Lead frame quality detection method, device and equipment
Through multi-angle image acquisition and comprehensive detection technology, the problems of slow speed and low accuracy of traditional detection methods are solved, and the comprehensive quality evaluation of the lead frame is achieved, which improves the speed and accuracy of detection.
Patent Information
- Application Number
- CN202411130317.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-12
AI Technical Summary
Traditional lead frame detection methods are slow, have low accuracy, difficult to identify minor defects and quality deviations, and cannot comprehensively evaluate the quality of all aspects of the lead frame.
Using multi-angle image acquisition of industrial cameras, combined with image processing algorithms, machine learning models, optical measurement technology and atomic absorption spectroscopy, the geometric shape, size, surface texture, material composition and electrical properties of the lead frame are detected, and the weighted summing method is carried out. Comprehensive assessment.
It improves the detection speed and accuracy, can accurately identify small defects and quality deviations, meet the needs of large-scale production, provide comprehensive quality assessment, and improves the fairness and reliability of inspections.
Smart Images

Figure CN118937338A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electronic component quality inspection, and in particular to a lead frame quality inspection method, device and equipment. Background Art
[0002] The lead frame plays a vital role in integrated circuits. It not only provides a stable support structure for the chip, but also realizes the effective connection between the chip and the external circuit. It can be said that the lead frame is an indispensable component of integrated circuits. In recent years, the electronics industry has shown a rapid development trend. The integration of integrated circuits continues to increase, the functions are becoming more and more complex, and the application scenarios are becoming more and more extensive, from consumer electronics to industrial control, from communication equipment to aerospace and other fields. This series of developments and changes have put forward higher requirements on the performance, reliability and stability of integrated circuits, and thus have significantly improved the quality standards of lead frames as their key components.
[0003] However, traditional lead frame inspection methods have gradually exposed many shortcomings: 1. Traditional testing often relies on manual operation or simple instruments and equipment, with slow testing speed, which is difficult to meet the needs of large-scale production. In terms of accuracy, due to the limitations of testing methods and interference from human factors, it is easy to make misjudgments or miss judgments, and it is impossible to accurately identify minor defects and quality deviations; Second, traditional methods may only focus on certain specific indicators or test items and fail to cover all aspects of lead frame quality, such as complex surface defects, fine dimensional parameters, internal structure of materials and electrical properties, making it difficult to provide a comprehensive and accurate quality assessment.
[0004] To this end, a lead frame quality inspection method, device and equipment are proposed. Summary of the invention
[0005] In view of this, the embodiments of the present invention hope to provide a lead frame quality inspection method, device and equipment to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.
[0006] In order to solve the above technical problems, a technical solution adopted in the present application is: a lead frame quality detection method, comprising the following steps: Step 1: Use an industrial camera to capture images of the lead frame from multiple angles to obtain appearance images of the lead frame at multiple angles; Step 2: performing denoising and enhancement processing on the appearance images of the lead frame at multiple angles obtained; Step 3: Using image processing algorithms to extract the geometric shape, size, and surface texture features of the lead frame; Step 4: Based on the extracted geometric shape, size, and surface texture features of the lead frame, use machine learning or deep learning models to detect scratches, dents, protrusions, and oxidation defects on the surface of the lead frame, and compare them with preset standard values to obtain defect scores; Step 5: Use optical measurement technology to measure the key dimensions of the lead frame and compare them with the preset standard values to obtain the dimension deviation score; Step 6: Use atomic absorption spectroscopy or inductively coupled plasma emission spectroscopy to perform quantitative analysis based on the relationship between the intensity of characteristic spectral lines and element concentrations, and compare with preset standard values to obtain material analysis scores; Step 7: Connect the test circuit, measure the resistance, capacitance, and inductance parameters of the lead frame, and compare them with the preset standard values to obtain the electrical performance score; Step 8. Integrate the defect score, dimensional deviation score, material analysis score and electrical performance score, conduct a comprehensive evaluation of the quality of the lead frame according to the preset quality standards, obtain a comprehensive evaluation score, and divide the quality grade according to the comprehensive evaluation score of the lead frame.
[0007] As a further preferred embodiment of the present technical solution, in step 2, the denoising process uses a median filtering algorithm, and its formula is:
[0008] in, is the pixel value after filtering, is the original pixel value, is the filter window; The enhancement process uses a histogram equalization algorithm, and its probability density function is:
[0009] in, is the gray level, is the number of pixels of this gray level, is the total number of image pixels.
