CN111681235A - IC welding spot defect detection method based on learning mechanism - Google Patents
IC welding spot defect detection method based on learning mechanism Download PDFInfo
- Publication number
- CN111681235A CN111681235A CN202010529781.9A CN202010529781A CN111681235A CN 111681235 A CN111681235 A CN 111681235A CN 202010529781 A CN202010529781 A CN 202010529781A CN 111681235 A CN111681235 A CN 111681235A
- Authority
- CN
- China
- Prior art keywords
- sample
- welding spot
- samples
- qualified
- training set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003466 welding Methods 0.000 title claims abstract description 109
- 238000001514 detection method Methods 0.000 title claims abstract description 55
- 230000007547 defect Effects 0.000 title claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 101
- 238000011156 evaluation Methods 0.000 claims abstract description 55
- 238000013179 statistical model Methods 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 26
- 229910000679 solder Inorganic materials 0.000 claims description 24
- 238000007689 inspection Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 description 6
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005476 soldering Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24143—Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an IC welding spot defect detection method based on a learning mechanism, which comprises the following steps: establishing a local statistical model according to a plurality of qualified IC welding spot samples; collecting an IC welding spot sample, inputting the IC welding spot sample into a local statistical model, comparing to obtain a potential defect image as a training set, wherein the training set comprises a qualified sample and an unqualified sample; training the plurality of classifiers by using the training subsets; evaluating the training set by using a plurality of trained classifiers, determining the weight of each classifier, calculating the average disqualification probability of each sample, and calculating an evaluation threshold; the method comprises the steps of collecting an IC welding spot sample picture to be detected, obtaining the average unqualified probability of the IC welding spot sample by using a local statistical model and a plurality of classifiers, and comparing the average unqualified probability with an evaluation threshold value to obtain a detection result. The method provided by the invention adaptively sets the weight value and the evaluation threshold value of the classifier, improves the accuracy and the robustness of the IC welding spot defect detection, and has more reasonable evaluation result.
Description
Technical Field
The invention relates to the field of digital image processing application, in particular to an IC welding spot defect detection method based on a learning mechanism.
Background
Surface Mount Technology (SMT) is a Circuit connecting Technology in which a Surface-mounted device (SMC/SMD, or chip device, for short) with no leads or short leads is mounted on the Surface of a Printed Circuit Board or other substrate and then soldered or assembled by reflow soldering or dip soldering.
Integrated circuit IC packages are mounted or placed directly on the surface of a PCB using SMT techniques, and since the IC pins are bridges of the electrical connection between the IC components and the PCB, their quality is critical to the reliability and practicality of the PCB and the electronic device. With the rapid development of microelectronic manufacturing industry, the density of IC devices on a PCB is higher and smaller, so that manual detection of IC pads becomes difficult.
With the development of computer vision technology, Automated Optical Inspection (AOI) systems are widely used for product surface defect inspection to evaluate the quality of many industrial products. In order to realize AOI systems of different products, the detection method of the AOI system needs to be adjusted according to the characteristics of the object. At present, algorithms for detecting IC welding spot defects on an AOI system are few.
Most of the existing methods for detecting IC welding spot defects firstly model acceptable samples to obtain models of the acceptable samples, and then compare the models with other samples to obtain a binary image; the pixel is matched with the template and is 0 in the binary image, and the pixel is not matched with the template and is 1; and then, evaluating the sample by combining weighted summation of pixel points in the binary image of the sample to be detected.
In the existing scheme, only qualified samples are used as training samples, and the establishment of the model and the threshold are determined based on the qualified samples, so that the modes of unqualified samples are ignored, and the detection result can only determine the qualified sample mode learned in the training samples, namely, the detection result can only determine the qualified samples within a certain range, and the qualified samples exceeding the certain range are easily considered as unqualified samples.
In the existing scheme, the weights of pixel points are difficult to reasonably set, in the process of evaluating the quality of a sample, the pixel points of a sample binary image are generally subjected to weighted summation, and the determination mode of the weights is either artificially set according to prior knowledge or determined according to a defect frequency image. Both the former and the latter depend on the quality of the qualified sample model, and if the number of training samples of the qualified sample model is small, the detection result is liable to be deteriorated.
