducing errors from subjective judgment and making inflection point identification more objective and consistent, thus meeting the needs of automated detection. By accumulating the actual number of inflection points and comparing it with the standard number of inflection points under normal conditions, abnormal abrupt changes in the curve, such as extra inflection points, can be intuitively identified. These abnormal inflection points directly correspond to diode performance abnormalities such as breakdown, providing clear characteristic clues for subsequent analysis of damage types and making fault diagnosis more targeted.

[0021] Furthermore, in determining the damage type, a multi-dimensional fault identification system is constructed through the coordinated judgment of the number of inflection points and the range of segmented slopes. Diodes with short circuits, open circuits, and performance degradation exhibit distinctly different current-voltage characteristics: short circuits have no inflection points and the slope exceeds the limit; open circuits have multiple inflection points and the slope in the cutoff region is abnormal; and degradation has normal inflection points but the slope deviates. This targeted judgment logic can accurately distinguish different damage types, avoiding misjudgments caused by single-feature judgments and ensuring the accuracy of fault nature identification. In determining the number of damaged diodes, based on the characteristic contribution values ​​of a single diode, such as the short-circuit slope increment, the single-diode conduction current, and the slope deviation threshold, the deviation of the overall electrical characteristics is transformed into a quantifiable number of damaged diodes. This deepens the detection results from qualitative judgment to quantitative data, providing a concrete basis for fault screening, repair, and replacement in mass production, and enhancing the practical value of the detection. The entire process fully integrates the material characteristics of the diodes, such as the standard range for setting the conduction threshold of silicon diodes, making the judgment logic compatible with the inherent properties of the device, enhancing the versatility of the method, and making it applicable to the detection of different types of diodes. Meanwhile, standardized judgment rules and calculation methods reduce reliance on human experience, adapt to automated testing processes, help improve the efficiency and consistency of batch testing, and provide reliable support for production quality control. Attached Figure Description

[0022] Figure 1 This is a flowchart of the diode conductivity testing method based on machine vision positioning in this embodiment; Figure 2 This is a flowchart illustrating the process of determining the diode conductivity detection using the machine vision-based positioning method in this embodiment. Figure 3 This is a flowchart illustrating the process of determining the actual number of inflection points in the diode conductivity testing method based on machine vision positioning in this embodiment. Figure 4 This is a flowchart illustrating the process of determining the diode damage type using the machine vision-based diode conductivity testing method in this embodiment. Detailed Implementation

[0023] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0024] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0026] Furthermore, it should be noted that, in the description of this invention, 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; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0027] Please see Figure 1 As shown, it is a flowchart of the diode conductivity testing method based on machine vision positioning in this embodiment; This embodiment provides a method for testing the conductivity of diodes based on machine vision positioning, including: Step S1: Obtain image data of the diode group based on the image detection module to obtain the diode group image; determine the arrangement spacing of the diode group based on the pin distance of any diode in the diode group image; and determine the arrangement direction of the diode group based on the pin markings of the diodes. Step S2: Determine the actual arrangement deviation of the diode group based on the arrangement spacing and arrangement direction; Step S3: Compare the actual arrangement deviation degree with the standard arrangement deviation degree threshold to obtain the arrangement deviation comparison result, and determine whether to perform diode conductivity performance testing based on the arrangement deviation comparison result; Step S4: Based on the condition that the arrangement deviation comparison result is qualified, the conductivity of the parallel diode group is tested based on the preset voltage to obtain the parallel current value. Based on the parallel current value and the parallel voltage value, a parallel volt-ampere characteristic curve is obtained. The parallel volt-ampere characteristic curve is segmented to obtain a segmented volt-ampere characteristic curve. The actual segment slope is determined based on the segmented volt-ampere characteristic curve. Step S5: Determine the actual number of inflection points based on the actual segmented slope and the parallel volt-ampere characteristic curve; Step S6: Determine the diode damage type and corresponding damage quantity based on the actual segment slope and the actual number of inflection points; Wherein, the pin distance is the average distance from any diode pin to its adjacent diode pin.

[0028] In step S1, this embodiment uses the diode array soldered on the PCB board of mass production for image acquisition, providing raw image data for subsequent appearance defect detection and location positioning. It uses a 1.3-megapixel industrial camera with a resolution of 1280×960, a frame rate of 20fps, and supports global shutter to avoid motion blur. It is equipped with an 8mm fixed-focus lens and a working distance of 30-50cm to ensure that the diode array size of about 5×10cm is fully imaged with clear details and the pin diameter of about 0.5mm can be distinguished. A ring-shaped white LED light source is installed at a 45° angle directly below the camera to avoid glare from the diode's metal pins, while uniformly illuminating the plastic casing to distinguish polarity markings. The PCB stage has positioning pins to ensure that the positional deviation is ≤0.5mm each time. The camera and light source are fixed on the gantry frame, and their relative positions remain unchanged.

