An image recognition-based automatic detection method for virtual soldering of GPU solder joints

Abstract

The present application relates to the technical field of material detection, and discloses a graphics card solder joint false welding automatic detection method based on image recognition, which comprises the following steps: first, comparing a visible light image of a graphics card after gray scale normalization with an infrared thermal distribution map after noise suppression to obtain surface texture and thermodynamic characteristics of the solder joint; second, splicing the two to obtain an associated feature weight matrix, extracting texture and thermodynamic characteristic vectors of the solder joint area from the matrix, and splicing the vectors into a fusion feature matrix according to the channel dimension; finally, combining a standard solder joint feature library to locate the coordinates of the false welding solder joint, and generating a graphics card solder joint false welding defect report according to the coordinates and the texture fracture index. This process realizes accurate false welding detection and report generation through multi-feature fusion and comparison, and can improve the accuracy and efficiency of the graphics card solder joint false welding detection based on image recognition, thereby providing a high-precision automatic solution for graphics card solder joint quality detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of material detection, and particularly relates to a graphics card solder joint false welding automatic detection method based on image recognition. BACKGROUND

[0002] The prior art relies on a single type of image information for analysis, and only visible light images are used to extract surface texture features, or only infrared thermal images are used to obtain temperature distribution features. This single feature analysis method cannot fully reflect the complex state of false welding joints. False welding may cause surface texture abnormalities such as cracking and deformation, and may also cause local temperature abnormalities due to poor contact. Relying on only one of these features may result in missed detection or misjudgment, and the subtle differences between normal and false welding joints cannot be accurately distinguished, which limits the detection accuracy.

[0003] The prior art has deficiencies in feature fusion and processing. Although some methods attempt to combine multiple features, they lack an effective dynamic weight distribution mechanism, making it difficult to adjust the weight according to the actual effectiveness of the features, resulting in a large amount of noise interference in the fused feature matrix and weakening of key features. In the feature extraction process, the positioning of the welding point area is not accurate enough, and the deformation state of the welding point cannot be adaptively matched, resulting in a large deviation between the extracted feature vector and the actual welding point state, which ultimately affects the accuracy of false welding coordinate positioning and the reliability of defect reports, making it difficult to meet the needs of high-precision graphics card production detection. SUMMARY

[0004] The present application provides a graphics card solder joint false welding automatic detection method based on image recognition to solve the problems raised in the background.

[0005] To achieve the above-mentioned purpose, the present application provides a graphics card solder joint false welding automatic detection method based on image recognition, comprising:

[0006] S1, comparing the visible light image of the graphics card after gray scale normalization and the infrared thermal distribution image of the graphics card after noise suppression position by position to obtain the surface texture features and thermodynamic features of the welding points in the graphics card;

[0007] S2, splicing the surface texture features and the thermodynamic features to obtain the associated feature weight matrix of the graphics card welding points;

[0008] S3, extracting the texture feature vector and the thermodynamic feature vector of the welding point area in the associated feature weight matrix according to the welding point area of the graphics card;

[0009] S4, splicing the texture feature vector and the thermodynamic feature vector in the channel dimension to obtain the fusion feature matrix of the graphics card welding points;

[0010] S5, positioning coordinates of the false solder joint in the solder joint region based on the fusion feature matrix and a standard solder joint feature library;

[0011] S6, generating a false solder joint defect report of the graphics card according to the coordinates and a texture fracture index of the false solder joint.

[0012] In a preferred embodiment, when the surface texture feature and the thermodynamic feature of the solder joint in the graphics card are obtained by aligning the gray scale normalized visible light image of the graphics card and the noise suppressed infrared thermal distribution map of the graphics card position by position, the method comprises:

[0013] extracting texture key points in the visible light image and thermal gradient vector fields of each pixel point in the infrared thermal distribution map;

[0014] filtering abnormal curvature features in the thermal gradient vector fields to obtain an effective solder joint region of the graphics card;

[0015] aligning the texture key points and the thermal gradient vector fields according to the effective solder joint region to obtain the surface texture feature and the thermodynamic feature of the solder joint in the graphics card.

[0016] In a preferred embodiment, the filtering of the abnormal curvature features in the thermal gradient vector fields to obtain the effective solder joint region of the graphics card comprises:

[0017] divergence value points in the thermal gradient vector fields exceeding a preset curvature threshold are taken as isotherm curvature extreme points of the solder joint of the graphics card;

[0018] regions where the isotherm curvature extreme points coincide with the texture key points are taken as the effective solder joint region of the graphics card.

[0019] In a preferred embodiment, the splicing of the surface texture feature and the thermodynamic feature to obtain the correlation feature weight matrix of the solder joint of the graphics card comprises:

[0020] splicing the surface texture feature and the thermodynamic feature along the feature dimension to obtain a high-dimensional tensor of the solder joint of the graphics card;

[0021] suppressing high-frequency noise in the high-dimensional tensor;

[0022] dynamically allocating weights of the surface texture feature and the thermodynamic feature according to the suppressed high-frequency noise to obtain a mask coefficient of the surface texture feature and the thermodynamic feature;

[0023] stacking the mask coefficient and spatial feature vectors of the high-dimensional tensor according to pixel point positions to obtain a correlation feature weight matrix of the solder joint of the graphics card.

[0024] In a preferred embodiment, the extracting the texture feature vector and the thermodynamic feature vector of the solder region in the correlation feature weight matrix according to the solder region of the graphics card comprises:

[0025] Driving the sampling points of the correlation feature weight matrix to adaptively offset in the solder deformation direction to obtain a sampling feature map of the graphics card solder;

[0026] Taking the affine transformed sampling coordinate point as a reference, extracting the feature value within the contour boundary of the graphics card solder from the sampling feature map, and taking the feature value as the texture feature vector and the thermodynamic feature vector set of the graphics card solder;

[0027] Selecting the feature vector with a thermodynamic confidence weight greater than a threshold value in the thermodynamic feature vector set as the thermodynamic feature vector of the graphics card solder.

[0028] In a preferred embodiment, the calculation formula of the thermodynamic confidence weight is:

[0029]

[0030] In the formula, is the thermodynamic confidence weight, is the thermal conductivity variation, is the temperature gradient module, is the specific heat capacity, is the material density.