The evaluation method in the existing scheme lacks overall consideration, is too simple, is difficult to evaluate the quality of the sample on the whole, and is difficult to determine an evaluation threshold value, so that more training samples are needed to obtain a reliable detection result.
Disclosure of Invention
The invention aims to provide an IC welding spot defect detection method based on a learning mechanism, which aims to solve the technical problems that the quality of a sample is difficult to be integrally evaluated and an evaluation threshold value is difficult to determine in the existing method.
The purpose of the invention can be realized by the following technical scheme:
establishing a local statistical model according to a plurality of qualified IC welding spot samples;
collecting pictures of an IC welding spot sample, inputting the pictures into the local statistical model for comparison, and obtaining a potential defect image of the IC welding spot sample as a training set; the training set comprises qualified samples and unqualified samples, wherein the qualified samples are potential defect images of the qualified IC welding spot samples, and the unqualified samples are potential defect images of the unqualified IC welding spot samples;
respectively training a plurality of classifiers by using different training subsets; wherein the training subset consists of all unqualified samples and the same number of qualified samples in the training set;
respectively evaluating each sample in the training set by using the trained classifiers, determining the weight of each classifier, and calculating the average disqualification probability of each sample so as to obtain the evaluation threshold of the training set sample;
and acquiring an IC welding spot sample picture to be detected, obtaining the average unqualified probability of the IC welding spot sample by using the local statistical model and the plurality of classifiers, and comparing the average unqualified probability of the IC welding spot sample with the evaluation threshold value to obtain the detection result of the IC welding spot sample.
Optionally, determining the average probability of failure for the IC solder joint sample using the local statistical model and the plurality of classifiers further comprises: inputting the IC welding spot sample picture into the local statistical model to obtain a potential defect image of the IC welding spot sample, and evaluating the potential defect image by utilizing a plurality of classifiers to obtain the average failure probability of the IC welding spot sample.
Optionally, comparing the average failure probability of the IC solder joint sample with the evaluation threshold to obtain the detection result of the IC solder joint sample further comprises: and judging whether the average disqualification probability of the IC welding spot sample is smaller than the evaluation threshold value, if so, determining that the IC welding spot sample is qualified, otherwise, determining that the IC welding spot sample is disqualified.
Optionally, determining the weight of each of the classifiers further comprises calculating the weight of each of the classifiers using the following formula:
wherein ,wiWeight of the ith classifier, eiIndicating the accuracy of the ith classifier, emaxRepresenting the highest accuracy in the classifier, eminRepresenting the lowest accuracy in the classifier.
Optionally, the classifier is a KNN classifier.
Optionally, calculating the average probability of failure for each sample further comprises calculating the average probability of failure for each sample using the following equation:
wherein ,MUPXRepresenting the mean probability of failure, KNN, of sample Xi(X) represents the probability of the ith KNN classifier failing the sample X, and N represents the number of KNN classifiers.
Optionally, the evaluation threshold is calculated using the following formula:
UPT＝αUmin+(1+α)Qmax；
wherein α represents the proportion of qualified samples in the training set, UminRepresents the minimum value of MUP in the failed sample, QmaxRepresents the maximum value of MUP in the qualified samples, Q represents the qualified samples, U represents the unqualified samples, num ({ Q }) represents the number of qualified samples in the training set, and num ({ Q } + { U }) represents the number of samples in the training set.
Optionally, building the local statistical model from the plurality of qualified IC solder joint samples further comprises: and establishing a local statistical model, initializing the local statistical model by utilizing a first qualified IC welding spot sample, and updating the local statistical model according to a subsequent qualified IC welding spot sample.
Optionally, the local statistical model is a VIBE model.
Optionally, the training subset is composed of all unqualified samples and the same number of qualified samples in the training set, and further comprises: the qualified samples are randomly selected.