[0010] As a further preferred embodiment of the present technical solution, in step three, the geometric shape of the lead frame is extracted using an edge detection algorithm, and the edge is determined by calculating the gradient amplitude and direction; the size of the lead frame is extracted using a minimum circumscribed rectangle algorithm, and the length and width of the rectangle are calculated to represent the size of the lead frame; the surface texture features of the lead frame are extracted using a grayscale co-occurrence matrix.
[0011] As a further preferred embodiment of the present technical solution, in step 4, the machine learning model uses a decision tree algorithm, and its information gain formula is:
[0012] in, For the dataset The information entropy of Characterized by Features The value is subset of data.
[0013] As a further preferred embodiment of the present technical solution, in step 5, the key dimensions include length, width and pin spacing; the length and width are measured by structured light measurement, and the three-dimensional coordinates of the surface points of the object are calculated according to the triangulation principle to obtain the length and width dimensions; the pin spacing is measured using an image sub-pixel algorithm; Dimensional deviation The calculation formula is:
[0014] in, is the measured value, is the standard value.
[0015] As a further preferred embodiment of the present technical solution, in step seven, the resistance of the lead frame By voltammetry:
[0016] in, is the voltage, is the electric current; The capacitance and inductance of the lead frame are measured using an LCR tester and calculated according to the test frequency and the measured value.
[0017] As a further preferred embodiment of the present technical solution, in step eight, the comprehensive evaluation score Using the weighted summation method, the formula is:
[0018] in, , , , They are defect score, deviation score, material analysis score and electrical performance score. , , , are the weights of defect score, deviation score, material analysis score and electrical performance score, respectively, and ; The quality grades include four grades: excellent, good, qualified and unqualified.
[0019] To solve the above technical problems, another technical solution adopted in the present application is: a lead frame quality inspection device, comprising: an image acquisition module, an image processing module, a dimension measurement module, a material analysis module, an electrical performance test module and a data integration and evaluation module; The image acquisition module includes a plurality of industrial cameras for acquiring appearance images of the lead frame from different angles; The image processing module is used to perform denoising, enhancement, feature extraction and defect detection on the collected images to obtain defect scores; The dimension measurement module includes an optical measuring instrument and a related algorithm for measuring the key dimensions of the lead frame and obtaining a dimension deviation score; The material analysis module is used to analyze the material composition of the lead frame and obtain a material analysis score; The electrical performance test module is used to measure the resistance, capacitance and inductance parameters of the lead frame to obtain an electrical performance score; The data integration and evaluation module is used to integrate the defect score, deviation score, material analysis score and electrical performance score, and to perform comprehensive evaluation scores and divide quality grades according to quality standards.
[0020] In order to solve the above technical problems, another technical solution adopted by the present application is: an electronic device, the electronic device comprising: Processor; and A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to perform the above-mentioned lead frame quality detection method by executing the executable instructions.
[0021] In order to solve the above technical problem, another technical solution adopted by the present application is: the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned lead frame quality detection method is implemented.
[0022] The embodiment of the present invention has the following advantages due to the adoption of the above technical solution: 1. The present invention uses multi-angle image acquisition with industrial cameras, combines advanced image processing algorithms, automated detection processes, precise optical measurement technology, advanced spectral analysis methods, and machine learning and deep learning models for defect detection, which greatly improves the detection speed, reduces human interference, and can accurately identify tiny defects and quality deviations, meeting the needs of large-scale production while improving detection accuracy; 2. The present invention covers multiple aspects of lead frame detection, including geometry, size, surface texture, material composition and electrical properties. It uses machine learning and deep learning models to continuously learn and optimize detection standards, adapt to lead frame detection of different types and specifications, and can comprehensively evaluate quality, with stronger adaptability and intelligence. 3. The present invention objectively and uniformly evaluates the quality of the lead frame by presetting clear quality standards and evaluation systems, avoiding the inconsistency caused by subjective judgment in traditional methods, improving the fairness and reliability of quality evaluation, and at the same time, being able to record and store the detection process and results in detail, facilitating quality traceability and problem analysis, and contributing to continuous improvement of production processes and product quality.
[0023] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0025] Figure 1 A schematic diagram of a process of a lead frame quality inspection method of the present invention; Figure 2 It is a functional module schematic diagram of a lead frame quality inspection device of the present invention; Figure 3 The figure is a schematic structural diagram of an electronic device of the present invention. DETAILED DESCRIPTION
[0026] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.
[0027] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.
[0028] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.
[0029] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.
[0030] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.