[0031] In a preferred embodiment, the splicing the texture feature vector and the thermodynamic feature vector according to the channel dimension to obtain the fusion feature matrix of the graphics card solder comprises:

[0032] Adding a corresponding channel identifier to the texture feature vector and the thermodynamic feature vector;

[0033] Connecting the texture feature vector and the thermodynamic feature vector with the channel identifier in the channel direction to obtain an initial fusion tensor of the graphics card solder;

[0034] Flattening the spatial dimension of the initial fusion tensor into a column vector, and splicing the column vector with the channel dimension of the initial fusion tensor as a row index to obtain the fusion feature matrix of the graphics card solder.

[0035] In a preferred embodiment, the positioning the coordinates of the false solder in the solder region based on the fusion feature matrix and a standard solder feature library comprises:

[0036] Extracting the cosine similarity of the fusion feature matrix and the standard feature library;

[0037] detecting an abnormal region with the cosine similarity lower than a preset threshold, eliminating a concave region in the abnormal region, and connecting adjacent abnormal regions to obtain an expansion region of the GPU solder joint;

[0038] taking a center coordinate of a circumscribed rectangle of the expansion region as a coordinate of a fake solder joint in the solder joint region.

[0039] In a preferred embodiment, the generating of the solder joint fake solder defect report of the GPU based on the coordinate and the texture fracture index of the fake solder joint comprises:

[0040] converting the coordinate into a physical position of the fake solder joint in a coordinate system of a circuit board;

[0041] calibrating a profile fracture amount of the GPU solder joint based on the physical position to obtain a texture fracture index of the fake solder joint;

[0042] associating the texture fracture index with a standard defect level to obtain the solder joint fake solder defect report of the GPU.

[0043] In a preferred embodiment, the calibrating of the profile fracture amount of the GPU solder joint based on the physical position to obtain the texture fracture index of the fake solder joint comprises:

[0044] obtaining an edge line of the fake solder joint through connection of multi-level gradient filtering and an adaptive threshold;

[0045] identifying a continuity interruption section of the edge line, and accumulating a geometric distance of pixels between end points of the interruption section to obtain a fracture length value;

[0046] taking a ratio of the fracture length and a solder joint perimeter as the texture fracture index of the fake solder joint.

[0047] Compared with the prior art, the present application has the following beneficial effects:

[0048] 1. By comparing the gray scale normalized visible light image and the noise suppressed infrared thermal distribution map position by position, combining the accurate alignment of texture key points and thermal gradient vector field, and effectively filtering abnormal curvature features to determine the effective solder joint region, the accuracy of the extraction of the solder joint surface texture features and thermodynamic features is improved. At the same time, by dynamically allocating feature weights to form an associated feature weight matrix, the correlation between different features is enhanced, and the extracted solder joint features are more in line with the actual state, providing a more reliable basis for subsequent detection.

[0049] 2. In the feature vector extraction, an adaptive offset sampling is adopted to accurately obtain the texture and thermodynamic feature vectors of the solder joint area, the virtual solder coordinates are located by combining the cosine similarity comparison, and the defect level is calibrated through the texture fracture index, thereby improving the positioning accuracy of the virtual solder joint and the accuracy of the defect evaluation. The generated virtual solder defect report of the solder joint can clearly reflect the virtual solder position and defect degree, thereby enhancing the accuracy of the detection result and being helpful for efficiently identifying the virtual solder problem of the graphics card. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 A flowchart of a graphics card solder virtual solder automatic detection method based on image recognition provided by an embodiment of the present application is shown.

[0051] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein are merely intended to explain the present application and not to limit the present application.

[0053] The embodiment of the present application provides a graphics card solder virtual solder automatic detection method based on image recognition. The execution subject of the graphics card solder virtual solder automatic detection method based on image recognition includes but is not limited to at least one of electronic devices such as a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the present application. In other words, the graphics card solder virtual solder automatic detection method based on image recognition can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster and the like. The server can be a stand-alone server, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and big data and artificial intelligence platforms and the like basic cloud computing services.

[0054] Referring to Figure 1 A flowchart of a graphics card solder virtual solder automatic detection method based on image recognition provided by an embodiment of the present application is shown. In the embodiment, the graphics card solder virtual solder automatic detection method based on image recognition includes:

[0055] In the embodiment of the present application, when the visible light image of the graphics card after gray scale normalization and the infrared thermal distribution map of the graphics card after noise suppression are compared position by position, the surface texture features and thermodynamic features of the solder joints in the graphics card are obtained, and the surface texture features and thermodynamic features are specifically used for:

[0056] extracting texture key points in the visible light image and thermal gradient vector field of each pixel point in the infrared thermal distribution map;

[0057] filtering abnormal curvature features in the thermal gradient vector field to obtain an effective solder region of the graphics card;

[0058] aligning the texture key points and the thermal gradient vector field according to the effective solder region to obtain surface texture features and thermodynamic features of the solder of the graphics card.

[0059] Specifically, the texture key points in the visible light image are extracted to realize preliminary classification and identification by using the uniqueness of different object surface textures, and to remain stable under interference such as light change, scale scaling, slight rotation, etc., to provide reliable anchor points for subsequent processing, and to focus on local significant texture area to retain microstructure details of the object surface; the thermal gradient vector field of each pixel point in the infrared thermal distribution map is extracted to quantify the spatial features of the thermal distribution from the dynamic dimension of temperature, and the size of the vector can intuitively reflect the severity of temperature change, which can quickly locate abnormal heat areas, and the direction indicates the dominant transmission direction of heat flow to analyze the heat source diffusion path and heat conduction efficiency, and the distribution mode of the vector field can also distinguish stable heat field and dynamic heat field to provide quantitative basis for thermal state analysis.

[0060] Specifically, by eliminating the distorted curvature information in the vector field caused by noise, interference or abnormal thermal characteristics of non-solder regions, the thermal gradient distribution law specific to the solder region is accurately focused on, false features caused by uneven heat dissipation, external heat source interference and other factors are excluded, and the interference of invalid data on subsequent analysis is reduced, at the same time, the continuity and consistency features of the thermal gradient vector of the solder region are strengthened to provide clear thermal feature basis for accurately defining the spatial range of the effective solder, and to ensure that the extracted solder region has high purity and integrity.