[0029] By employing machine vision positioning technology and an image detection module, the spacing and orientation of diode arrays are precisely acquired, enabling automated and high-precision identification of the spatial arrangement of diode arrays. This provides a reliable positioning benchmark for subsequent conductivity performance testing, reducing detection errors caused by positional deviations. By comparing the actual arrangement deviation with the standard deviation to determine whether conductivity performance testing is necessary, situations that might affect testing accuracy or damage testing equipment due to abnormal arrangement can be preemptively eliminated, improving the stability and safety of the testing process and avoiding invalid testing operations. Combining segmented analysis of the volt-ampere characteristic curve, slope, and number of inflection points to determine the type and quantity of damage, visual positioning and electrical performance testing are integrated, achieving integrated testing from appearance positioning to performance evaluation. This ensures comprehensive testing while improving the accuracy of fault identification.

[0030] Specifically, the process of determining the arrangement spacing of the diode group based on the pin distance of any diode in the diode group image includes, Select any diode and determine its pin position to obtain the actual pin position. Based on the actual pin position, determine the pin distance between it and the pins of surrounding diodes. Based on the pin distance and the pin distance threshold, determine the arrangement spacing.

[0031] The diode group to be tested consists of through-hole silicon diodes soldered in a single row on a PCB board. Each diode has two metal leads, each about 8mm long and 0.5mm in diameter. The positive and negative leads are located on the left and right sides, respectively. The standard spacing between the leads on the same side of adjacent diodes, such as the right lead of the previous diode and the left lead of the next diode, is 5mm. The pre-processed grayscale image of the diode group has been obtained through the image detection module.

[0032] The third diode is randomly selected from the image as a reference to avoid errors caused by PCB boundary interference of edge diodes. The pin outline of the diode is extracted. The metal pin appears as a dark strip in the image, which has high contrast with the light PCB background. The pin edge is fitted by Hough line detection, and the midline of the two edges is taken as the pin centerline. The Hough line detection fitting of the pin edge is an existing technology and will not be described in detail here. The pin endpoints are determined using the coordinates of the two ends of the central axis, i.e., the image pixel coordinates: the left pin endpoint is (x1, y1), and the right pin endpoint is (x2, y2). Based on the image coordinate system, the x-axis is horizontal, consistent with the diode arrangement direction. Using the right pin of the reference diode as a reference, the left pin of the fourth diode adjacent to its right is found. Similarly, the endpoint coordinates (x3, y3) of the left pin of the fourth diode are extracted. The Euclidean distance between the right pin endpoint (x2, y2) and the fourth left pin endpoint (x3, y3) is calculated. Since the y-coordinate deviation is extremely small in a single column arrangement, it is approximately a horizontal distance. The pin distance d1 = |x3 - x2| is in pixels, converted to physical distance through camera calibration, 1 pixel ≈ 0.02 mm. Similarly, the distance d2 between the left pin of the diode and the right pin of the second diode to the left is calculated. A total of 3 sets of adjacent pin distances d1, d2, and d3 are obtained, covering 1-2 adjacent devices on each side of the diode. Based on the standard spacing, a preset pin distance threshold of 5mm is used, with an allowable error of ±0.3mm, i.e. 4.7mm-5.3mm, to filter out abnormal values ​​caused by pin bending or soldering misalignment. After converting the three sets of distances d1, d2, and d3 into physical distances, they are filtered out, and outliers exceeding the threshold are removed, such as a set where d3 = 5.5 mm. If the value exceeds the threshold, it is excluded. Take the average value of the remaining effective distance: if d1=4.9mm and d2=5.1mm, then the average pin distance is (4.9+5.1) / 2=5.0mm, that is, the arrangement spacing of the diode group is determined to be 5.0mm.

[0033] By selecting the diode in the middle position as a reference, the interference from PCB boundaries on edge devices is effectively avoided, providing a more reliable benchmark for subsequent pin positioning and reducing errors caused by boundary effects. Combining mature contour extraction and Hough line detection technology to locate pin edges, and determining the pin's centerline and endpoints through centerline calculation, fully utilizes the high contrast between the pin and the background, ensuring the accuracy of pin position identification and providing a reliable coordinate basis for distance calculation. By acquiring multiple sets of distances between adjacent pins and covering the devices on both sides of the reference diode, the random errors of single distance measurements can be effectively reduced. Setting reasonable thresholds to filter outomas can eliminate extreme data caused by pin bending, soldering misalignment, etc., further improving the effectiveness of distance data. Averaging the remaining effective distances results in a more representative arrangement spacing that truly reflects the overall layout pattern of the diode group.

[0034] Specifically, the process of determining the arrangement direction of a diode group based on the diode pin markings includes, The actual pin identifier is obtained by determining the pin identifier on any side of any diode based on the diode group image, and the arrangement direction is determined based on the actual pin identifier and the standard pin identifier.

[0035] Specifically, the process of determining the actual arrangement deviation of the diode group based on the arrangement spacing and arrangement direction includes, Regional diode group images are obtained by performing region segmentation based on diode group images. The regional diode group images are compared with standard diode group images to obtain feature points. The actual arrangement deviation of the diode group is determined based on the number of feature points. The feature points include points with abnormal spacing and points with abnormal pin markings.

