A method and system for identifying the filling state of a plastic encapsulation material of an integrated circuit lead frame

Abstract

The present application belongs to the technical field of image detection, and particularly relates to a plastic sealing material filling state identification method and system for an integrated circuit lead frame, which constructs a four-dimensional feature system combining macroscopic integrity and microscopic texture quality with whole area coverage index, pin gap coverage index, surface gray scale non-uniformity index and edge profile irregularity index as the core. Various defects such as large-area material shortage, pin gap material shortage, surface gray scale non-uniformity and edge profile irregularity can be identified synchronously, multi-dimensional accurate evaluation is realized, the surface gray scale non-uniformity index and the edge profile irregularity index are introduced, the surface gray scale non-uniformity problem caused by bubbles and foreign matters is quantified through the gray scale parameter, the edge profile distortion defect caused by overflow and punching is quantified through curvature fitting, the traditional single detection mode of only detecting the filling area is broken through, the dual detection of macroscopic filling quality and microscopic forming quality can be realized, and the defect detection type and detection precision can be significantly improved.

Description

Technical Field

[0001] This invention belongs to the field of image detection technology, specifically relating to a method and system for identifying the molding compound filling state of integrated circuit lead frames. Background Technology

[0002] In integrated circuit packaging, the quality of lead frame molding compound filling directly affects the reliability of the device. Common defects in the molding process include large-area material shortage, incomplete filling of lead gaps, abnormal texture on the surface of the molding compound caused by bubbles or foreign objects, and poor edge contours caused by overflow and wire breakage. These various defect types coexist on both macroscopic and microscopic scales, which places high demands on the comprehensiveness of the detection methods.

[0003] In the field of industrial vision inspection, existing solutions use image acquisition, preprocessing, threshold segmentation, and morphological processing to assess the quality of product surfaces. For the inspection of molding compound filling, common practices include judging whether there is missing material by area ratio or template comparison, or assessing surface non-uniformity by grayscale variance. However, large-area missing material and local micro-missing material at pin gaps are significantly different in spatial scale, and a single area index is difficult to take into account at the same time. Grayscale anomalies caused by surface bubbles and foreign objects and contour distortions caused by edge overflow and wire punching are different types of defects with different physical properties, requiring different features for characterization.

[0004] Some solutions attempt to introduce multiple features for comprehensive evaluation. However, if the independent detection results of each feature are simply superimposed without unifying and normalizing the indicators of different dimensions, the features with different value ranges are prone to weight imbalance during fusion, which can lead to the masking of certain defect types. In addition, the weights and judgment thresholds of each feature in existing solutions mostly rely on manual experience to set, making it difficult to adaptively adjust according to changes in different batches or working conditions.

[0005] To address the aforementioned issues, this application presents a method and system for identifying the molding compound filling state of integrated circuit lead frames. Summary of the Invention

[0006] To address the shortcomings of the prior art mentioned in the background section, this application proposes a method and system for identifying the molding compound filling state of integrated circuit leadframes. By extracting overall area coverage indicators, pin gap coverage indicators, surface grayscale non-uniformity indicators, and edge contour irregularity indicators, a four-dimensional feature system combining macroscopic coverage integrity and microscopic surface texture quality is constructed. This system simultaneously detects large-area material shortages, pin gap material shortages, grayscale non-uniformity caused by surface bubbles or foreign matter, and edge contour irregularity caused by overflow or wire punching. This enables a multi-dimensional and comprehensive evaluation of the molding compound filling state of leadframes, thereby solving the problems in the background section.

[0007] Firstly, to achieve the above objectives, this application provides a method for identifying the molding compound filling state of an integrated circuit lead frame, which includes the following specific steps: S1. Obtain surface inspection image data and structural parameter data of the lead frame after plastic sealing; S2. Perform image segmentation processing on the surface inspection image data to distinguish between the molding compound filling area and the exposed metal area of ​​the lead frame, and extract the contour data of the molding compound filling area. S3. Based on the contour data and structural parameter data, obtain the overall area coverage index and pin gap coverage index of the lead frame, and obtain the filling coverage integrity analysis results. S4. Based on the surface detection image data, calculate the surface gray level non-uniformity index and the edge contour irregularity index to obtain the surface image texture feature analysis results. S5. Based on the results of the filling coverage integrity analysis and the surface image texture feature analysis, after unifying the direction of each indicator, the indicators are fused according to the preset comprehensive scoring weights to calculate the comprehensive score of the molding compound filling status. S6. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

[0008] In a preferred embodiment based on the above scheme, step S2 includes: S21. Extract surface inspection image data of the lead frame after plastic sealing; S22. Based on the surface detection image data, a grayscale threshold segmentation algorithm is used to perform image segmentation processing on the molding compound filling area and the exposed metal area of ​​the lead frame to obtain the contour data of the molding compound filling area.