[0061] Specifically, by aligning the texture key points and the thermal gradient vector field, the surface texture features and thermodynamic features of the graphics card solder are obtained, by matching the region range, the texture information from the visible light image and the thermal information of the infrared thermal distribution map are accurately associated to the same solder region, eliminating the spatial misplacement caused by the differences in collection angle and scale between the two modal data, making the structure details of the solder surface reflected by the texture key points and the temperature change law embodied by the thermal gradient vector field form a one-to-one spatial mapping relationship, realizing the fusion expression of double features on a single solder region, and providing complete and related basic data for subsequent feature analysis.

[0062] Further, the visible light texture key points provide static structural information such as the geometric shape and surface details of the target, which are used for positioning, contour extraction and identity recognition of the target, and the infrared thermal gradient vector field supplements the temperature variation characteristics of the target, which is used for analyzing the functional state of the target, especially in insufficient light, shielding and other visible light failure scenarios, and can independently provide a judgment basis in the thermal dimension; the two work together through multi-modal feature fusion to overcome the limitations of a single modality and improve the reliability of the scheme in complex scenarios.

[0063] Further, to provide accurate regional positioning basis for subsequent, through filtering of effective welding point area obtained by abnormal curvature characteristics, the analysis range is focused on the key connection parts of the display card, which provides a clear target area for subsequent welding point state evaluation, avoids misjudgment caused by invalid area interference, and at the same time, the area serves as a related node of thermal and structural characteristics, which can connect the extraction results of the texture key points and the thermal gradient vector field in the early stage, and provide specific action objects for multi-modal feature fusion analysis.

[0064] Further, the texture key points and the thermal gradient vector field extracted in the early stage are spatially aligned, so that the scheme can obtain the structural health and thermal stability of the welding point at the same time, break the limitations of single modal feature analysis, and provide more comprehensive judgment basis for the scheme, and the aligned features can be directly used as the input of subsequent classification, recognition or fault diagnosis model, thereby improving the accuracy and reliability of the scheme for welding point state evaluation.

[0065] In summary, visible light is vulnerable to environmental lighting in special scenarios, and infrared lacks texture details, while the extraction of texture key points and thermal gradient vector field realizes two-dimensional feature coverage of structural texture and temperature dynamics, provides complete data basis for subsequent fusion analysis, and improves feature expression capability; it provides a reusable technical paradigm for cross-modal feature fusion, establishes a double-modal feature extraction standard by clearly defining the difference between feature dimensions, and provides a theoretical reference for feature fusion in more complex multi-modal scenarios.

[0066] In summary, the abnormal feature filtering improves the signal-to-noise ratio of the thermal gradient vector field data, solves the problem of multiple thermal distribution interference in complex electronic equipment and the problem that effective features are easily hidden, and perfects the fine processing method from raw thermal data to target area, thereby providing technical support for high-precision thermal feature analysis, and the accurate extraction of the effective welding point area can improve the efficiency and accuracy of welding point quality detection, discover potential connection faults in time, provide accurate guidance for fault troubleshooting, and reduce costs and time.

[0067] Overall, the accurate alignment of the texture and thermal features by the effective area constraint provides technical support for the multi-feature collaborative analysis of the solder joints, improves the utilization efficiency and analysis depth of the feature data, and meanwhile, the double-feature abnormal association analysis in the video card maintenance link can quickly locate the faulty solder joint and shorten the troubleshooting time.

[0068] In the embodiment of the present application, when the abnormal curvature features in the thermal gradient vector field are filtered to obtain the effective solder joint area of the video card, the following is specifically used:

[0069] The divergence value points in the thermal gradient vector field that exceed the preset curvature threshold are taken as the isotherm curvature extreme points of the video card solder joint;

[0070] The area where the isotherm curvature extreme points coincide with the texture key points is taken as the effective solder joint area of the video card.

[0071] Specifically, by setting a scientific curvature threshold to screen the divergence value points in the thermal gradient vector field, the key points with significantly abnormal curvature changes in the isotherm of the solder joint are accurately identified. These points can intuitively reflect the mutation characteristics of the temperature distribution of the solder joint area and are the landmark positions of heat flow concentration or disorder. By extracting such extreme points, the flat area information in the thermal gradient vector field can be removed, and the thermal abnormal feature points with diagnostic value can be focused on, providing a clear target object for subsequent analysis of the thermal stability of the solder joint.

[0072] Specifically, the isotherm curvature extreme points reflect the significant changes in the thermal characteristics of the solder joint, and the texture key points mark the key texture features of the solder joint surface structure. The overlapping area of the two means the area that has both significant thermal characteristics and unique structural features. This screening process can effectively exclude interference areas with only single modal features, ensuring that the extracted effective solder joint area has clear solder joint features in terms of structure and thermal properties, providing a high-reliability target area for subsequent analysis.

[0073] Further, to provide key thermal abnormality identification basis for subsequent analysis, by determining the isotherm curvature extreme points, the thermal feature analysis of the solder joint area can be focused on the local mutation points from the overall distribution. The scheme can compare the distribution density, position characteristics of these points with the preset standard to quickly lock the solder joint area that may have quality problems, provide accurate thermal abnormality positioning reference for the subsequent fusion analysis of texture features and thermal features, and enhance the early identification ability of the scheme for the solder joint fault.

[0074] Further, by locking the overlapping area of the isotherm curvature extremum point and the texture key point, a welding point area definition with structural integrity and thermal effectiveness is provided for subsequent analysis, enabling the scheme to focus on the welding point area that truly has functional significance, avoiding analysis deviation caused by ambiguous area definition, and at the same time, this area as the intersection of structural features and thermal features provides a precise spatial range for subsequent fusion analysis of the correlation between surface texture defects and thermal anomalies of the welding point, improving the accuracy of the scheme for judging the quality of the welding point.

[0075] In general, a quantitative screening standard for effective feature points in a thermal gradient vector field is established, and the automatic and accurate extraction of thermal anomaly points is realized by predefining a curvature threshold, solving the problems of large data volume and difficulty in stripping effective information in the thermal gradient vector field, and improving the efficiency and pertinence of thermal feature analysis.