[0036] In this embodiment, a group of 6 single-row through-hole diodes is combined on the PCB board. The standard arrangement direction is with the positive terminal facing left, and the positive terminal of each diode is marked with a white color ring. By segmenting by color threshold, the white ring is brighter than the outer shell in the grayscale image. The bright area near the diode pin is located and the closed outline is extracted and identified as the white ring of the positive electrode. Record the identification position. The color ring is located at the root of the left pin of the diode in the image coordinates. The x-coordinate of the left pin is smaller than that of the right pin, that is, the color ring is closer to the left pin. The positive lead with the white ring should be located on the left side of the diode, that is, the arrangement direction is positive on the left and negative on the right, and the entire diode group should be consistent along the single column direction with all positive terminals facing left; If the white ring of the diode is located on the left pin, and the color rings of the adjacent 1st and 3rd diodes are also located on the left, then the arrangement direction is determined to be in the same positive direction as the standard pin markings; if the color ring of the diode is located on the right pin, then the direction is determined to be reversed, i.e., abnormal arrangement. Based on the established arrangement spacing of 5mm, the overall diode group image is divided into 6 independent sub-regions along the arrangement direction. Each region corresponds to one diode, with a size of approximately 3×8mm, covering the two pins and the casing of a single diode. In the standard diode group image, the characteristic references for each sub-region are: the positive electrode mark is located on the left pin, and the center-to-center distance between adjacent sub-regions is 5mm with an error of ≤0.3mm. An abnormal spacing point was found. The center-to-center distance between the 4th and 5th diodes was measured to be 5.6mm, exceeding the 0.3mm threshold. This was marked as one abnormal spacing point. The pin markings indicate an abnormal point. Upon inspection, the white ring of the 6th diode was located on the right pin, which is inconsistent with the standard. This is marked as one abnormal point.

[0037] Please see Figure 2 This is a flowchart illustrating the process of determining the diode conductivity detection using the machine vision-based positioning method in this embodiment. Specifically, the process of determining whether to perform diode conductivity testing based on the arrangement deviation comparison results includes, If the actual arrangement deviation is less than the standard arrangement deviation threshold, the conductivity of the diode is tested. If the actual arrangement deviation is greater than or equal to the standard arrangement deviation threshold, the conductivity performance of the diode will not be tested.

[0038] In mass production, to avoid poor contact of the detection probes due to excessive diode arrangement deviations, such as short circuits caused by pin misalignment or probes failing to align with the pins, a preset standard arrangement deviation threshold of 20% is set. That is, when the proportion of abnormal points to the total number of diodes is ≤20%, the deviation is considered acceptable. If the actual layout deviation is less than the standard deviation threshold, conductivity testing is performed. After image detection, the feature points of a certain diode group are extracted as follows: A total of 6 diodes; Abnormal location: Only the spacing between the 3rd and 4th diodes is 5.4mm, exceeding the standard threshold of 0.3mm, but there is only 1 abnormal spacing location; Actual layout deviation: 1 / 6 ≈ 16.7% < 20% standard threshold; Since the actual deviation did not exceed the standard, it indicates that the pin position and arrangement of the diode group are controllable as a whole. The test probe can be aligned with the pin by fine adjustment. At this time, the conductivity test process is started: control the probe to move down and contact the positive and negative pins of each diode respectively, apply a preset positive voltage, and collect current data to generate the current-voltage characteristic curve. If the actual layout deviation is greater than or equal to the standard deviation threshold, conductivity testing will not be performed. The characteristic point detection results of another diode group are as follows: Total number of diodes: 6; Abnormal points: The second diode has one abnormal marking (reverse orientation); the fifth and sixth diodes are 6.0mm apart (an abnormal spacing), for a total of two abnormal points. Actual layout deviation: 2 / 6 ≈ 33.3% ≥ 20% of the standard threshold; If the actual deviation exceeds the standard, it indicates that some diode pins are misaligned or incorrectly oriented. Forcing testing may cause the probe to contact adjacent pins, leading to a short circuit if the spacing is too small, or incorrect voltage application due to reverse markings, such as applying reverse voltage to the forward pin, potentially damaging the diode or the testing equipment. In this case, conductivity testing will not be performed, and an alarm will be triggered, prompting manual review and adjustment.

[0039] When determining the arrangement direction based on pin markings, identifying characteristic markings such as white color rings and comparing them with the standard direction allows for accurate judgment of the overall consistency of the diode's positive and negative terminals. This avoids reverse voltage application during subsequent conductivity performance testing due to incorrect orientation, ensuring the fundamental correctness of electrical testing. When determining the actual degree of arrangement deviation, the diode group is divided into independent sub-regions through area segmentation. Spacing and abnormal points are extracted by comparing with standard images, and the quantified number of abnormalities reflects the degree of deviation. This makes the identification of arrangement problems more systematic and objective, avoiding the ambiguity of subjective judgment and providing a clear basis for subsequent decisions. The comparison between the degree of deviation and a threshold determines whether to perform conductivity performance testing. This ensures the normal progress of the testing process when the deviation is controllable, using fine-tuning to ensure precise probe contact and improve testing efficiency. Conversely, it promptly stops testing when the deviation is too large, preventing short circuits and incorrect voltage application caused by pin misalignment, protecting the diodes and testing equipment. Furthermore, alarm-triggered manual review reduces ineffective operations and potential losses in mass production.