[0009] In a preferred embodiment based on the above scheme, step S22 includes: S221. Statistically analyze the gray value distribution of each pixel in the surface detection image and construct a gray level histogram; S222. Based on the grayscale histogram, identify the bimodal distribution characteristics of the molding compound filling area and the exposed metal area of ​​the lead frame, and determine the segmentation threshold. S223. Binarize the surface detection image according to the segmentation threshold to divide the plastic sealant filling area and the lead frame metal exposed area; S224. Perform morphological closing operation on the binarized molding compound filling area to fill the internal micro-holes and obtain the contour data of the molding compound filling area.

[0010] In a preferred embodiment based on the above scheme, step S3 includes: S31. Based on the structural parameter data of the lead frame, obtain the total area At of the target filling region; S32. Calculate the actual coverage area Ac based on the contour data of the molding compound filling area; S33. Calculate the overall area coverage index Ra based on the actual coverage area and the total area of ​​the target filling area. The specific calculation formula is as follows: ; S34. Based on the structural parameter data of the lead frame, obtain the distribution data of the gap area between the pins. Determine whether each gap area is completely covered and filled according to the contour data of the molding compound filling area. Calculate the pin gap coverage index Rg by counting the number of gap areas that are not completely covered and filled (Nu) and the total number of gap areas (Ng). The specific calculation formula is as follows: ; The results of the fill coverage integrity analysis include the overall area coverage index Ra and the pin gap coverage index Rg.

[0011] In a preferred embodiment based on the above scheme, step S4 includes: S41. Extract the grayscale distribution data within the molding compound filling area; S42. Based on grayscale distribution data, calculate the grayscale mean μ and grayscale standard deviation σ. The specific calculation formulas for the grayscale mean μ and grayscale standard deviation σ are as follows: ; ; Where Np is the total number of pixels. Let be the grayscale value of the i-th pixel; S43. Based on the grayscale mean and grayscale standard deviation, calculate the surface grayscale non-uniformity index U. The specific calculation formula is as follows: ; Where z is a preset regularization constant; S44. Extract the edge contour line of the molding compound filling area, perform curve fitting, and calculate the curvature value at each sampling point along the edge contour line to obtain the curvature value sequence. S45. Based on the curvature value sequence and the design reference curvature in the lead frame structural parameter data, calculate the edge profile irregularity index J. The calculation formula is as follows: ; Where n is the total number of sampling points on the edge contour line. Let be the curvature value at the i-th sampling point. The design reference curvature for all sampling points; The surface image texture feature analysis results include the surface gray-level non-uniformity index U and the edge contour irregularity index J.

[0012] In a preferred embodiment based on the above scheme, step S5 includes: S51. Normalize the surface gray-level non-uniformity index U and the edge contour irregularity index J, mapping them to the [0, 1] interval to obtain the normalized surface gray-level non-uniformity index. and the normalized edge contour irregularity index ; S52, based on the overall area coverage index Ra, pin gap coverage index Rg, and normalized surface grayscale non-uniformity index. and the normalized edge contour irregularity index Overall score of molding compound filling condition The calculation is performed using the following formula: ; Where α, β, γ, and δ are the weighting coefficients for overall area coverage index, pin gap coverage index, surface gray level non-uniformity, and edge contour irregularity, respectively.

[0013] In a preferred embodiment based on the above scheme, step S6 includes: S61. Obtain the comprehensive score of the molding compound filling status and the preset sorting threshold; S62. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold. If the comprehensive score of the molding compound filling state is greater than or equal to the preset sorting threshold, output the first sorting control signal to indicate that it is a qualified product. If the overall score of the molding compound filling status is less than the preset sorting threshold, a second sorting control signal will be output to indicate that the product is unqualified.

[0014] Secondly, this application provides a molding compound filling state identification system for integrated circuit leadframes, including a data acquisition module for performing step S1 to acquire surface detection image data and structural parameter data of the leadframe after molding. The image segmentation processing module is used to perform step S2, perform image segmentation processing on the surface detection image data, distinguish between the molding compound filling area and the lead frame metal exposed area, and extract the contour data of the molding compound filling area. The fill coverage integrity analysis module is used to execute step S3, which, based on contour data and structural parameter data, obtains the overall area coverage index and pin gap coverage index of the lead frame, and obtains the fill coverage integrity analysis results. The surface image texture feature analysis module is used to execute step S4, which calculates the surface gray level non-uniformity index and the edge contour irregularity index based on the surface detection image data, and obtains the surface image texture feature analysis results. The filling status comprehensive scoring module is used to execute step S5. Based on the filling coverage integrity analysis results and the surface image texture feature analysis results, the direction of each indicator is unified and then fused according to the preset comprehensive scoring weight to calculate the comprehensive score of the molding compound filling status. The product sorting control signal output module is used to execute step S6, compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

[0015] Thirdly, this application provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the above-described method for identifying the molding compound filling state of an integrated circuit lead frame by calling a computer program stored in the memory.