The invention provides an IC welding spot defect detection method based on a learning mechanism, which comprises the following steps: establishing a local statistical model according to a plurality of qualified IC welding spot samples; collecting pictures of an IC welding spot sample, inputting the pictures into the local statistical model for comparison, and obtaining a potential defect image of the IC welding spot sample as a training set; the training set comprises qualified samples and unqualified samples, wherein the qualified samples are potential defect images of the qualified IC welding spot samples, and the unqualified samples are potential defect images of the unqualified IC welding spot samples; respectively training a plurality of classifiers by using different training subsets; wherein the training subset consists of all unqualified samples and the same number of qualified samples in the training set; respectively evaluating each sample in the training set by using the trained classifiers, determining the weight of each classifier, and calculating the average disqualification probability of each sample so as to obtain the evaluation threshold of the training set sample; and acquiring an IC welding spot sample picture to be detected, obtaining the average unqualified probability of the IC welding spot sample by using the local statistical model and the plurality of classifiers, and comparing the average unqualified probability of the IC welding spot sample with the evaluation threshold value to obtain the detection result of the IC welding spot sample.
The IC welding spot defect detection method based on the learning mechanism has the following beneficial effects:
(1) based on a mechanism from local statistical modeling to global evaluation detection, the method not only can well grasp the characteristic information in the training set samples, but also can reveal the relation among the training set samples to globally evaluate the quality of the IC welding spot samples, so that the method can complete the detection task of the IC welding spot samples under the condition of insufficient and unbalanced training samples.
(2) The weight of the classifier is set in a self-adaptive mode, the self-adaptive weight can improve the influence of the classifier with good detection effect, and the influence of the classifier with poor detection effect on global evaluation is reduced, so that the robustness of the whole detection scheme can be improved, and the detection accuracy is also improved.
(3) The evaluation threshold value is set in a self-adaptive mode, and due to the fact that unqualified samples are added into the training set, the limit of the qualified samples and the limit of the unqualified samples can be considered at the same time, and evaluation is more reasonable. Meanwhile, the proportion of the qualified sample limit and the unqualified sample limit in the evaluation threshold is adjusted according to the proportion of the qualified samples in the training set, and the evaluation threshold is changed along with the composition proportion of the qualified samples and the unqualified samples in the training set, so that the evaluation result is more reasonable.
Drawings
FIG. 1 is a schematic diagram of a detection process of an IC solder joint defect detection method based on learning according to the present invention.
Detailed Description
The embodiment of the invention provides an IC welding spot defect detection method based on a learning mechanism, which aims to solve the technical problems that the quality of a sample is difficult to be integrally evaluated and an evaluation threshold value is difficult to determine in the existing method.
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The embodiment of the invention provides an IC welding spot defect detection method based on a learning mechanism, which comprises the following steps:
establishing a local statistical model according to a plurality of qualified IC welding spot samples;
collecting pictures of an IC welding spot sample, inputting the pictures into the local statistical model for comparison, and obtaining a potential defect image of the IC welding spot sample as a training set; the training set comprises qualified samples and unqualified samples, wherein the qualified samples are potential defect images of the qualified IC welding spot samples, and the unqualified samples are potential defect images of the unqualified IC welding spot samples;
respectively training a plurality of classifiers by using different training subsets; wherein the training subset consists of all unqualified samples and the same number of qualified samples in the training set;
respectively evaluating each sample in the training set by using the trained classifiers, determining the weight of each classifier, and calculating the average disqualification probability of each sample so as to obtain the evaluation threshold of the training set sample;
and acquiring an IC welding spot sample picture to be detected, obtaining the average unqualified probability of the IC welding spot sample by using the local statistical model and the plurality of classifiers, and comparing the average unqualified probability of the IC welding spot sample with the evaluation threshold value to obtain the detection result of the IC welding spot sample.
The detection flow of the embodiment of the invention is shown in fig. 1, please refer to fig. 1, and the embodiment of the invention provides an IC solder joint defect detection method based on a learning mechanism, based on a local to global detection mechanism, firstly detecting potential defect pixel points in an IC solder joint sample through a local statistical model, and then evaluating the quality of the IC solder joint sample through a plurality of KNN classifiers. Based on the mechanism, the method provided by the embodiment of the invention not only can grasp the local characteristic information of the IC welding spot sample, does not leave any detail for detecting the IC welding spot sample, but also can grasp the relation among the IC welding spot samples, and integrally evaluates the quality of the IC welding spot sample.
The IC welding spot defect detection method based on the learning mechanism provided by the embodiment of the invention mainly comprises two parts: (1) local statistical modeling; (2) and (4) a global evaluation model based on ensemble learning.