[0040] The process of detecting the conductivity of a parallel diode group based on a preset voltage to obtain the parallel current value includes: applying a preset voltage to the parallel diode group and controlling the preset voltage to increase, while detecting the current change of the parallel diode group to obtain the parallel current value.

[0041] The process of verifying the arrangement direction based on the actual segmented slope includes: Determine the forward and reverse voltage ranges of the parallel current-voltage characteristic curve; Compare the actual segmented slope of the forward voltage range with the standard segmented slope range of forward conduction, and compare the actual segmented slope of the reverse voltage range with the standard segmented slope range of reverse cutoff. If the actual segmented slope of the forward voltage range is within the range of the standard segmented slope of the forward conduction and the actual segmented slope of the reverse voltage range is within the range of the standard segmented slope of the reverse cutoff, then the arrangement direction is consistent with the arrangement direction determined based on the pin identifier. If the actual segmented slope of the forward voltage range is within the range of the reverse cutoff standard segmented slope and the actual segmented slope of the reverse voltage range is within the range of the forward conduction standard segmented slope, then the arrangement direction is opposite to the arrangement direction determined based on the pin identifier, and the output arrangement direction verification result is abnormal.

[0042] In this embodiment, the test voltage range is set from -5V reverse to +5V forward. Forward voltage range: Voltage > 0.7V, which is higher than the conduction threshold, theoretically the diode should conduct in the forward direction; Reverse voltage range: When the voltage is <0V, theoretically the diode should be reverse-biased and cut off when a reverse voltage is applied. Based on the nominal characteristics and the number of diodes connected in parallel, 10 diodes are configured to be connected in parallel. In this embodiment, the standard slope range for forward conduction is 50-80mA / V. During forward conduction, the current increases rapidly with the voltage, and the slope is relatively large. When the reverse cutoff standard segment slope range is set to 0-2mA / V, the leakage current is extremely small, the current hardly changes with the voltage, and the slope is close to 0. Obtain the actual segmented slope, apply a preset voltage to the parallel diode group, measure the current-voltage characteristic curve, and then segment it. Therefore, the actual slope of the positive voltage range, for example, 1V-5V, is 65mA / V. The actual slope of the reverse voltage range, such as -5V to 0V, is 0.8mA / V. The actual slope of the forward voltage range of 65 mA / V falls within the standard forward conduction range of 50-80 mA / V, and the actual slope of the reverse voltage range of 0.8 mA / V falls within the standard reverse cutoff range of 0-2 mA / V. The arrangement direction is consistent with the left-facing positive terminal and left-facing white-marked pin side determined by image detection, verification passed; If the image detection misjudges the actual diode's positive terminal to be to the right, contrary to the label's judgment, the measurement will be... The forward voltage range is 1V-5V, and the actual applied reverse voltage has a segmented slope of 0.5mA / V that falls within the reverse cutoff standard range. The reverse voltage range is -5V to 0V, and the actual applied forward voltage has a segmented slope of 70mA / V that falls within the standard range for forward conduction. If the arrangement direction is opposite to the image detection result, the output arrangement direction verification is abnormal, and the pin identification needs to be reconfirmed through image detection.

[0043] Cross-validation of diode electrical characteristics and visual positioning results significantly improves the accuracy and reliability of orientation determination, effectively compensating for potential limitations in machine vision recognition. In image detection, factors such as blurred pin markings and lighting interference can easily lead to orientation misjudgment. The inherent unidirectional conductivity of diodes causes significant differences in the slope characteristics between forward conduction and reverse cutoff. By comparing the actual segmented slope with the standard range, the visual positioning results can be objectively verified, reducing errors from single recognition methods. Orientation directly affects the interpretation of the current-voltage characteristic curve; orientation misjudgment can lead to distortion in curve feature analysis, such as mistaking reverse characteristics for forward characteristics. By inferring the correctness of orientation through electrical characteristics, subsequent slope analysis, inflection point counting, and damage assessment are ensured to be based on correct electrical logic. This improves the rigor and reliability of the overall testing process, reduces invalid detections or misjudgments caused by orientation errors, and enhances the practicality of the method in industrial scenarios.

[0044] Specifically, the process of determining the actual number of inflection points based on the actual piecewise slope and the parallel volt-ampere characteristic curve includes, Select any actual segment slope as the actual segment slope of this segment, determine the actual segment slopes of its adjacent upper and lower segments based on the actual segment slope of this segment, determine the absolute value of the first slope difference based on the actual segment slope of this segment and the actual segment slope of the upper segment, determine the absolute value of the second slope difference based on the actual segment slope of this segment and the actual segment slope of the lower segment, and determine the actual number of inflection points based on the absolute value of the difference and the absolute value threshold of the difference.