[0016] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method for identifying the molding compound filling state of an integrated circuit lead frame.

[0017] Compared with the prior art, the beneficial effects of the present invention are: First, this invention utilizes the grayscale difference between the molding compound and the metal plating layer, combined with grayscale threshold segmentation, binarization, and morphological closing operations, to accurately distinguish between the molding compound-filled area and the exposed metal area of ​​the lead frame. This effectively removes image noise, repairs micro-holes, and accurately extracts the contour data of the molding compound-filled area, avoiding the problems of blurred boundaries and feature loss in traditional segmentation. It can provide a reliable image data foundation for the accurate calculation of subsequent core indicators.

[0018] Secondly, this invention breaks through the limitations of traditional single-dimensional detection technology by constructing a four-dimensional feature system that combines macroscopic integrity with microscopic texture quality, with overall area coverage index, pin gap coverage index, surface gray level unevenness index, and edge contour irregularity index as the core. It can simultaneously identify various defects such as large-area material shortage, pin gap material shortage, surface gray level unevenness, and edge contour irregularity, and realize multi-dimensional accurate evaluation of the molding compound filling status. It can effectively solve the problems of missed detection and false detection in traditional detection, and can significantly improve the coverage and accuracy of defect detection.

[0019] Third, this invention simultaneously calculates the overall area coverage index and the pin gap coverage index, which can detect large-area material shortage defects in the molding compound, and accurately identify hidden defects such as incomplete pin gap filling and local micro-material shortages. This makes up for the shortcomings of traditional testing, which only checks the overall area and ignores the quality of local gaps. It can effectively ensure the integrity of the lead frame molding structure and avoid quality risks such as leakage and cracking in the package.

[0020] Fourth, this invention introduces surface grayscale non-uniformity index and edge contour irregularity index. By using grayscale parameters, it quantifies the surface grayscale non-uniformity caused by bubbles and foreign objects. By using curvature fitting, it quantifies edge contour distortion defects caused by overflow and wire punching. This breaks through the traditional single detection mode that only detects the filling area. It can realize dual detection of macroscopic filling quality and microscopic forming quality, which can significantly improve the types and accuracy of defect detection.

[0021] Fifth, this invention normalizes the surface grayscale non-uniformity index and the edge contour irregularity index, combines the optimal weight coefficient obtained by regression fitting of historical data, and integrates four core indicators to construct a comprehensive scoring model for the molding compound filling status. This model can simultaneously quantify filling integrity defects and surface texture quality defects, replacing the traditional single threshold judgment method. The evaluation standard is objective and unified, which can significantly improve the accuracy of identifying the molding compound filling status. Attached Figure Description

[0022] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the overall process of the method for identifying the molding compound filling state of an integrated circuit lead frame according to the present invention. Figure 2 This is a flowchart illustrating step S3 in the method for identifying the molding compound filling state of an integrated circuit lead frame according to the present invention. Figure 3 This is a schematic diagram of the framework of a molding compound filling state identification system for an integrated circuit lead frame according to the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1: To address the technical problems raised in the background art, this application provides a preferred embodiment: such as Figure 1 - Figure 3 As shown, a method for identifying the molding compound filling state of an integrated circuit lead frame includes the following specific steps: S1. Obtain surface inspection image data and structural parameter data of the lead frame after plastic sealing; It should be noted that the surface detection image data in step S1 is acquired by the image acquisition device, and the structural parameter data comes from the lead frame design file. Both must be unified to the same coordinate system to ensure the accuracy of the subsequent calculation of the overall area coverage index and the pin gap coverage index.

[0025] The registration process is as follows: Before detection, the conversion relationship between image pixels and actual physical size is established by using the positioning features of the known physical size on the standard calibration plate or lead frame. During detection, the preset positioning reference features on the lead frame are identified in the surface detection image. Based on the correspondence between the reference features in the image coordinate system and the structural parameter coordinate system, a transformation matrix between the two coordinate systems is established. Through the transformation matrix, the boundary of the target filling area and the gap area of ​​each pin in the structural parameter data are mapped to the image coordinate system, so that the structural parameter data and the surface detection image data are aligned in the same coordinate system. After registration, the actual coverage area calculated from the contour data has a spatial correspondence with the total area of ​​the target filling area obtained from the structural parameter data, and the coverage determination of each pin gap area also has an accurate coordinate basis. S2. Perform image segmentation processing on the surface inspection image data to distinguish between the molding compound filling area and the exposed metal area of ​​the lead frame, and extract the contour data of the molding compound filling area. S3. Based on the contour data and structural parameter data, obtain the overall area coverage index and pin gap coverage index of the lead frame, and obtain the filling coverage integrity analysis results. S4. Based on the surface detection image data, calculate the surface gray level non-uniformity index and the edge contour irregularity index to obtain the surface image texture feature analysis results. S5. Based on the results of the filling coverage integrity analysis and the surface image texture feature analysis, after unifying the direction of each indicator, the indicators are fused according to the preset comprehensive scoring weights to calculate the comprehensive score of the molding compound filling status. S6. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