[0031] Figure 1 1 is a flow chart of a lead frame quality inspection method according to an embodiment of the present invention. It should be noted that if there are substantially the same results, the method of the present application is not limited to Figure 1 The process sequence shown is limited. Figure 1 As shown: A lead frame quality inspection method, comprising the following steps: Step 1: Use an industrial camera to capture images of the lead frame from multiple angles to obtain appearance images of the lead frame from multiple angles. There are multiple industrial cameras, and the multiple angles include the front, side, top, etc., to ensure that all key areas of the lead frame are covered; Step 2: De-noising and enhancing the acquired appearance images of the lead frame at multiple angles, wherein: The denoising process uses the median filter algorithm. For each pixel point, the median of the pixel values in the surrounding window is taken as the pixel value after filtering to reduce noise interference; Enhancement processing uses the histogram equalization algorithm to adjust the contrast of the image to make the image clearer and easier to process later; Step 3: Use image processing algorithms to extract the geometric shape, size, and surface texture features of the lead frame, where: The geometry uses edge detection algorithms (such as the Canny edge detector) to identify the boundaries of the lead frame and extract its shape by calculating the geometric features of the boundary (such as straight line segments, circular arcs, etc.); The minimum enclosing rectangle algorithm is used to determine the minimum enclosing rectangle of the lead frame, and its length and width are calculated as the size feature. The pin spacing is accurately measured using the image sub-pixel algorithm. Surface texture uses gray-level co-occurrence matrix (GLCM) to analyze the gray-level dependency between pixels in an image and their neighboring pixels and extract surface texture features; Step 4: Based on the extracted geometry, size, and surface texture features of the lead frame, use machine learning or deep learning models to detect scratches, dents, protrusions, and oxidation defects on the surface of the lead frame, and compare them with the preset standard values to obtain defect scores. Specifically: Take the extracted geometric shape, size and surface texture features as input data, prepare a sample data set containing various known defects (scratches, dents, protrusions, oxidation, etc.), these samples should be labeled with the defect type and its severity, select a suitable machine learning or deep learning model, for complex image recognition tasks, deep learning models (such as convolutional neural networks CNN) usually perform better, use the labeled sample data set to train the model, during the training process, the model will learn how to identify different types of defects based on the input features, use the sample data set that did not participate in the training to evaluate the model to test its accuracy and generalization ability in identifying defects, adjust and optimize the model according to the evaluation results to improve its performance; The trained model is applied to the lead frame image to be inspected. The model will automatically analyze the features in the image and identify possible defects. For each detected defect, the model may output a confidence score or probability value to indicate the possibility of the defect. Based on the output of the model, a score is assigned to each detected defect. This score can be determined based on the severity, area, number or confidence of the model. These defect scores are compared with preset standard values. The standard value can be a threshold set based on product specifications, industry standards or customer requirements. If the score of a defect exceeds the corresponding standard value threshold, the defect is considered unacceptable and the lead frame may need to be further processed or eliminated. All detected defects and their scores are recorded for subsequent analysis and traceability; Step 5: Use optical measurement technology to measure the key dimensions of the lead frame and compare them with the preset standard values to obtain the dimension deviation score. Optical measurement technology is a non-contact measurement method with the advantages of high precision, high efficiency, and non-destructiveness. It is very suitable for the dimension measurement of tiny parts such as lead frames. Among the many optical measurement technologies, we can choose the following: Laser rangefinder: uses a laser beam to measure the distance of an object surface, and determines the size by calculating the round-trip time or phase difference of the laser. This method is suitable for measuring longer distances or when a quick response is required.
[0032] Structured light measuring instrument: It measures the size by projecting a specific light pattern (such as stripes, gratings, etc.) onto the object to be measured and observing its deformation or displacement. This method is suitable for measuring complex shapes and surfaces.