[0045] Please see Figure 3 This is a flowchart illustrating the process of determining the actual number of inflection points in the diode conductivity testing method based on machine vision positioning in this embodiment. Specifically, the process of determining the actual number of inflection points based on the absolute value of the difference and the threshold value of the difference includes: If the absolute value of the difference is less than the absolute value threshold of the difference, then the actual segment slope of this segment does not change significantly from the actual segment slope of the adjacent segment, and there is no inflection point between this segment and the adjacent segment. If the absolute value of the difference is greater than or equal to the threshold of the absolute value of the difference, then there is a significant change between the actual segment slope of this segment and the actual segment slope of the adjacent segment. There is an inflection point between this segment and the adjacent segment, and this inflection point is included in the actual number of inflection points.

[0046] After conductivity testing, the parallel current-voltage characteristic curve of this diode group was divided into four consecutive segments. The actual slope of each segment, in mA / V, is as follows: In the reverse low voltage region of segment A: kA=0.05, the reverse leakage current is extremely small, and the slope is close to 0; When segment B conducts forward through the transition region: kB=8.2, after the forward voltage reaches the conduction threshold, the current rises sharply and the slope increases abruptly. In the forward stable conduction region of segment C: kC=7.9, the current increases gradually with the voltage, and the slope is slightly lower than that of segment B; Segment D forward high voltage abnormal region: kD=0.3 After a certain diode breaks down, the current increase suddenly stops and the slope drops sharply; In this embodiment, the preset absolute value threshold for the difference is 3. Based on the characteristics of silicon diodes, when the slope difference between adjacent segments exceeds this value, it is determined to be an inflection point of abrupt change in characteristics. Using segment B as the actual slope of this segment, The adjacent upper segment is segment A, and the lower segment is segment C; The absolute value of the first slope difference: |kB-kA|=|8.2-0.05|=8.15; The absolute value of the second slope difference: |kB-kC|=|8.2-7.9|=0.3; Threshold judgment: 8.15≥3, then there is an inflection point between segment A and segment B; 0.3<3, then there is no inflection point between segment B and segment C. Therefore, segment B contributes 1 inflection point. Let segment C be the actual slope of this segment. The adjacent upper segment is segment B, and the lower segment is segment D; The absolute value of the first slope difference: |kC-kB|=|7.9-8.2|=0.3; The absolute value of the second slope difference: |kC-kD|=|7.9-0.3|=7.6; Threshold judgment: 7.6≥3, then there is an inflection point between segment C and segment D, and segment C contributes 1 inflection point; Segment A is the starting segment with no upper segment, and segment D is the ending segment with no lower segment. No additional calculation is needed. The final cumulative number of actual inflection points is 2. Under normal circumstances, this diode group should only have 1 inflection point from segment A to segment B. The fact that 2 inflection points actually occurred indicates that the abrupt change in slope from segment C to segment D caused an abnormality that led to the forward breakdown of a certain diode.

[0047] Continuous voltage increases comprehensively cover the entire operating range of the diode from cutoff to conduction, ensuring the capture of current change characteristics under different voltages and avoiding omissions of characteristics due to voltage point dispersion. For overall testing of parallel diode groups, it directly reflects the combined conductivity characteristics of the assembly, eliminating the need for individual testing and aggregation, significantly improving testing efficiency, especially suitable for mass production scenarios. Synchronous acquisition of voltage and current ensures data consistency, providing continuous and complete raw data for subsequent plotting of the volt-ampere characteristic curve, ensuring the accuracy of the curve shape. Standardized control of preset voltage reduces the randomness of human operation, making test data from different batches and groups comparable, providing a unified benchmark for subsequent fault analysis. By selecting the actual slope of each segment, calculating the absolute value of the slope difference with adjacent upper and lower segments, and combining it with preset thresholds to determine the existence of inflection points, precise capture of abrupt changes in the volt-ampere characteristic curve is achieved. This segment-by-segment analysis method systematically covers all intervals of the curve, avoiding omission of any possible characteristic abrupt changes and ensuring the comprehensiveness and accuracy of the statistical count of inflection points. By comparing the absolute value of the difference with a threshold as the criterion, the fuzzy concept of significant change is transformed into a quantifiable standard, reducing errors from subjective judgment and making inflection point identification more objective and consistent, thus meeting the needs of automated detection. By accumulating the actual number of inflection points and comparing it with the standard number of inflection points under normal conditions, abnormal abrupt changes in the curve, such as extra inflection points, can be intuitively identified. These abnormal inflection points directly correspond to diode performance abnormalities such as breakdown, providing clear characteristic clues for subsequent analysis of damage types and making fault diagnosis more targeted.