[0076] In general, the spatial coincidence verification of thermal features and structural features solves the problem that single modal feature positioning is easily disturbed, improves the accuracy and robustness of effective area extraction, and perfects the definition technology of key areas of complex electronic components; at the same time, accurate effective welding point area definition can improve the judgment efficiency of welding point quality, reduce false positives and false negatives, reduce production costs, and quickly lock the welding point position that needs to be detected.

[0077] In the embodiment of the present application, when the surface texture feature and the thermodynamic feature are spliced to obtain the correlation feature weight matrix of the graphics card welding point, it is specifically used for:

[0078] Splice the surface texture feature and the thermodynamic feature along the feature dimension to obtain the high-dimensional tensor of the graphics card welding point;

[0079] Suppress high-frequency noise in the high-dimensional tensor;

[0080] According to the dynamic allocation of the suppressed high-frequency noise, the weights of the surface texture feature and the thermodynamic feature are obtained to obtain the mask coefficient of the surface texture feature and the thermodynamic feature;

[0081] Stack the mask coefficient and the spatial feature vector of the high-dimensional tensor according to the pixel point position to obtain the correlation feature weight matrix of the graphics card welding point.

[0082] Specifically, the originally dispersed two modal features are structured and integrated, and a unified data carrier containing welding point surface microstructure information and thermal dynamic information is formed through expansion of the feature dimension. This splicing method retains the original attributes and details of each feature, establishes the correlation between different feature dimensions, avoids the problem of isolated single feature dimension information, enables the high-dimensional tensor to comprehensively and compactly present the multi-dimensional feature panorama of the welding point, and provides a structured and high-information data basis for subsequent depth analysis.

[0083] Specifically, by suppressing high-frequency noise in the high-dimensional tensor, interference information can be effectively eliminated, making the original data of surface texture features and thermodynamic features purer, and reducing the adverse effects of noise on feature analysis. At the same time, based on the dynamic allocation of the weights of the two features and the mask coefficients after suppressing high-frequency noise, the weight allocation can be more in line with the actual situation of the data, avoiding the problem of unreasonable feature emphasis caused by fixed weights.

[0084] Specifically, the mask coefficients and the spatial feature vectors of the high-dimensional tensor are stacked according to the pixel point positions, which can organically integrate the mask coefficients reflecting the importance of surface texture features and thermodynamic features with the spatial feature vectors embodying spatial position information, forming a matrix containing feature weights and spatial position correlation. The precise matching of weight information and spatial information is realized, and the feature weight of each pixel point can correspond to its spatial position.

[0085] Further, by constructing a high-dimensional tensor to organically integrate surface texture features and thermodynamic features, the scheme can simultaneously utilize the structural features and thermal features of the solder joint for joint analysis, breaking through the limitations of single feature analysis, supporting subsequent feature mining, pattern recognition or fault classification, etc., providing data support for the scheme to output accurate evaluation results, and improving the analysis ability of the scheme for complex solder joint states.

[0086] Further, the optimized feature weights and mask coefficients are provided for subsequent analysis, decision-making and other links based on surface texture features and thermodynamic features, ensuring that the subsequent links can more accurately utilize effective features, and improving the reliability and effectiveness of the entire scheme.

[0087] Further, the mask coefficients obtained in the early stage are integrated with the spatial feature vectors into an associated feature weight matrix, which not only retains the weight information of different features, but also maintains the spatial distribution relationship of the pixel points, so that subsequent feature extraction, anomaly detection and other operations of the graphics card solder joint can be based on complete spatial-weight correlation data, ensuring the coherence of the analysis process and the integrity of the data.

[0088] In summary, the high-dimensional tensor form preserves the integrity and relevance of the features, provides an adaptive input form for deep learning, etc., promotes the technical leap of solder joint feature analysis from single dimension to multi-dimension and deep level, solves the problem of difficult collaborative analysis of multi-source features, and at the same time, the high-dimensional tensor can provide more comprehensive feature basis for the automatic detection system, improving the accuracy of solder joint quality grading.

[0089] Overall, the suppression of high-frequency noise ensures the quality of the feature data, laying the foundation for accurate extraction and analysis of surface texture and thermodynamic features, and helping to dig out more real and valuable information. At the same time, dynamic allocation of weights and mask coefficients are obtained, which realizes flexible adjustment of the importance of different features, so that the scheme can better adapt to the dynamic changes of data, and improve the adaptability and accuracy of the scheme in processing complex, variable and high-dimensional tensor data.

[0090] Overall, it provides accurate data support for identifying key areas of welds and judging weld quality. It avoids the fragmentation of weight information and spatial information, allowing subsequent analysis to focus more on important features and key locations, thereby improving the accuracy and efficiency of GPU weld detection and evaluation.

[0091] In the embodiment of the application, when the texture feature vector and the thermodynamic feature vector of the weld area in the associated feature weight matrix are extracted according to the weld area of the GPU, it is specifically used for:

[0092] Driving the sampling points of the associated feature weight matrix to adaptively offset in the deformation direction of the weld, to obtain a sampling feature map of the GPU weld;

[0093] Taking the affine transformed sampling coordinate points as the reference, extracting feature values within the contour boundary of the GPU weld from the sampling feature map, and taking the feature values as the texture feature vector and the thermodynamic feature vector set of the GPU weld;

[0094] Selecting the feature vector with a thermodynamic confidence weight greater than a threshold value in the thermodynamic feature vector set as the thermodynamic feature vector of the GPU weld.

[0095] Specifically, by driving the sampling points of the associated feature weight matrix to adaptively offset in the deformation direction of the weld, the sampling points can be more accurately focused on the key area of the weld deformation. This offset is not random, but is dynamically adjusted according to the actual deformation direction of the weld, so that the sampling points can follow the deformation part, thereby capturing more information that can reflect the deformation characteristics of the weld. The sampling feature map generated finally can more accurately present the state and details of the weld deformation.

[0096] Specifically, the interference of irrelevant information outside the contour is excluded to ensure that the extracted feature values are completely derived from the weld itself, and the feature values are clearly distinguished as texture feature vectors and thermodynamic feature vector sets, realizing clear separation and accurate extraction of different types of features, and providing pure and classified data for subsequent feature analysis.

[0097] Specifically, the feature vectors with low confidence and insufficient reliability are removed, ensuring that the final selected thermodynamic feature vectors have high reliability and representativeness, avoiding interference of low-quality feature vectors on subsequent analysis, thereby improving the effectiveness and accuracy of feature data.