[0033] Microscope measurement system: Combine high-magnification microscope and image processing technology to accurately measure tiny objects. This method is suitable for occasions that require high-precision measurement; Step 6: Using atomic absorption spectrometry or inductively coupled plasma emission spectrometry, quantitative analysis is performed based on the relationship between the intensity of characteristic spectral lines and element concentrations, and the results are compared with preset standard values to obtain a material analysis score, wherein atomic absorption spectrometry or inductively coupled plasma emission spectrometry is used to perform element content analysis on the lead frame sample, and the content of a specific element is calculated based on the relationship between the intensity of characteristic spectral lines and element concentrations; Step 7. Connect the test circuit, measure the resistance, capacitance, and inductance parameters of the lead frame, and compare them with the preset standard values to obtain the electrical performance score. The preset standard values of the resistance, capacitance, and inductance parameters of the lead frame can be obtained from the product design specification, industry standards, or customer requirements. Compare the measured resistance, capacitance, and inductance values with the preset standard values, calculate the deviation between the measured value and the standard value of each parameter (which can be an absolute deviation or a relative deviation), and set the scoring criteria according to the size and importance of the deviation. For example, different deviation ranges can be set to correspond to different scoring intervals. The electrical performance score of the lead frame is calculated according to the scoring criteria. The score can be calculated based on the deviation of a single parameter, or it can be weighted by comprehensively considering the deviations of multiple parameters. Step 8. Integrate the defect score, deviation score, material analysis score and electrical performance score, conduct a comprehensive evaluation of the quality of the lead frame according to the preset quality standards, obtain a comprehensive evaluation score, and divide the quality level according to the comprehensive evaluation score of the lead frame. The comprehensive evaluation score uses the weighted summation method to assign different weights to each score according to its importance in the comprehensive evaluation. Then, add these weighted scores to get the comprehensive evaluation score. The allocation of weights should be based on product design requirements, industry standards, customer expectations, and the degree of influence of various performance factors on the overall quality.
[0034] In one embodiment, specifically, in step 2, the denoising process uses a median filtering algorithm, and its formula is:
[0035] in, is the pixel value after filtering, is the original pixel value, is the filter window; Median filtering is a nonlinear filtering technique that is often used to remove noise in images or signals, especially salt-and-pepper noise. The principle of median filtering is: for each pixel in the image, take the median of all pixel values in the neighborhood centered on it, and assign this median to the current pixel as the filtered pixel value; the filtering window The size of is an important parameter, which determines the effect of median filtering. A larger window can smooth the noise better, but it may also cause the loss of image details; a smaller window may not be enough to completely remove the noise, but it can better preserve the image details. The enhancement process uses a histogram equalization algorithm, and its probability density function is:
[0036] in, is the gray level, is the number of pixels of this gray level, is the total number of image pixels; Histogram equalization is a method to enhance the contrast of an image by redistributing the pixel values of the image to make the histogram distribution of the image more uniform. This method is particularly suitable for images where both the background and foreground are too bright or too dark, and it can enhance the contrast of the image by expanding the dynamic range of the image.
[0037] In one embodiment, specifically, in step three, the geometric shape of the lead frame is extracted using an edge detection algorithm to determine the edge by calculating the gradient amplitude and direction; specifically, edge detection is a technology used in image processing to identify areas with significant brightness changes in an image. When extracting the geometric shape of the lead frame, the edge detection algorithm determines the edge position by calculating the gradient amplitude and direction of each pixel in the image, wherein: Gradient magnitude: represents the intensity of the change in image brightness. It is usually calculated using the Sobel operator, Prewitt operator, or Canny edge detector, which estimate the gradient magnitude by considering the grayscale changes within the pixel and its neighborhood. Gradient direction: refers to the direction in which the image brightness changes fastest, that is, the direction of the edge. The calculation of the gradient direction is usually performed simultaneously with the gradient amplitude to provide useful information for subsequent edge connection or shape analysis; The output of the edge detection algorithm is a binary image where edge locations are marked as white (or highlighted) and non-edge locations are marked as black (or dimmed). This binary image can be used to further analyze the geometry of the lead frame. The size of the lead frame is extracted by using a minimum bounding rectangle algorithm to calculate the length and width of a rectangle to represent the size of the lead frame; specifically, the minimum bounding rectangle is the smallest rectangular frame that can completely contain the target object. First, the boundary points of the target object (i.e., the lead frame) are determined according to the edge detection result. Then, algorithms such as the rotating caliper method and the minimum area method are used to calculate the smallest rectangular frame that can contain these boundary points. Finally, the length and width of the rectangle are output as the size of the lead frame. The minimum bounding rectangle algorithm is widely used in image processing, especially in the fields of target detection, shape analysis, and image recognition. The surface texture features of the lead frame are extracted through a grayscale co-occurrence matrix. Specifically, the grayscale co-occurrence matrix is a statistical method for analyzing image texture. It extracts the texture features of the image by calculating the frequency of occurrence of grayscale combinations of adjacent pixel pairs at a certain distance and direction in the image. First, an appropriate distance and direction parameter is selected; then, each pixel point in the image is traversed to find adjacent pixel points according to the selected distance and direction; then, the number of occurrences of the grayscale combination of each pixel point and its adjacent pixel points is counted to form a grayscale co-occurrence matrix; finally, various texture features such as energy, contrast, correlation, entropy, etc. are extracted from the grayscale co-occurrence matrix; the grayscale co-occurrence matrix can capture the spatial relationship between pixels in the image, thereby effectively describing the texture features of the image. When extracting the surface texture features of the lead frame, the grayscale co-occurrence matrix is a very useful tool.