[0048] Please see Figure 4 This is a flowchart illustrating the process of determining the diode damage type using the diode conductivity testing method based on machine vision positioning in this embodiment. Specifically, the process of determining the diode damage type and corresponding damage quantity based on the actual segmented slope and the actual number of inflection points includes, If the actual number of inflection points is 0, and there exists any actual segment slope that is greater than the maximum value of the standard slope range of the forward conduction area, then it is determined that there is a short-circuited damaged diode. If the actual number of inflection points is greater than the standard number of inflection points, and there is at least one segment where the actual slope is less than the minimum value of the standard slope range of the forward cutoff region, then it is determined that there is an open-circuit damaged diode. If the actual number of inflection points is equal to the standard number of inflection points, and the actual segmented slope is not within its corresponding slope range, then the diode is determined to have performance degradation. The standard slope range of the forward conduction region is the range when the corresponding voltage is higher than the conduction threshold; the standard slope range of the forward cutoff region is the range when the corresponding voltage is lower than the conduction threshold; the conduction threshold voltage is determined based on the material of the test diode.

[0049] Specifically, the process of determining the diode damage type and corresponding damage quantity based on the actual segment slope and the actual number of inflection points also includes: Based on the diodes with short-circuit damage, the short-circuit slope increment is determined based on the slope contribution value of a single short-circuit diode under forward voltage. Based on the slope difference between the actual segmented slope and the maximum value of the standard slope range in the forward conduction region, the number of short-circuit damaged diodes is determined based on the short-circuit slope increment and the slope difference. Based on the open-circuit damaged diodes, obtain the standard current value of normal parallel diodes under the same voltage, determine the current difference based on the actual current value and the standard current value, use the conduction current of a single normal diode as the preset single-tube current, and determine the number of open-circuit damaged diodes based on the current difference and the preset single-tube current. Based on diodes with performance degradation, a slope deviation threshold for a single performance degradation diode is set. The total deviation is determined based on the actual segmented slope and the standard slope. The number of performance degradation diodes is determined based on the total deviation and the slope deviation threshold.

[0050] In this embodiment, the six parallel silicon diodes are configured with a conduction threshold of 0.7V and a standard inflection point of 1. The standard slope range of the forward conduction region is 4-7 mA / V, corresponding to a voltage > 0.7V; the standard slope range of the forward cutoff region is 0-0.3 mA / V, corresponding to a voltage < 0.7V. The actual number of inflection points is 0, with no abrupt change in slope from cutoff to conduction. Furthermore, in the voltage > 0.7V region, the actual segmented slope is 10mA / V, exceeding the standard maximum value of 7mA / V in the forward conduction region. This indicates a short-circuit damaged diode. A normal diode has a clear inflection point between forward cutoff and conduction (one inflection point). At cutoff, the slope is extremely small with low leakage current. After conduction, the slope increases significantly, and the current rises rapidly with the voltage. If the diode is short-circuited, its unidirectional conductivity is completely lost. Regardless of whether the voltage exceeds the conduction threshold, it will exhibit a low-resistance conduction state. This causes the current-voltage characteristic curve to lose the abrupt change from cutoff to conduction, resulting in 0 inflection points. Moreover, the slope in the conduction region is much larger than the normal range due to the extremely small short-circuit resistance. Therefore, the combination of 0 inflection points and an excessively high slope in the conduction region directly corresponds to the core characteristics of a short circuit. The actual number of inflection points is 2, which is greater than the standard number of inflection points of 1. Furthermore, in the voltage < 0.7V region, a certain segment's actual slope is -0.1mA / V, which is less than the standard minimum value of 0 in the forward cutoff region. This indicates the presence of open-circuit damaged diodes. An open-circuit diode is equivalent to an open circuit, leading to abnormal overall conductivity in the parallel circuit. Normally, the volt-ampere characteristic of a parallel diode group should exhibit a smooth transition with increasing voltage, showing a single-inflection-point characteristic. If an open circuit exists, the failure of some diodes will cause abrupt segmental changes in the overall characteristics. For example, in a certain voltage range, the effective conduction count may suddenly decrease, causing the slope to drop abruptly, resulting in an increased number of inflection points exceeding the standard value. Simultaneously, in the cutoff region, where the voltage is below the conduction threshold, the leakage current of a normal diode is extremely low, and the slope is close to 0. However, an open circuit further weakens conductivity, even showing a negative deviation from the standard value, with the slope falling below the standard minimum value for the cutoff region. Therefore, the combination of an excessive number of inflection points and an abnormal cutoff region slope accurately reflects the characteristic disorder caused by an open circuit. The actual number of inflection points is 1, which equals the standard number of inflection points. However, in the voltage > 0.7V region, the actual segmented slope is 2mA / V, which is outside the standard range of 4-7mA / V for the forward conduction region. This indicates a diode with performance degradation. A diode with performance degradation is not completely failed and still retains the basic characteristics of cutoff and conduction. Therefore, the number of inflection points is consistent with the standard, but its conduction capability decreases due to increased internal resistance. This causes the current in the conduction region to increase more slowly with voltage, and the slope is less than the normal range or abnormally increased due to aging. The core issue is deviation from the standard range. Therefore, the combination of a normal number of inflection points and a slope outside the corresponding range not only distinguishes between incompletely damaged states but also quantifies the degree of performance degradation through slope deviation, consistent with the characteristics of degradation.