[0098] Further, the adaptive offset processing of the sampling points combines the abstract weight-related information with the actual deformation characteristics of the welding points, providing intuitive and targeted feature image data for subsequent analysis and evaluation of the welding point quality, playing a role in data conversion and information deepening, and ensuring that the scheme can transition from abstract data to concrete feature expression.

[0099] Further, the key region information focused on by the adaptive offset of the sampling points is converted into structured texture and thermodynamic feature vector sets through accurate extraction of the coordinate reference after affine transformation, providing directly usable feature data for subsequent defect detection of the graphics card welding points.

[0100] Further, threshold screening is used to purify the feature vectors, providing high-quality thermodynamic feature data for subsequent comprehensive analysis combining texture feature vectors and thermodynamic feature vectors, playing a role in optimizing data input and improving subsequent analysis efficiency.

[0101] In summary, the offset of the sampling points in the deformation direction can highlight the features of the deformation region of the welding points, avoiding the omission of effective information caused by the fixation of the sampling points in non-key regions, and improving the targeting and effectiveness of feature extraction; meanwhile, the adaptive offset feature enables the analysis to adapt to different deformation conditions of different welding points, enhancing the adaptability of the scheme to complex and variable welding point states. Through the generated sampling feature map, it can more accurately determine whether the welding point deformation is within a reasonable range and whether there are potential defects, providing a reliable basis for quality detection and performance guarantee of the graphics card welding points.

[0102] In summary, the extraction of feature values within the contour based on the coordinate points after affine transformation ensures the spatial accuracy and regional targeting of the features, avoids feature mis-extraction caused by coordinate deviation, improves the reliability of feature data, and distinguishes feature values into texture and thermodynamic feature vector sets, enabling subsequent analysis to respectively evaluate the appearance texture state and thermal performance characteristics of the welding points, which is helpful for more comprehensive and detailed judgment of the welding point quality.

[0103] In summary, the correlation between feature data and actual thermal state of the welding points is enhanced, providing a reliable basis for accurately evaluating key indicators such as thermal stability and heat dissipation performance of the welding points, while threshold screening reduces the number of feature vectors, reduces the complexity of subsequent data processing, and improves the running efficiency of the scheme.

[0104] In the embodiment of the present application, the calculation formula of the thermodynamic confidence weight is:

[0105]

[0106] In the formula, is the thermodynamic confidence weight, is the change of thermal conductivity, is the temperature gradient module, is the specific heat capacity, is the material density.

[0107] Specifically, is the inherent property of the material's ability to conduct heat, is the change value of the thermal conductivity under the change of the material state, and the temperature gradient is the vector describing the rate of change of temperature in space, and the module is the scalar thereof, is the heat required for the temperature of unit mass of matter to rise by 1℃, is the mass of unit volume of matter.

[0108] Further, reflects the correlation between thermal properties and heat transfer driving force: embodies the change of thermal conductivity, embodies the degree of non-uniformity of the temperature field, and the product of the two is the coupling strength of the change of thermal properties and the activity of heat transfer. is the volumetric specific heat capacity, reflecting the material's ability to store heat: The greater it is, the stronger the material's ability to store heat, and the more obvious the buffering effect on thermal processes.

[0109] Overall, is positively correlated with . When it increases, i.e., the material's thermal conductivity fluctuates more violently, increases; When it decreases, i.e., the thermal conductivity is more stable, decreases. is positively correlated with . The greater the temperature gradient, i.e., the stronger the heat transfer driving force, the greater it is; the smaller the temperature gradient, i.e., the more uniform the temperature field, the smaller it is. is negatively correlated with . The greater the specific heat capacity, i.e., the stronger the material's heat storage ability, the smaller it is; the smaller the specific heat capacity, i.e., the weaker the heat storage ability, the greater it is. is negatively correlated with The greater the density, that is, the greater the mass per unit volume, the greater the indirect enhancement of the heat storage capacity, The smaller the density, that is, the smaller the mass per unit volume, The greater the density, that is, the greater the mass per unit volume, the greater the indirect enhancement of the heat storage capacity, The greater the change trend of the thermal conductivity, the greater the temperature gradient modulus, the greater the specific heat capacity, and the greater the density.

[0110] In the embodiment of the present application, when the texture feature vector and the thermodynamic feature vector are spliced in the channel dimension to obtain the fusion feature matrix of the graphics card solder joint, it is specifically used for:

[0111] Adding corresponding channel identifiers to the texture feature vector and the thermodynamic feature vector;

[0112] Connecting the texture feature vector and the thermodynamic feature vector with channel identifiers in the channel direction to obtain the initial fusion tensor of the graphics card solder joint;

[0113] Flattening the spatial dimension of the initial fusion tensor into a column vector, and splicing the column vector with the channel dimension of the initial fusion tensor as the row index to obtain the fusion feature matrix of the graphics card solder joint.

[0114] Specifically, the texture feature vector and the thermodynamic feature vector are respectively added with corresponding channel identifiers, realizing the clear distinction and identification of the two different types of feature vectors. Through the exclusive channel identifier, the category to which each feature vector belongs can be accurately defined, avoiding feature confusion in subsequent processing, ensuring that the properties of the feature vector are traceable, and providing an intuitive basis for the separate call of the feature.

[0115] Specifically, the information of the two types of features is retained, and they form a unified whole through the association of the channel dimension, so that the originally dispersed texture and thermodynamic features can be associated with each other in the same data structure, providing a complete data carrier for subsequent comprehensive analysis of the features.

[0116] Specifically, the tensor form originally containing spatial and channel information is converted into a more concise and regular matrix form, which not only retains the feature details of the spatial dimension and the category distinction of the channel dimension, but also makes the data present a clear structure feature, facilitating subsequent matrix operation and data analysis processing.

[0117] Further, the two types of features are marked by adding channel identifiers, providing a structured basis for subsequent fusion, analysis or input into the model, ensuring that different types of features can be accurately recognized and processed in subsequent links.

[0118] Further, the two types of features are converted from an independent state to a fusion state through a connection operation, providing basic data for further processing of the fusion tensor, and being an important transition from feature separation processing to feature comprehensive analysis.