[0038] In one embodiment, specifically, in step 4, the machine learning model uses a decision tree algorithm, and its information gain formula is:
[0039] in, For the dataset The information entropy of Characterized by Features The value is A subset of data; specifically: First, we need to define the information entropy of a data set, which represents the measure of uncertainty or "purity" of samples in the data set. Datasets of categories , its information entropy Defined as:
[0040] in, It is the first The proportion of samples in the same class; Information gain is obtained by selecting a feature After splitting, the dataset The reduction in information entropy. Assuming the feature have Different values , using the feature For the dataset Dividing it, we get Subset ,in Representation characteristics The upper value is A sample subset, then, using the feature Information gain for splitting The calculation is as follows:
[0041] in, and Represents the data set and subset The number of samples in ; is the dataset before splitting The information entropy of represents the uncertainty of the original data set; is the weighted average information entropy of the split data set, indicating the weighted average information entropy of the split data set according to the characteristics. The average uncertainty of the partitioned data set; The difference between the two is the information gain, which indicates that The amount by which the uncertainty of the dataset is reduced after the split.
[0042] In one embodiment, specifically, in step five, the key dimensions include length, width and pin spacing; the length and width are measured by structured light measurement, and the three-dimensional coordinates of the surface points of the object are calculated according to the triangulation principle to obtain the length and width dimensions; specifically: The structured light measurement method projects a specific pattern of light (such as laser stripes or gratings) onto the surface of the object to be measured, and then uses a camera to capture the image formed by these lights on the surface of the object. According to the principle of triangulation (i.e., using the geometric relationship between the light source, the surface point of the object and the camera), the three-dimensional coordinates of the surface point of the object can be calculated. Through these three-dimensional coordinates, the length and width of the object can be further calculated. The structured light measurement method has the advantages of non-contact, high precision, and high speed, and is particularly suitable for the measurement of complex shapes and small sizes; The pin spacing is measured using an image sub-pixel algorithm; specifically: First, the image of the pin area is captured by a camera. Then, the image is preprocessed using image processing software, such as denoising and contrast enhancement. Next, a sub-pixel algorithm is used for edge detection to obtain the precise position of the pin edge. Finally, the pin spacing is calculated based on the edge position. Dimensional deviation refers to the difference between the measured value and the standard value, which is used to evaluate whether the product size meets the design requirements. The calculation formula is:
[0043] in, is the measured value, is the standard value.
[0044] In one embodiment, specifically, in step seven, the resistance of the lead frame By voltammetry:
[0045] in, is the voltage, is the electric current; The capacitance and inductance of the lead frame are measured using an LCR tester, and are calculated based on a test frequency and a measured value. Specifically, a desired test frequency is set on the LCR tester, and the selection of the test frequency depends on the characteristics of the component under test and the test requirements. The lead frame is connected to the test port of the LCR tester as the component under test. The LCR tester is started for testing, and the tester will automatically output a test signal and measure the response of the component under test. After the test is completed, the LCR tester will display the inductance, capacitance and resistance values of the component under test.
[0046] In one embodiment, specifically, in step eight, the comprehensive evaluation score Using the weighted summation method, the formula is:
[0047] in, , , , They are defect score, deviation score, material analysis score and electrical performance score. , , , are the weights of defect score, deviation score, material analysis score and electrical performance score, respectively, and ; Assume that we conduct quality inspection on a batch of lead frames and obtain the following scores: Defect Score : After testing the lead frame surface for defects such as scratches, dents, protrusions and oxidation, and comparing them with the preset standard values, the average defect score was 75 points (out of 100 points); Deviation score :The key dimensions are measured by optical measurement technology and compared with the preset standard values. The average dimensional deviation score is 80 points; Material Analysis Score : Quantitative analysis is performed using atomic absorption spectroscopy or inductively coupled plasma emission spectroscopy, with an average material analysis score of 90 points; Electrical performance score :Connect the test circuit to measure the resistance, capacitance and inductance parameters, and compare them with the preset standard values. The average electrical performance score is 85 points; At the same time, we set the weights for each score as follows: Defect score weight : 0.2 Weight of the bias score : 0.3 Weight of material analysis score : 0.3 Weight of electrical performance score : 0.2 The comprehensive evaluation score for:
[0048] According to the preset quality standards, the comprehensive evaluation score between 80-89 points is classified as a good grade, and the quality grade of this batch of lead frames is judged to be good.