[0051] The slope contribution of a single short-circuited diode under forward voltage is calibrated to be 3mA / V. Under normal conditions, the single-diode conduction slope is approximately 1mA / V. After short-circuiting, the single-diode slope increases to approximately 4mA / V, with an increment of 3mA / V. The difference between the actual segmented slope of 10 mA / V and the standard maximum slope of 7 mA / V in the forward conduction zone is 10 - 7 = 3 mA / V; The number of short-circuit damaged units is equal to the slope difference / short-circuit slope increment, which equals 1 unit. Under the same forward voltage of 1V, the standard current value of 6 normally connected diodes in parallel is 12mA, and the conduction current of a single normally connected diode is 2mA. That is, the preset single diode current is 2mA, and the actual measured current value is 8mA. The current difference is the standard current value minus the actual current value = 12 - 8 = 4mA; The number of open-circuit damaged units is calculated as current difference / preset single-tube current = 2. The slope deviation threshold for a single performance decay diode is set to 1 mA / V. Under normal conditions, the slope of a single diode is 1 mA / V. After decay, the slope of a single diode drops to 0 mA / V, resulting in a single-diode deviation of 1 mA / V. The total deviation between the actual segmented slope of 2 mA / V and the standard minimum value of 4 mA / V in the forward conduction region is 4 - 2 = 2 mA / V. The number of units with performance degradation is the total deviation / single tube slope deviation threshold = 2.

[0052] In determining the damage type, a multi-dimensional fault identification system is constructed through the coordinated judgment of the number of inflection points and the range of segmented slopes. Diodes with short circuits, open circuits, and performance degradation exhibit distinctly different current-voltage characteristics: short circuits have no inflection points and the slope exceeds the limit; open circuits have multiple inflection points and the slope in the cutoff region is abnormal; and degradation has normal inflection points but the slope deviates. This targeted judgment logic can accurately distinguish different damage types, avoiding misjudgments caused by single-feature judgments and ensuring the accuracy of fault nature identification. In determining the number of damaged diodes, based on the characteristic contribution values ​​of a single diode, such as the short-circuit slope increment, single-diode conduction current, and slope deviation threshold, the deviation of the overall electrical characteristics is transformed into a quantifiable number of damaged diodes. This deepens the detection results from qualitative judgment to quantitative data, providing a concrete basis for fault screening, repair, and replacement in mass production, and enhancing the practical value of the detection. The entire process fully integrates the material characteristics of the diodes, such as the standard range for setting the conduction threshold of silicon diodes, making the judgment logic compatible with the inherent properties of the device, enhancing the versatility of the method, and making it applicable to the detection of different types of diodes. Meanwhile, standardized judgment rules and calculation methods reduce reliance on human experience, adapt to automated testing processes, help improve the efficiency and consistency of batch testing, and provide reliable support for production quality control.

[0053] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for testing the conductivity of diodes based on machine vision positioning, characterized in that, include, The diode group image is obtained by acquiring image data of the diode group based on the image detection module. The arrangement spacing of the diode group is determined based on the pin distance of any diode in the diode group image. The arrangement direction of the diode group is determined based on the pin markings of the diodes. The actual arrangement deviation of the diode group is determined based on the arrangement spacing and arrangement direction. Compare the actual arrangement deviation with the standard arrangement deviation threshold to obtain the arrangement deviation comparison result, and determine whether to perform diode conductivity testing based on the arrangement deviation comparison result. If the arrangement deviation comparison results are satisfactory, the conductivity of the parallel diode group is tested based on a preset voltage to obtain the parallel current value. Based on the parallel current value and the parallel voltage value, a parallel volt-ampere characteristic curve is obtained. The parallel volt-ampere characteristic curve is segmented to obtain a segmented volt-ampere characteristic curve. The actual segment slope is determined based on the segmented volt-ampere characteristic curve. The arrangement direction is verified based on the actual segment slope. The actual number of inflection points is determined based on the actual segmented slope and the parallel volt-ampere characteristic curve. The diode damage type and corresponding damage quantity are determined based on the actual segment slope and the actual number of inflection points. Wherein, the pin distance is the average distance from any diode pin to its adjacent diode pin.

2. The diode conductivity testing method based on machine vision positioning according to claim 1, characterized in that, The process of determining the arrangement spacing of the diode group based on the pin distance of any diode in the diode group image includes, Select any diode and determine its pin position to obtain the actual pin position. Based on the actual pin position, determine the pin distance between it and the pins of surrounding diodes. Based on the pin distance and the pin distance threshold, determine the arrangement spacing.

3. The diode conductivity testing method based on machine vision positioning according to claim 2, characterized in that, The process of determining the arrangement direction of a diode group based on the diode pin markings includes, The actual pin identifier is obtained by determining the pin identifier on any side of any diode based on the diode group image, and the arrangement direction is determined based on the actual pin identifier and the standard pin identifier.