[0119] Further, the fusion feature matrix is converted through structural reorganization, providing an adaptive data format for subsequent feature extraction, model training or quality evaluation based on matrix operation, ensuring smooth connection of the data processing process, and enabling the pre-fused feature information to participate in subsequent analysis in a more operable form.

[0120] In summary, the addition of the channel identifier makes the boundary between the texture feature and the thermodynamic feature clearer, facilitating targeted attention to a certain type of feature and comparative study of the two types of features in subsequent analysis, improving the flexibility and accuracy of feature processing; at the same time, this standardized identification method lays a foundation for efficient management and application of feature vectors, and can reduce errors caused by feature confusion, and has an important supporting role for accurate analysis and application of GPU solder features.

[0121] In summary, the formation of the initial fusion tensor breaks down the barriers between texture features and thermodynamic features, enabling the information of the two to work together, which helps to uncover associated information that is difficult to find in single feature analysis, improves the comprehensiveness of GPU solder feature description, and the channel connection method preserves the channel identifier information of the features, enabling the use of the overall features after fusion in subsequent analysis, and tracing back to the original texture or thermodynamic features through the channel.

[0122] In summary, the complexity of subsequent data processing is reduced, enabling the matrix to be directly applied to feature analysis, improving the efficiency of data processing, while the spatial and channel information in the initial fusion tensor is completely preserved, ensuring the integrity and correlation of the features, enabling subsequent analysis to be based on comprehensive feature data.

[0123] In the embodiment of the present application, when the coordinates of the virtual solder joint in the solder joint region are located based on the fusion feature matrix and the standard solder joint feature library, it is specifically used for:

[0124] Extracting the cosine similarity of the fusion feature matrix and the standard feature library;

[0125] Detecting abnormal regions with a cosine similarity lower than a preset threshold, eliminating concave regions in the abnormal regions, and connecting adjacent abnormal regions to obtain an inflation region of the GPU solder joint;

[0126] Taking the center coordinates of the circumscribed rectangle of the inflation region as the coordinates of the virtual solder joint in the solder joint region.

[0127] Specifically, the cosine similarity can effectively reflect the similarity of two vectors in the direction, and is not affected by the length of the vector. Here, the degree of fit between the analyzed graphics card solder joint fusion features and the standard feature library can be accurately measured, and the similarity value obtained can be used as a direct quantitative index to determine whether the solder joint features meet the standard.

[0128] Specifically, first, accurately locate the abnormal area with a cosine similarity lower than the preset threshold. These areas are often where the graphics card solder joint may have defects. Then, eliminate the concave areas in the abnormal area to avoid interference from the details of the concave part in the overall abnormal area judgment, making the area shape more regular. Next, connect adjacent abnormal areas to integrate scattered small abnormal areas into complete large areas, and finally obtain the inflation area.

[0129] Specifically, by taking the center coordinates of the circumscribed rectangle of the inflation area as the virtual solder joint coordinates, the irregular shape of the inflation area can be ignored, and a unified coordinate point is used to lock the virtual solder joint position, avoiding positioning ambiguity caused by complex area shape, and ensuring that each virtual solder joint area has a unique and easily identifiable position reference.

[0130] Further, with the help of cosine similarity calculation with the standard feature library, abstract feature data is converted into similarity results with clear judgment significance, providing a key basis for subsequent judgment of graphics card solder joint quality, defect identification, etc.

[0131] Further, by detecting and processing the abnormal area, abstract similarity data is converted into concrete abnormal area shape, providing a clear area basis for subsequent accurate identification of graphics card solder joint defects, severity assessment, etc., and playing a transitional role from abnormal judgment to defect visualization.

[0132] Further, the inflation area is converted into specific coordinate information, providing direct position guidance for subsequent virtual solder joint confirmation, repair operations, etc., and is a key link connecting abnormal area analysis and actual production application.

[0133] Overall, the extraction of cosine similarity achieves objective and quantitative comparison of graphics card solder joint features and standard features, avoiding subjective judgment bias and improving the accuracy and consistency of quality assessment. At the same time, this quantitative method can sensitively capture subtle differences between features, helping to timely discover potential defects or deviations from the standard in the solder joint.

[0134] Overall, detecting and processing the abnormal area can accurately lock the problem area of the graphics card solder joint, and eliminating the concave area and connecting adjacent areas can avoid missed detection or misjudgment, improving the accuracy and integrity of abnormal area identification. At the same time, the inflation area obtained directly shows the range and shape of the solder joint anomaly, providing a reliable reference for analyzing the causes of defects and developing repair solutions.

[0135] Overall, the unified coordinate positioning method ensures the standardization and traceability of the detection results, facilitates statistical analysis of false welding problems, and provides data support for reducing the incidence of false welding. It upgrades false welding detection from qualitative judgment to quantitative positioning.

[0136] In the embodiment of the application, when the GPU's solder false welding defect report is generated according to the coordinates and the texture fracture index of the false welding solder, it is specifically used for:

[0137] Converting the coordinates into a physical position of the false welding solder in a coordinate system of a circuit board;

[0138] Calibrating the profile fracture amount of the GPU's solder based on the physical position to obtain the texture fracture index of the false welding solder;

[0139] Correlating the texture fracture index with a standard defect level to obtain the GPU's solder false welding defect report.

[0140] Specifically, the digital coordinates are converted into specific positions that can directly correspond to the physical space of the circuit board. The differences between different coordinate systems are eliminated, and the coordinate information obtained by detection can correspond to the actual structure and component layout of the circuit board, providing a direct basis for finding the accurate position of the false welding solder on the physical circuit board.

[0141] Specifically, through physical position calibration, the calculation error of the profile fracture amount caused by position deviation is eliminated, and the obtained fracture amount is more in line with the actual situation. The texture fracture index converts the abstract profile fracture information into a quantifiable index, which directly reflects the degree of texture fracture of the false welding solder, and provides a specific basis for judging the severity of false welding.

[0142] Specifically, correlating the texture fracture index with the standard defect level can convert the quantified texture fracture data into a defect level with clear evaluation criteria, and finally form a systematic and standardized GPU solder false welding defect report. It realizes the conversion from specific data to a conclusive report, and the severity of false welding defects has a unified evaluation standard. The report content contains both quantitative basis and clear level division, which is easy to understand and use.