[0049] The quality grades include four levels: excellent, good, qualified and unqualified. Among them, the excellent comprehensive evaluation score is greater than or equal to 90 points, the good comprehensive evaluation score is between 80-89 points, the qualified comprehensive evaluation score is between 60-79 points, and the unqualified comprehensive evaluation score is less than 60 points.
[0050] In combination with the above embodiments, the present application provides a specific implementation of a lead frame quality inspection method: Inspection preparation: Select a suitable industrial camera, set the resolution to 2048×1536 pixels, and the frame rate to 30 frames per second. Arrange 6 industrial cameras to capture images of the lead frame from the front, back, both sides, top, and bottom. Prepare structured light measurement equipment, atomic absorption spectrometer, LCR tester, and computers for data processing and model training.
[0051] Image acquisition: Place the lead frame to be inspected on a fixed workbench and ensure that it is within the camera's field of view. Start the industrial camera and simultaneously shoot the lead frame from various angles to obtain 6 sets of appearance images.
[0052] Image denoising and enhancement: For the acquired image, use the median filter algorithm to denoise. For example, select a filter window size of 3×3 pixels, sort the pixel values of each pixel and its neighborhood, and take the middle value as the pixel value after filtering. Use the histogram equalization algorithm to enhance the image. Calculate the number of pixels at each gray level, obtain the probability density function, and then redistribute the gray value to enhance the image contrast.
[0053] Feature extraction: Use the Canny edge detection algorithm to extract the geometric shape of the lead frame. Calculate the gradient amplitude and direction of the pixels in the image, set the high threshold to 100 and the low threshold to 50 to determine the edge. Use the minimum bounding rectangle algorithm to extract the size, and calculate the length and width of the rectangle to be 10mm and 5mm respectively. Extract surface texture features through the gray level co-occurrence matrix, and calculate the eigenvalues such as contrast and correlation.
[0054] Defect detection: Based on the extracted features, a decision tree algorithm is used to build a machine learning model. In the training data set, information gain is calculated based on the features, and the best split feature is selected. The features to be detected are input into the model to detect defects such as scratches and bumps on the surface. Compared with the preset standard value, if the scratch length exceeds 2mm, it is unqualified, and a defect score is given according to the defect situation, assuming it is 70 points (out of 100 points).
[0055] Dimension measurement: For length and width, the structured light measurement method is used. Stripe structured light is projected onto the lead frame. According to the principle of triangulation, the deformed light stripe image is obtained by the camera, and the length measurement value is calculated to be 9.8mm and the width measurement value is 4.9mm. The pin spacing is measured using the image sub-pixel algorithm, and the measurement value is 0.51mm. Compared with the preset standard value, the dimension deviation score is calculated, assuming it is 85 points.
[0056] Material analysis: Atomic absorption spectroscopy is used for material analysis. The sample is placed in a spectrometer to generate characteristic spectral lines. The element concentration is calculated based on the intensity of the spectral lines and compared with the standard value to obtain the material analysis score, which is assumed to be 90 points.
[0057] Electrical performance measurement: Connect the test circuit, measure the resistance by volt-ampere method, apply 5V voltage, measure the current as 1A, and calculate the resistance as 5Ω. Use LCR meter to measure capacitance and inductance at 1kHz frequency, and get capacitance value of 100pF and inductance value of 1μH respectively. Compare with the preset standard value to get the electrical performance score, assuming it is 95 points.
[0058] Comprehensive evaluation: Assume that the weights of defect score, deviation score, material analysis score, and electrical performance score are 0.3, 0.2, 0.3, and 0.2, respectively; Comprehensive evaluation score = 0.3×70+0.2×85+0.3×90+0.2×95=84 points.
[0059] According to the preset quality grade classification standard, 80-90 points are good, and the quality grade of the lead frame is judged to be good.