[0026] Furthermore: In an optional embodiment, step S2 includes: S21. Extract surface inspection image data of the lead frame after plastic sealing; S22. Based on the surface detection image data, a grayscale threshold segmentation algorithm is used to perform image segmentation processing on the molding compound filling area and the exposed metal area of ​​the lead frame to obtain the contour data of the molding compound filling area.

[0027] In an optional embodiment, step S22 includes: S221. Statistically analyze the gray value distribution of each pixel in the surface detection image and construct a gray level histogram; S222. Based on the grayscale histogram, identify the bimodal distribution characteristics of the molding compound filling area and the exposed metal area of ​​the lead frame, and determine the segmentation threshold. It should be noted that: the molding compound filling area is an epoxy resin-based black molding compound area, and the exposed metal area of ​​the lead frame is an area with a bright metal coating on the surface. The bimodal distribution characteristic of the gray-level histogram is formed based on the gray-level difference between the black molding compound and the bright metal. The effectiveness of the bimodal threshold segmentation of gray-level in this scheme is based on the following imaging premise: the molding compound of the lead frame to be detected is an epoxy resin-based black molding compound, and the exposed metal area of ​​the lead frame has a bright metal coating on the surface. The two have a significant inherent gray-level difference under visible light or specific light source illumination. Under this premise, in the surface detection image obtained, the black molding compound area presents a low gray-level value, and the bright metal area presents a high gray-level value. Two distinguishable peaks are formed on the gray-level histogram, corresponding to the molding compound pixel group and the metal pixel group, respectively. Under the above imaging conditions, the segmentation threshold can be determined by the following rule: find the gray value corresponding to the valley position between the two peaks of the gray-level histogram as the segmentation threshold. In specific implementation, the gray level corresponding to the two peaks can be determined on the smoothed gray-level histogram, and the gray level with the smallest occurrence frequency in this interval can be searched as the segmentation threshold. This rule is simple and reliable and is suitable for standard imaging scenarios with obvious double-peak separation. Finally, it should be noted that the segmentation method based on the bimodal feature of the grayscale histogram in this solution is suitable for situations where there is a significant difference in grayscale between the molding compound and the metal surface of the lead frame. If the molding compound used is light in color or the metal coating surface has been specially treated to make the grayscale of the two similar, the bimodal feature may not be significant or may even disappear. In this case, a segmentation method based on texture, color space or other features should be used instead, or active light source illumination should be introduced to enhance the contrast. The advantages of the above embodiments are: the grayscale bimodal threshold segmentation is deeply adapted to the specific imaging scene of epoxy resin black molding compound and lead frame with bright metal coating on the surface, and the morphological closing operation is combined to fill the micro holes, which ensures the stability and accuracy of image segmentation under specific material combinations, and can lay a reliable foundation for subsequent feature extraction. S223. Binarize the surface detection image according to the segmentation threshold to divide the plastic sealant filling area and the lead frame metal exposed area; S224. Perform morphological closing operation on the binarized molding compound filling area to fill the internal micro-holes and obtain the contour data of the molding compound filling area.

[0028] It should be noted that the purpose of the morphological closing operation in this scheme is only to eliminate isolated micro-holes inside the molding compound filling area after binarization caused by image noise or micro-texture of the molding compound surface, and to obtain complete and connected molding compound filling area contour data, so as to provide accurate area boundaries for subsequent calculation of overall area coverage index and pin gap coverage index.

[0029] In practice, based on the quality requirements of the molding process and the image resolution, a threshold for retaining the hole area can be preset. Before the closing operation, connected component analysis is performed on all holes in the binarized image. Only holes with an area smaller than the retention threshold are filled, while holes with an area greater than or equal to the retention threshold are retained, so that they are correctly reflected as missing fills in the subsequent calculation of area fill rate and gap fill rate.

[0030] In an optional embodiment, step S3 includes: S31. Based on the structural parameter data of the lead frame, obtain the total area At of the target filling region; S32. Calculate the actual coverage area Ac based on the contour data of the molding compound filling area; S33. Calculate the overall area coverage index Ra based on the actual coverage area and the total area of ​​the target filling area. The specific calculation formula is as follows: ; S34. Based on the structural parameter data of the lead frame, obtain the distribution data of the gap area between the pins. Determine whether each gap area is completely covered and filled according to the contour data of the molding compound filling area. Calculate the pin gap coverage index Rg by counting the number of gap areas that are not completely covered and filled (Nu) and the total number of gap areas (Ng). The specific calculation formula is as follows: ; The results of the fill coverage integrity analysis include the overall area coverage index Ra and the pin gap coverage index Rg.