[0143] Further, through coordinate system conversion, it is mapped to the physical space of the circuit board, providing accurate position guidance for subsequent automated and other actual operations, completing the key transition from data information to physical position, and ensuring the practicality and operability of the whole scheme.

[0144] Further, by calibrating the profile fracture amount and calculating the texture fracture index, the physical position information is combined with the texture characteristics of the solder, providing feature data for subsequent quality rating and repair priority determination of the false welding solder.

[0145] Further, the association of the previously obtained texture fracture index and the standard defect grade is carried out to comprehensively judge the virtual welding condition of the graphics card solder joint and form a report, thereby providing a final decision basis for subsequent links.

[0146] In general, the virtual welding solder joint can be accurately found on the circuit board, the problems caused by the non-corresponding coordinates are avoided, the maintenance efficiency is improved, and the physical position information in the circuit board coordinate system is unified, thereby providing a spatial coordinate basis for tracing the virtual welding causes.

[0147] In general, the calibration based on the physical position ensures the accuracy of the profile fracture amount, the introduction of the texture fracture index realizes the quantitative evaluation of the texture characteristics of the virtual welding, the judgment of the virtual welding severity is upgraded from qualitative description to quantitative analysis, the objectivity and precision of the evaluation are improved, meanwhile, the index provides more detailed reference basis for subsequent analysis, and it is helpful to develop differentiated repair strategies according to the size of the fracture index.

[0148] In general, the deviation in information transmission is reduced, meanwhile, the texture fracture index and the standard defect grade associated in the report provide detailed reference data for tracing the root cause of the virtual welding problem, which is helpful to reduce the virtual welding rate, the clear defect grade also provides a basis for determining the maintenance priority, and the utilization efficiency of the maintenance resources is improved.

[0149] In the embodiment of the application, when the profile fracture amount of the graphics card solder joint is calibrated based on the physical position to obtain the texture fracture index of the virtual welding solder joint, the calibration is specifically used for:

[0150] The edge line of the virtual welding solder joint is obtained through multi-level gradient filtering and adaptive threshold connection.

[0151] The continuity interruption section of the edge line is recognized, and the geometric distance of the pixels between the interruption section endpoints is accumulated to obtain a fracture length value.

[0152] The ratio of the fracture length to the solder joint circumference is taken as the texture fracture index of the virtual welding solder joint.

[0153] Specifically, the multi-level gradient filtering is used to enhance the difference between the virtual welding solder joint edge and the surrounding area, highlight the edge features, and then the adaptive threshold is used to connect the edge segments, so that the complete and continuous virtual welding solder joint edge line is finally formed, the noise interference is effectively filtered, the edge details are accurately captured, the problems of edge fracture and blur under single filtering or fixed threshold are solved, and the profile boundary of the virtual welding solder joint is clear.

[0154] Specifically, the discontinuous feature of the edge line is converted into a specific numerical index, which clearly reflects the degree and range of edge fracture of the virtual welding spot, avoids the vague description of edge integrity, and provides an intuitive and comparable quantitative basis for judging the edge damage condition.

[0155] Specifically, the problem of absolute value of the fracture length being incomparable due to size difference of different welding spots is eliminated, so that the relative severity of the texture fracture of the welding spot can be measured by a unified index regardless of the size of the welding spot.

[0156] Further, the edge line is extracted to further depict the geometric shape of the virtual welding spot, thereby providing accurate edge basis for subsequent contour parameter calculation, defect shape classification and the like, and playing a transitional role from area positioning to shape quantification.

[0157] Further, the shape feature of the edge line is further converted into a quantifiable parameter, thereby providing important feature data for subsequent evaluation of the severity of the virtual welding defect and division of the defect grade, and improving the capture of multi-dimensional features of the virtual welding spot.

[0158] Further, the specific information of the edge fracture is converted into an index with clear evaluation significance, thereby providing a key quantitative basis for subsequent correlation with the standard defect grade and generation of a defect report, and playing a conversion role from original data to evaluation index.

[0159] In summary, the edge feature can analyze whether the welding spot has problems such as deformation and defect, thereby improving the fineness of defect recognition; meanwhile, the combination of the multi-level gradient filtering and the adaptive threshold value enables the process to adapt to the welding spot image under different illumination and material, thereby enhancing the adaptability of the scheme to complex scenes. The accurate extraction of the edge line provides a basis for subsequent calculation of the welding spot size and shape parameter, and is helpful for more scientific evaluation of the severity of the virtual welding defect, and plays an important supporting role in perfecting the GPU welding spot quality detection system.

[0160] In summary, the scheme is convenient for quickly distinguishing the severity of the defect, thereby providing data support for formulating a differentiated processing strategy, and meanwhile, the numerical value can accurately reflect the actual size of the edge fracture, thereby being helpful for in-depth analysis of the causes of the virtual welding.

[0161] In summary, the scheme objectively reflects the damage degree of the virtual welding on the texture integrity of the welding spot, thereby avoiding misjudgment caused by different sizes of the welding spot itself, improving the accuracy and fairness of defect evaluation, and meanwhile, the index provides a unified standard for transverse comparison between different virtual welding spots and longitudinal monitoring of different stages of the same welding spot, thereby being convenient for statistical analysis and trend judgment of the virtual welding problem in production.

[0162] Compared with the prior art, the scheme has the following beneficial effects:

[0163] 1. By comparing the gray scale normalized visible light image with the noise suppressed infrared thermal distribution, combined with the precise alignment of texture key points and thermal gradient vector field, the effective filter abnormal curvature feature is determined to determine the effective welding point area, which improves the accuracy of the extraction of the welding point surface texture feature and the thermodynamic feature. At the same time, by dynamically allocating feature weights to form an associated feature weight matrix, the correlation between different features is enhanced, making the extracted welding point features more in line with the actual state, providing a more reliable basis for subsequent detection.

[0164] 2. In the feature vector extraction, adaptive offset sampling is used to accurately obtain the texture and thermodynamic feature vectors of the welding point area, combined with cosine similarity comparison to locate the false welding coordinates, and through the texture fracture index to calibrate the defect level, which improves the accuracy of false welding point positioning and the accuracy of defect evaluation. The generated false welding defect report of the welding point can clearly reflect the false welding position and defect degree, enhance the accuracy of the detection result, and help to efficiently identify the false welding problem of the graphics card welding point.