[0060] Figure 2 is a functional module diagram of a lead frame quality inspection device according to an embodiment of the present application. Figure 2As shown, a lead frame quality inspection device includes: an image acquisition module, an image processing module, a dimension measurement module, a material analysis module, an electrical performance test module and a data integration and evaluation module; The image acquisition module includes multiple industrial cameras for acquiring appearance images of the lead frame from different angles. The appearance images of the lead frame are photographed from multiple angles by multiple industrial cameras to ensure high-resolution image acquisition without blind spots. Multi-angle photography helps to discover hidden defects, such as scratches, cracks, stains, etc., and improves the accuracy and reliability of detection. The image processing module is used to perform denoising, enhancement, feature extraction and defect detection on the collected image to obtain a defect score. The denoising process is used to reduce the noise in the image and improve the image quality. The enhancement process is used to enhance the image contrast and sharpen the edges to make the defects more obvious. The feature extraction process is used to extract key features in the image, such as edges, shapes, textures, etc. Based on the extracted features, the defects in the image are identified and located, and the defect score is calculated. The dimension measurement module includes an optical measuring instrument and a related algorithm, which is used to measure the key dimensions of the lead frame and obtain the dimension deviation score, wherein the key dimensions include length, width, pin spacing, etc.; The material analysis module is used to analyze the material composition of the lead frame and obtain a material analysis score, wherein the material composition of the lead frame includes metal type, purity, impurity content, etc.; The electrical performance test module is used to measure the resistance, capacitance and inductance parameters of the lead frame to obtain an electrical performance score; The data integration and evaluation module is used to integrate defect scores, deviation scores, material analysis scores and electrical performance scores, and to perform comprehensive evaluation scores and divide quality grades according to quality standards, and finally output a quality evaluation report for the lead frame, including information such as comprehensive evaluation scores, quality grades, specific defects and suggestions.
[0061] For other details of the technical solutions implemented by each module in the device of the above embodiment, reference may be made to the description of a lead frame quality detection method in the above embodiment, which will not be repeated here.
[0062] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0063] like Figure 3 The present invention provides a schematic diagram of the structure of an electronic device according to an embodiment of the present invention, which is suitable for implementing the electronic device in the embodiment of the present invention. Figure 3The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0064] like Figure 3 As shown, the electronic device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device are also stored. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.
[0065] Typically, the following devices can be connected to the I / O interface: input devices such as sensors or visual information acquisition devices; output devices such as display screens; storage devices such as tapes, hard disks, etc.; and communication devices. The communication device allows the electronic device to communicate with other devices (such as edge computing devices) wirelessly or by wire to exchange data. Figure 3 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.
[0066] Suppose we have a specially designed electronic device for performing a lead frame quality inspection method. The processor of this electronic device can be a high-performance multi-core central processing unit (CPU), such as Intel Core i7 or AMD Ryzen 7 series. The memory can include high-speed DDR4 or DDR5 memory, and a large-capacity solid-state drive (SSD) to store the operating system, executable instructions, and data generated during the inspection process.
[0067] In actual operation, when receiving the lead frame to be inspected, the device first obtains image data through the image acquisition module. The processor executes executable instructions and calls the image processing algorithm to denoise and enhance the image. For example, the median filter algorithm is used to remove noise in the image, and the contrast of the image is enhanced by histogram equalization. Then, the edge detection algorithm is used to extract the geometric shape of the lead frame, the minimum enclosing rectangle algorithm is used to obtain the size information, and the gray level co-occurrence matrix is used to extract the surface texture features. Next, based on the extracted features, the decision tree algorithm is used to detect surface defects, and the defect score is obtained by comparing with the preset standard value. For size measurement, the processor controls the optical measuring instrument, obtains the key size through the structured light measurement method and the image sub-pixel algorithm, and calculates the size deviation score. In terms of material analysis, the processor controls the atomic absorption spectrometer or the inductively coupled plasma emission spectrometer to perform elemental analysis and obtain the material analysis score. For electrical performance testing, the processor controls the test circuit and the LCR tester to measure the resistance, capacitance and inductance parameters and obtain the electrical performance score. Finally, the processor integrates all scores according to the weighted summation method, performs a comprehensive evaluation according to the preset quality standards, divides the quality levels, and outputs the results for display or storage.
[0068] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.
[0069] According to the computer-readable storage medium of the embodiment of the present disclosure, non-transitory computer-readable instructions are stored thereon. When the non-transitory computer-readable instructions are executed by a processor, all or part of the steps of a lead frame quality detection method of each embodiment of the present disclosure are executed.
[0070] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).
[0071] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.
[0072] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.
[0073] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, devices, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.
[0074] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.
[0075] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.
[0076] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.