[0031] In an optional embodiment, step S4 includes: S41. Extract the grayscale distribution data within the molding compound filling area; S42. Based on grayscale distribution data, calculate the grayscale mean μ and grayscale standard deviation σ. The specific calculation formulas for the grayscale mean μ and grayscale standard deviation σ are as follows: ; ; Where Np is the total number of pixels. Let be the grayscale value of the i-th pixel; S43. Based on the grayscale mean and grayscale standard deviation, calculate the surface grayscale non-uniformity index U. The specific calculation formula is as follows: ; Where z is a preset regularization constant; S44. Extract the edge contour line of the molding compound filling area, perform curve fitting, and calculate the curvature value at each sampling point along the edge contour line to obtain the curvature value sequence. S45. Based on the curvature value sequence and the design reference curvature in the lead frame structural parameter data, calculate the edge profile irregularity index J. The calculation formula is as follows: ; Where n is the total number of sampling points on the edge contour line. Let be the curvature value at the i-th sampling point. The design reference curvature for all sampling points; The surface image texture feature analysis results include the surface gray-level non-uniformity index U and the edge contour irregularity index J.

[0032] In an optional embodiment, step S5 includes: S51. Normalize the surface gray-level non-uniformity index U and the edge contour irregularity index J, mapping them to the [0, 1] interval to obtain the normalized surface gray-level non-uniformity index. and the normalized edge contour irregularity index ; It should be noted that this scheme uses a linear normalization method based on the statistical extreme values ​​of the training samples, and the specific formula is as follows: ; ; in, and These represent the maximum and minimum values ​​of the surface gray-level non-uniformity index in the historical training samples, respectively. and These represent the maximum and minimum values ​​of the edge contour irregularity index in the historical training samples, respectively. During online testing, if the surface gray-scale non-uniformity index U or the edge contour irregularity index J of a certain lead frame sample exceeds the statistical range of the training samples, i.e., the surface gray-scale non-uniformity index U is less than... or greater than Or the edge contour irregularity index J value is less than or greater than If the normalization result is truncated to the boundary of the [0, 1] interval, that is, 0 is taken when it is less than 0 and 1 is taken when it is greater than 1. This truncation strategy avoids the normalization value from deviating significantly from the normal range due to individual abnormal samples, and ensures the stability and comparability of the comprehensive score. S52, based on the overall area coverage index Ra, pin gap coverage index Rg, and normalized surface grayscale non-uniformity index. and the normalized edge contour irregularity index Overall score of molding compound filling condition The calculation is performed using the following formula: ; Where α, β, γ, and δ are the weighting coefficients for overall area coverage index, pin gap coverage index, surface gray level non-uniformity, and edge contour irregularity, respectively.

[0033] It should be noted that the overall score for the molding compound filling condition is as follows: This comprehensively reflects both fill coverage integrity defects and surface texture quality defects. The overall area coverage index Ra and pin gap coverage index Rg characterize fill coverage integrity, while the normalized surface grayscale non-uniformity index... and the normalized edge contour irregularity index Characterizes surface texture quality defects, including uneven grayscale caused by bubbles or foreign matter, and irregular edge contours caused by overflow or wire breakage. The relative magnitudes of α and β reflect the weighting of overall area coverage and pin gap coverage in the fill integrity evaluation. They complement each other and together constitute the evaluation of the macro coverage dimension. The relative magnitudes of γ and δ reflect the weight distribution of surface gray-level non-uniformity and edge irregularity in the evaluation of surface texture quality. The two complement each other and together constitute the evaluation of the micro-texture dimension. The relative magnitudes of the two sets of weights reflect the degree of emphasis placed on macroscopic coverage integrity and microscopic surface texture quality in the overall score.

[0034] The four categories of indicators are organically integrated into a single comprehensive score through the aforementioned weighting system. The indicators are not simply superimposed after independent judgment, but rather work together under a unified scoring framework. When an indicator is abnormal, the comprehensive score can sensitively reflect the abnormality. When multiple indicators are abnormal at the same time, the deductions of each indicator are accumulated, causing the comprehensive score to drop significantly. When all indicators are in a good or excellent state, the comprehensive score reaches a high level.

[0035] In an optional embodiment, step S6 includes: S61. Obtain the comprehensive score of the molding compound filling status and the preset sorting threshold; S62. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold. If the comprehensive score of the molding compound filling state is greater than or equal to the preset sorting threshold, output the first sorting control signal to indicate that it is a qualified product. If the overall score of the molding compound filling status is less than the preset sorting threshold, a second sorting control signal will be output to indicate that the product is unqualified.