[0165] In several embodiments provided by the present application, it should be understood that the disclosed method can be implemented in other ways.

[0166] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0167] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An image recognition-based automatic detection method for virtual soldering of a graphics card solder joint, characterized in that, The method comprises: S1, aligning the visible light image of the graphics card after gray scale normalization and the infrared thermal distribution map of the graphics card after noise suppression pixel by pixel to obtain the surface texture feature and the thermodynamic feature of the solder joint in the graphics card; S2, splicing the surface texture feature and the thermodynamic feature to obtain the correlation feature weight matrix of the solder joint of the graphics card, comprising: splicing the surface texture feature and the thermodynamic feature along the feature dimension to obtain the high-dimensional tensor of the solder joint of the graphics card; suppressing high-frequency noise in the high-dimensional tensor; dynamically allocating the weight of the surface texture feature and the thermodynamic feature according to the suppressed high-frequency noise to obtain the mask coefficient of the surface texture feature and the thermodynamic feature; stacking the mask coefficient and the spatial feature vector of the high-dimensional tensor according to the pixel point position to obtain the correlation feature weight matrix of the solder joint of the graphics card; S3, extracting the texture feature vector and the thermodynamic feature vector of the solder joint region in the correlation feature weight matrix according to the solder joint region of the graphics card, comprising: driving the sampling point of the correlation feature weight matrix to adaptively offset in the deformation direction of the solder joint to obtain the sampling feature map of the solder joint of the graphics card; taking the affine transformed sampling coordinate point as a reference, extracting the feature value within the contour boundary of the solder joint of the graphics card from the sampling feature map, and taking the feature value as the texture feature vector and the thermodynamic feature vector set of the solder joint of the graphics card; selecting the feature vector with a thermodynamic confidence weight greater than a threshold value in the thermodynamic feature vector set as the thermodynamic feature vector of the solder joint of the graphics card; S4, splicing the texture feature vector and the thermodynamic feature vector along the channel dimension to obtain the fusion feature matrix of the solder joint of the graphics card; S5, positioning the coordinates of the false solder joint in the solder joint region based on the fusion feature matrix and the standard solder joint feature library; S6, generating the solder joint false defect report of the graphics card according to the coordinates and the texture fracture index of the false solder joint.

2. The image recognition-based automatic detection method for virtual soldering of GPU solder joints according to claim 1, characterized in that, When aligning the visible light image of the graphics card after gray scale normalization and the infrared thermal distribution map of the graphics card after noise suppression pixel by pixel to obtain the surface texture feature and the thermodynamic feature of the solder joint in the graphics card, comprising: extracting the texture key point in the visible light image and the thermal gradient vector field of each pixel point in the infrared thermal distribution map; filtering the abnormal curvature feature in the thermal gradient vector field to obtain the effective solder joint region of the graphics card; aligning the texture key point and the thermal gradient vector field according to the effective solder joint region to obtain the surface texture feature and the thermodynamic feature of the solder joint in the graphics card.

3. The image recognition-based automatic detection method for virtual soldering of GPU solder joints according to claim 2, characterized in that, When filtering the abnormal curvature feature in the thermal gradient vector field to obtain the effective solder joint region of the graphics card, comprising: taking the divergence value point in the thermal gradient vector field exceeding the preset curvature threshold value as the isotherm curvature extreme point of the graphics card solder joint; taking the region where the isotherm curvature extreme point coincides with the texture key point as the effective solder joint region of the graphics card.

4. The image recognition-based automatic detection method for virtual soldering of GPU solder joints according to claim 1, characterized in that, The calculation formula of the thermodynamic confidence weight is: ； wherein, is the thermodynamic confidence weight, is the change in thermal conductivity, is the temperature gradient magnitude, is the specific heat capacity, is the material density.

5. The image recognition based automatic detection method for virtual soldering of GPU solder joints as claimed in claim 1 wherein, Splicing the texture feature vector and the thermodynamic feature vector along the channel dimension to obtain the fusion feature matrix of the solder joint of the graphics card, comprising: adding a corresponding channel identifier to the texture feature vector and the thermodynamic feature vector; connecting the texture feature vector and the thermodynamic feature vector with a channel identifier in a channel direction to obtain an initial fusion tensor of the GPU solder joint; flattening the spatial dimension of the initial fusion tensor into a column vector, and splicing the column vector with the channel dimension of the initial fusion tensor as a row index to obtain a fusion feature matrix of the GPU solder joint.

6. The image recognition based automatic detection method for virtual soldering of GPU solder joints as claimed in claim 1 wherein, The method for positioning the coordinates of the false solder joint in the solder joint region based on the fusion feature matrix and a standard solder joint feature library comprises: extracting the cosine similarity of the fusion feature matrix and the standard feature library; detecting an abnormal region with a cosine similarity lower than a preset threshold, eliminating concave regions in the abnormal region, and connecting adjacent abnormal regions to obtain an inflation region of the GPU solder joint; taking the center coordinates of the circumscribed rectangle of the inflation region as the coordinates of the false solder joint in the solder joint region.

7. The image recognition based automatic detection method for virtual soldering of GPU solder joints as claimed in claim 1 wherein, The method for generating a solder joint false solder defect report of the GPU according to the coordinates and the texture fracture index of the false solder joint comprises: converting the coordinates into a physical position of the false solder joint in a circuit board coordinate system; calibrating the contour fracture amount of the GPU solder joint based on the physical position to obtain the texture fracture index of the false solder joint; associating the texture fracture index with a standard defect level to obtain the solder joint false solder defect report of the GPU.

8. The image recognition-based automatic detection method for virtual soldering of GPU solder joints according to claim 7, characterized in that, The method for calibrating the contour fracture amount of the GPU solder joint based on the physical position to obtain the texture fracture index of the false solder joint comprises: obtaining an edge line of the false solder joint through multi-level gradient filtering and adaptive threshold connection; identifying a continuity interruption segment of the edge line, and accumulating the geometric distance of pixels between interruption segment endpoints to obtain a fracture length value; taking the ratio of the fracture length and the solder joint circumference as the texture fracture index of the false solder joint.

Images