[0077] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
[0078] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. A lead frame quality inspection method, characterized in that: The following steps are involved: Step 1: Use an industrial camera to capture images of the lead frame from multiple angles to obtain appearance images of the lead frame at multiple angles; Step 2: performing denoising and enhancement processing on the appearance images of the lead frame at multiple angles obtained; Step 3: Using image processing algorithms to extract the geometric shape, size, and surface texture features of the lead frame; Step 4: Based on the extracted geometric shape, size, and surface texture features of the lead frame, use machine learning or deep learning models to detect scratches, dents, protrusions, and oxidation defects on the surface of the lead frame, and compare them with preset standard values to obtain defect scores; Step 5: Use optical measurement technology to measure the key dimensions of the lead frame and compare them with the preset standard values to obtain the dimension deviation score; Step 6: Use atomic absorption spectroscopy or inductively coupled plasma emission spectroscopy to perform quantitative analysis based on the relationship between the intensity of characteristic spectral lines and element concentrations, and compare with preset standard values to obtain material analysis scores; Step 7: Connect the test circuit, measure the resistance, capacitance, and inductance parameters of the lead frame, and compare them with the preset standard values to obtain the electrical performance score; Step 8. Integrate the defect score, deviation score, material analysis score and electrical performance score, conduct a comprehensive evaluation of the quality of the lead frame according to the preset quality standards, obtain a comprehensive evaluation score, and divide the quality grade according to the comprehensive evaluation score of the lead frame.
2. A lead frame quality inspection method according to claim 1, characterized in that: In step 2, the denoising process uses a median filter algorithm, and its formula is: in, is the pixel value after filtering, is the original pixel value, is the filter window; The enhancement process uses a histogram equalization algorithm, and its probability density function is: in, is the gray level, is the number of pixels of this gray level, is the total number of image pixels.
3. A lead frame quality inspection method according to claim 1, characterized in that: In step three, the geometric shape of the lead frame is extracted using an edge detection algorithm, and the edge is determined by calculating the gradient amplitude and direction; the size of the lead frame is extracted using a minimum circumscribed rectangle algorithm, and the length and width of the rectangle are calculated to represent the size of the lead frame; the surface texture features of the lead frame are extracted using a grayscale co-occurrence matrix.
4. A lead frame quality inspection method according to claim 1, characterized in that: In step 4, the machine learning model uses the decision tree algorithm, and its information gain formula is: in, For the dataset The information entropy of Characterized by Features The value is subset of data.
5. A lead frame quality inspection method according to claim 1, characterized in that: In step 5, the key dimensions include length, width and pin spacing; the length and width are measured by structured light measurement, and the three-dimensional coordinates of the surface points of the object are calculated according to the triangulation principle to obtain the length and width dimensions; the pin spacing is measured using an image sub-pixel algorithm; Dimensional deviation The calculation formula is: in, is the measured value, is the standard value.
6. A lead frame quality inspection method according to claim 1, characterized in that: In step seven, the lead frame resistor By voltammetry: in, is the voltage, is the electric current; The capacitance and inductance of the lead frame are measured using an LCR tester and calculated according to the test frequency and the measured value.
7. A lead frame quality inspection method according to claim 1, characterized in that: In step eight, the comprehensive evaluation score Using the weighted summation method, the formula is: in, , , , They are defect score, deviation score, material analysis score and electrical performance score. , , , are the weights of defect score, deviation score, material analysis score and electrical performance score, respectively, and ; The quality grades include four grades: excellent, good, qualified and unqualified.
8. A lead frame quality inspection device, characterized in that: include: Image acquisition module, image processing module, dimension measurement module, material analysis module, electrical performance test module and data integration and evaluation module; The image acquisition module includes a plurality of industrial cameras for acquiring appearance images of the lead frame from different angles; The image processing module is used to perform denoising, enhancement, feature extraction and defect detection on the collected images to obtain defect scores; The dimension measurement module includes an optical measuring instrument and a related algorithm for measuring the key dimensions of the lead frame and obtaining a dimension deviation score; The material analysis module is used to analyze the material composition of the lead frame and obtain a material analysis score; The electrical performance test module is used to measure the resistance, capacitance and inductance parameters of the lead frame to obtain an electrical performance score; The data integration and evaluation module is used to integrate the defect score, deviation score, material analysis score and electrical performance score, and to perform comprehensive evaluation scores and divide quality grades according to quality standards.
9. An electronic device, characterized in that: The electronic device comprises: A processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute a lead frame quality detection method according to any one of claims 1 to 7 by executing the executable instructions.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a lead frame quality detection method according to any one of claims 1 to 7 is implemented.
Citation Information
Cited By
Iron box packaging quality detection method and system
CN121095192A