[0036] It should be noted that the preset comprehensive scoring weights and preset sorting thresholds are obtained as follows: Surface inspection image data and structural parameter data of multiple historical lead frames are acquired. Steps S2 to S5 are used to calculate the overall area coverage index Ra, pin gap coverage index Rg, surface grayscale non-uniformity index U, and edge contour irregularity index J for each historical lead frame. Then, the surface grayscale non-uniformity index U and the edge contour irregularity index J are normalized to obtain the normalized surface grayscale non-uniformity index. and the normalized edge contour irregularity index The actual molding compound filling status test results of each historical leadframe were obtained. Then, the overall area coverage index Ra, pin gap coverage index Rg, and normalized surface grayscale non-uniformity index were calculated. and the normalized edge contour irregularity index The corresponding actual detection results are imported into the fitting software for multivariate regression fitting to obtain α, β, γ, δ and preset sorting thresholds that maximize the accuracy of filling state identification.

[0037] The specific method for performing multiple regression fitting is as follows: Before multiple regression fitting, collect multiple lead frame samples from the historical production process. For each sample, perform steps S2 to S5 to extract the overall area coverage index, pin gap coverage index, surface gray level unevenness index, and edge contour irregularity index. Normalize the latter two. At the same time, obtain the actual plastic sealant filling status label for each sample, which is confirmed by manual review or authoritative testing equipment, i.e., qualified or unqualified. Then, use the above four index values ​​as independent variables and the actual label as dependent variable to construct a training dataset for regression fitting.

[0038] Next, the constructed training dataset is imported into the fitting software. The fitting software takes four index values ​​as input and uses the comprehensive scoring formula as the model framework. It automatically adjusts the weight coefficients of the overall area coverage index, pin gap coverage index, surface gray level unevenness, and edge contour irregularity through a multiple regression algorithm. The objective function of the fitting process is to maximize the accuracy of filling state identification, that is, to maximize the proportion of the comprehensive scoring judgment result that matches the actual label, with the preset sorting threshold as the boundary. The fitting software can traverse various weight combinations and threshold values, and find the global optimal solution through iterative optimization. Finally, it outputs a set of optimal weight coefficients and optimal preset sorting thresholds.

[0039] After fitting, the optimal weight coefficients are solidified into the calculation formula of the filling state comprehensive scoring module, and the optimal preset sorting threshold is configured into the comparison logic of the product sorting control signal output module. Subsequently, during online detection, the system scores and sorts based on this set of parameters, which can adaptively match the actual quality distribution characteristics of the current production line. If the production line process changes in the future, new historical data can be collected and refitted to quickly update the parameters without manual readjustment, thus achieving continuous optimization.

[0040] The advantages of the above embodiments are as follows: by normalizing the surface gray-scale non-uniformity index and the edge contour irregularity index before weighted comprehensive scoring, each index is mapped to a unified [0,1] interval, eliminating the difference in dimensions, so that the weight coefficients of the overall area coverage index, the pin gap coverage index, the surface gray-scale non-uniformity index, and the edge contour irregularity index can accurately reflect the true importance of each index, thus ensuring the scientificity and effectiveness of the comprehensive scoring. By employing a multivariate regression fitting method based on historical data, the system automatically learns the optimal weight coefficients and sorting thresholds, enabling it to achieve the highest accuracy in identifying the filling state through a data-driven approach, thus overcoming the limitations of traditional fixed thresholds and manual weighting.

[0041] Example 2: Based on the same inventive concept as Embodiment 1, such as Figure 3 As shown, this embodiment provides a molding compound filling state identification system for integrated circuit lead frames, including a data acquisition module for performing step S1 to acquire surface detection image data and structural parameter data of the lead frame after molding. The image segmentation processing module is used to perform step S2, perform image segmentation processing on the surface detection image data, distinguish between the molding compound filling area and the lead frame metal exposed area, and extract the contour data of the molding compound filling area. The fill coverage integrity analysis module is used to execute step S3, which, based on contour data and structural parameter data, obtains the overall area coverage index and pin gap coverage index of the lead frame, and obtains the fill coverage integrity analysis results. The surface image texture feature analysis module is used to execute step S4, which calculates the surface gray level non-uniformity index and the edge contour irregularity index based on the surface detection image data, and obtains the surface image texture feature analysis results. The filling status comprehensive scoring module is used to execute step S5. Based on the filling coverage integrity analysis results and the surface image texture feature analysis results, the direction of each indicator is unified and then fused according to the preset comprehensive scoring weight to calculate the comprehensive score of the molding compound filling status. The product sorting control signal output module is used to execute step S6, compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

[0042] The parameters and steps of each unit module in the above-described system for identifying the molding compound filling state of an integrated circuit lead frame according to the present invention can be referred to the parameters and steps in the embodiments of the method for identifying the molding compound filling state of an integrated circuit lead frame described above, and will not be repeated here.