4. The diode conductivity testing method based on machine vision positioning according to claim 3, characterized in that, The process of determining the actual arrangement deviation of the diode group based on the arrangement spacing and arrangement direction includes, Regional diode group images are obtained by performing region segmentation based on diode group images. The regional diode group images are compared with standard diode group images to obtain feature points. The actual arrangement deviation of the diode group is determined based on the number of feature points. The feature points include points with abnormal spacing and points with abnormal pin markings.

5. The diode conductivity testing method based on machine vision positioning according to claim 4, characterized in that, The process of determining whether to perform diode conductivity testing based on the arrangement deviation comparison results includes the following steps. If the actual arrangement deviation is less than the standard arrangement deviation threshold, the conductivity of the diode is tested. If the actual arrangement deviation is greater than or equal to the standard arrangement deviation threshold, the conductivity performance of the diode will not be tested.

6. The diode conductivity testing method based on machine vision positioning according to claim 5, characterized in that, The process of verifying the arrangement direction based on the actual segmented slope includes: Determine the forward and reverse voltage ranges of the parallel current-voltage characteristic curve; Compare the actual segmented slope of the forward voltage range with the standard segmented slope range of forward conduction, and compare the actual segmented slope of the reverse voltage range with the standard segmented slope range of reverse cutoff. If the actual segmented slope of the forward voltage range is within the range of the standard segmented slope of the forward conduction and the actual segmented slope of the reverse voltage range is within the range of the standard segmented slope of the reverse cutoff, then the arrangement direction is consistent with the arrangement direction determined based on the pin identifier. If the actual segmented slope of the forward voltage range is within the range of the reverse cutoff standard segmented slope and the actual segmented slope of the reverse voltage range is within the range of the forward conduction standard segmented slope, then the arrangement direction is opposite to the arrangement direction determined based on the pin identifier, and the output arrangement direction verification result is abnormal.

7. The diode conductivity testing method based on machine vision positioning according to claim 6, characterized in that, The process of determining the actual number of inflection points based on the actual piecewise slope and the parallel volt-ampere characteristic curve includes the following steps: Select any actual segment slope as the actual segment slope of this segment, determine the actual segment slopes of its adjacent upper and lower segments based on the actual segment slope of this segment, determine the absolute value of the first slope difference based on the actual segment slope of this segment and the actual segment slope of the upper segment, determine the absolute value of the second slope difference based on the actual segment slope of this segment and the actual segment slope of the lower segment, and determine the actual number of inflection points based on the absolute value of the difference and the absolute value threshold of the difference.

8. The diode conductivity testing method based on machine vision positioning according to claim 7, characterized in that, The process of determining the actual number of inflection points based on the absolute value of the difference and the threshold value of the difference includes: If the absolute value of the difference is less than the absolute value threshold of the difference, then the actual segment slope of this segment does not change significantly from the actual segment slope of the adjacent segment, and there is no inflection point between this segment and the adjacent segment. If the absolute value of the difference is greater than or equal to the threshold of the absolute value of the difference, then there is a significant change between the actual segment slope of this segment and the actual segment slope of the adjacent segment. There is an inflection point between this segment and the adjacent segment, and this inflection point is included in the actual number of inflection points.

9. The diode conductivity testing method based on machine vision positioning according to claim 8, characterized in that, The process of determining the diode damage type and corresponding damage quantity based on the actual segmented slope and the actual number of inflection points includes: If the actual number of inflection points is 0, and there exists any actual segment slope that is greater than the maximum value of the standard slope range of the forward conduction area, then it is determined that there is a short-circuited damaged diode. If the actual number of inflection points is greater than the standard number of inflection points, and there is at least one segment where the actual slope is less than the minimum value of the standard slope range of the forward cutoff region, then it is determined that there is an open-circuit damaged diode. If the actual number of inflection points is equal to the standard number of inflection points, and the actual segmented slope is not within its corresponding slope range, then the diode is determined to have performance degradation. The standard slope range of the forward conduction region is the range when the corresponding voltage is higher than the conduction threshold; the standard slope range of the forward cutoff region is the range when the corresponding voltage is lower than the conduction threshold; the conduction threshold voltage is determined based on the material of the test diode.

10. The diode conductivity testing method based on machine vision positioning according to claim 9, characterized in that, The process of determining the diode damage type and corresponding damage quantity based on the actual segment slope and the actual number of inflection points also includes: Based on the diodes with short-circuit damage, the short-circuit slope increment is determined based on the slope contribution value of a single short-circuit diode under forward voltage. Based on the slope difference between the actual segmented slope and the maximum value of the standard slope range in the forward conduction region, the number of short-circuit damaged diodes is determined based on the short-circuit slope increment and the slope difference. Based on the open-circuit damaged diodes, obtain the standard current value of normal parallel diodes under the same voltage, determine the current difference based on the actual current value and the standard current value, use the conduction current of a single normal diode as the preset single-tube current, and determine the number of open-circuit damaged diodes based on the current difference and the preset single-tube current. Based on diodes with performance degradation, a slope deviation threshold for a single performance degradation diode is set. The total deviation is determined based on the actual segmented slope and the standard slope. The number of performance degradation diodes is determined based on the total deviation and the slope deviation threshold.

Images