[0043] The specific implementation process of the molding compound filling state identification method for integrated circuit leadframes provided in this solution is as follows: First, the data acquisition module acquires surface inspection image data and structural parameter data of the lead frame after plastic sealing.

[0044] Then, through the image segmentation processing module, based on the grayscale difference between epoxy resin black molding compound and bright metal coating, grayscale threshold segmentation and binarization processing are performed on the surface detection image data to distinguish the molding compound filling area and the lead frame metal exposed area. Then, morphological closing operation is used to fill the micro holes and extract the contour data of the molding compound filling area.

[0045] Next, the filling coverage integrity analysis module calculates the overall area coverage index by comparing the actual coverage area with the total area of ​​the target filling area based on the contour data. At the same time, it determines whether each pin gap area is completely covered and filled, counts the number of gap areas that are not completely covered and filled, and calculates the pin gap coverage index.

[0046] Meanwhile, through the surface image texture feature analysis module, the mean gray value and standard deviation of gray value are calculated based on the gray value distribution data in the filled area, and then the surface gray value non-uniformity index is calculated. The edge contour line of the filled area is extracted, the curvature value sequence is calculated through curve fitting, and the edge contour irregularity index is calculated based on the average value of the curvature deviation.

[0047] Next, the surface grayscale non-uniformity index and edge contour irregularity index are normalized by the filling state comprehensive scoring module. Combined with the optimal weight coefficient obtained by fitting historical data, the overall area coverage index, pin gap coverage index and normalized index are weighted and calculated to obtain the comprehensive score of the molding compound filling state.

[0048] Finally, the product sorting control signal output module compares the comprehensive score of the molding compound filling status with the preset sorting threshold and outputs a sorting control signal for qualified or unqualified products to achieve automated sorting.

[0049] Example 3: Based on the same inventive concept as Embodiment 1, this embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the above-described method for identifying the molding compound filling state of an integrated circuit lead frame by calling a computer program stored in the memory.

[0050] It should be noted that all computer programs for the method of identifying the molding compound filling state of an integrated circuit lead frame are implemented using the C language.

[0051] Example 4: Based on the same inventive concept as in Embodiment 1, this embodiment proposes a computer-readable storage medium having an erasable and rewritable computer program stored thereon. When a computer program runs on a computer device, it causes the computer device to execute the aforementioned method for identifying the molding compound filling state of an integrated circuit lead frame.

[0052] For example, computer-readable storage media can be read-only memory, random access memory, read-only optical disc, magnetic tape, floppy disk, and optical data storage devices.

[0053] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are relatively simple in description because they are fundamentally similar to the method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0054] The systems, media, and methods provided in the embodiments of the present invention are in one-to-one correspondence. Therefore, the systems and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be repeated here.

[0055] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0057] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0060] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0061] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0062] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. The above are merely embodiments of the present invention and are 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 principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for identifying the molding compound filling state of an integrated circuit lead frame, characterized in that, Includes the following steps: S1. Obtain surface inspection image data and structural parameter data of the lead frame after plastic sealing; S2. Perform image segmentation processing on the surface inspection image data to distinguish between the molding compound filling area and the exposed metal area of ​​the lead frame, and extract the contour data of the molding compound filling area. S3. Based on the contour data and structural parameter data, obtain the overall area coverage index and pin gap coverage index of the lead frame, and obtain the filling coverage integrity analysis results. S4. Based on the surface detection image data, calculate the surface gray level non-uniformity index and the edge contour irregularity index to obtain the surface image texture feature analysis results. S5. Based on the results of the filling coverage integrity analysis and the surface image texture feature analysis, after unifying the direction of each indicator, the indicators are fused according to the preset comprehensive scoring weights to calculate the comprehensive score of the molding compound filling status. S6. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

2. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 1, characterized in that: Step S2 includes: S21. Extract surface inspection image data of the lead frame after plastic sealing; S22. Based on the surface detection image data, a grayscale threshold segmentation algorithm is used to perform image segmentation processing on the molding compound filling area and the exposed metal area of ​​the lead frame to obtain the contour data of the molding compound filling area.

3. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 2, characterized in that: Step S22 includes: S221. Statistically analyze the gray value distribution of each pixel in the surface detection image and construct a gray level histogram; S222. Based on the grayscale histogram, identify the bimodal distribution characteristics of the molding compound filling area and the exposed metal area of ​​the lead frame, and determine the segmentation threshold. S223. Binarize the surface detection image according to the segmentation threshold to divide the plastic sealant filling area and the lead frame metal exposed area; S224. Perform morphological closing operation on the binarized molding compound filling area to fill the internal micro-holes and obtain the contour data of the molding compound filling area.

4. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 3, characterized in that: Step S3 includes: S31. Based on the structural parameter data of the lead frame, obtain the total area At of the target filling region; S32. Calculate the actual coverage area Ac based on the contour data of the molding compound filling area; S33. Calculate the overall area coverage index Ra based on the actual coverage area and the total area of ​​the target filling area. The specific calculation formula is as follows: ; S34. Based on the structural parameter data of the lead frame, obtain the distribution data of the gap area between the pins. Determine whether each gap area is completely covered and filled according to the contour data of the molding compound filling area. Calculate the pin gap coverage index Rg by counting the number of gap areas that are not completely covered and filled (Nu) and the total number of gap areas (Ng). The specific calculation formula is as follows: ; The results of the fill coverage integrity analysis include the overall area coverage index Ra and the pin gap coverage index Rg.

5. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 4, characterized in that: Step S4 includes: S41. Extract the grayscale distribution data within the molding compound filling area; S42. Based on the gray-level distribution data, calculate the gray-level mean μ and the gray-level standard deviation σ; S43. Based on the mean gray level μ and the standard deviation of gray level σ, calculate the surface gray level non-uniformity index U. The specific calculation formula is as follows: ; Where z is a preset regularization constant; S44. Extract the edge contour line of the molding compound filling area, perform curve fitting, and calculate the curvature value at each sampling point along the edge contour line to obtain the curvature value sequence. S45. Based on the curvature value sequence and the design reference curvature in the lead frame structural parameter data, calculate the edge profile irregularity index J. The calculation formula is as follows: ; Where n is the total number of sampling points on the edge contour line. Let be the curvature value at the i-th sampling point. The design reference curvature for all sampling points; The surface image texture feature analysis results include the surface gray-level non-uniformity index U and the edge contour irregularity index J.

6. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 5, characterized in that: Step S5 includes: S51. Normalize the surface gray-level non-uniformity index U and the edge contour irregularity index J, mapping them to the [0, 1] interval to obtain the normalized surface gray-level non-uniformity index. and the normalized edge contour irregularity index ; S52, based on the overall area coverage index Ra, pin gap coverage index Rg, and normalized surface grayscale non-uniformity index. and the normalized edge contour irregularity index Overall score of molding compound filling condition The calculation is performed using the following formula: ; Where α, β, γ, and δ are the weighting coefficients for overall area coverage index, pin gap coverage index, surface gray level non-uniformity, and edge contour irregularity, respectively.

7. The method for identifying the molding compound filling state of an integrated circuit lead frame according to claim 6, characterized in that: Step S6 includes: S61. Obtain the comprehensive score of the molding compound filling status and the preset sorting threshold; S62. Compare the comprehensive score of the molding compound filling state with the preset sorting threshold. If the comprehensive score of the molding compound filling state is greater than or equal to the preset sorting threshold, output the first sorting control signal to indicate that it is a qualified product. If the overall score of the molding compound filling status is less than the preset sorting threshold, a second sorting control signal will be output to indicate that the product is unqualified.

8. A molding compound filling state identification system for an integrated circuit lead frame, implemented based on the molding compound filling state identification method for an integrated circuit lead frame as described in any one of claims 1-7, characterized in that, It includes a data acquisition module, used to perform step S1, to acquire surface inspection image data of the lead frame after molding and structural parameter data of the lead frame; The image segmentation processing module is used to perform step S2, perform image segmentation processing on the surface detection image data, distinguish between the molding compound filling area and the lead frame metal exposed area, and extract the contour data of the molding compound filling area. The fill coverage integrity analysis module is used to execute step S3, which, based on contour data and structural parameter data, obtains the overall area coverage index and pin gap coverage index of the lead frame, and obtains the fill coverage integrity analysis results. The surface image texture feature analysis module is used to execute step S4, which calculates the surface gray level non-uniformity index and the edge contour irregularity index based on the surface detection image data, and obtains the surface image texture feature analysis results. The filling status comprehensive scoring module is used to execute step S5. Based on the filling coverage integrity analysis results and the surface image texture feature analysis results, the direction of each indicator is unified and then fused according to the preset comprehensive scoring weight to calculate the comprehensive score of the molding compound filling status. The product sorting control signal output module is used to execute step S6, compare the comprehensive score of the molding compound filling state with the preset sorting threshold, and output the product sorting control signal to the packaging production line control device according to the comparison result.

9. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor, characterized in that: the processor executes a method for identifying the molding compound filling state of an integrated circuit lead frame as described in any one of claims 1-7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: The device stores instructions that, when executed on a computer, cause the computer to perform a method for identifying the molding compound filling state of an integrated circuit lead frame as described in any one of claims 1-7.

Images