According to the embodiment of the invention, a local statistical model is obtained by modeling a series of qualified IC welding spot samples, and the obtained statistical model is a qualified sample mode because the IC welding spot samples used for training the local statistical model are all qualified IC welding spot samples. Inputting the detected IC welding spot sample into a local statistical model, and comparing the detected IC welding spot sample with the local statistical model to obtain a Potential Defect Image (PDI) of the IC welding spot sample.
In the global evaluation stage based on ensemble learning, training each KNN classifier by utilizing a training subset, wherein each KNN classifier is trained by one training subset, and the training subset is composed of balanced unqualified samples and qualified samples. Unbalanced samples may result in a classification result biased towards a high number of classes, and a balanced training set may enable the KNN classifier to be trained well. The trained KNN classifier can be used for detecting the IC welding spot sample.
In a global evaluation phase based on ensemble learning, global evaluation is carried out on PDI of the IC welding spot samples through a plurality of weighted K-neighbor (KNN) classifiers, wherein the KNN classifiers are trained by PDI images of different IC welding spot samples. And carrying out weighted average on the detection result of each KNN classifier to obtain a score, obtaining the average Unqualified Probability (MUP) of the input IC welding spot sample, comparing the MUP with the evaluation threshold of the training set, wherein if the MUP is larger than the evaluation threshold, the IC welding spot sample is Unqualified, and otherwise, the IC welding spot sample is qualified.
It is noted that the evaluation threshold here is determined according to the limits of the unqualified samples and the qualified samples in the training set (the unqualified minimum value and the qualified maximum value).
In the embodiment of the invention, the process of local statistical modeling is as follows:
the local statistical model is established according to a series of input qualified IC welding spot samples, and the embodiment of the invention adopts a VIBE model for modeling: firstly, initializing a local statistical model according to a first input qualified IC welding spot sample; then, the model is continuously updated according to the qualified IC welding spot samples input subsequently. The VIBE model is adopted because the VIBE modeling speed is high and the accuracy rate is high. Since the local model is subjected to global evaluation after being processed, the requirement on the local model is not high, and a local statistical model based on the VIBE model is sufficient.
Initializing a local statistical model: six templates with the same size as the qualified IC welding spot sample are extracted from the qualified sample for training, wherein pixel points of each template are randomly selected from pixel points at corresponding positions in the qualified IC welding spot sample or pixel points in the neighborhood of the qualified IC welding spot sample.
The number of templates is a parameter and is determined empirically. When the template is extracted, extracting from a first qualified IC welding spot sample, selecting pixel points from each pixel point or the neighborhood thereof, assigning the pixel points to the pixel points at the corresponding position in the template, wherein the extracted template is also an image, and the size of the extracted template is the same as that of the qualified IC welding spot sample.
Updating the model: when a certain pixel point in an input qualified IC welding spot sample is matched with a sufficient template, the pixel point is a qualified pixel point, and the pixel point or a neighborhood pixel point is randomly selected to update a pixel point at a corresponding position in the template or a neighborhood pixel point; and if the pixel point is not a qualified pixel point, updating the pixel point at the corresponding position in a unmatched template by using the pixel point.
It should be noted that the number of matched templates is also a parameter, and is set empirically, and if a certain pixel matches 2 templates out of 6 templates, it can be considered that the pixel matches enough templates.
Matching strategies: taking the pixel value x of a certain pixel point of the input IC pad sample as an example, taking the pixel value of the certain pixel point as a center, and R as a range, the range can be set according to experience and experiments, for example, taking R as 30. And if the pixel point at the corresponding position of a certain template is in the range, the template is considered to be matched with the pixel point. When the number of matched templates reaches a matching threshold, the pixel point is considered to be matched with the local statistical model, and the pixel point is considered to be a qualified pixel point. After the input IC welding spot sample is compared with the local statistical model, a binary image called as a Potential Defect Image (PDI) of the IC welding spot sample can be obtained, wherein the pixel value of the qualified pixel point is 0, and the pixel value of the unqualified pixel point is 1.
For example, if the pixel value of a gray-scale pixel is 10 and R is 5, and the pixel values of the pixels at the positions corresponding to the templates are 1, 10, 15, 20, 25, and 30, respectively, the templates of 10 and 15 match with the pixel, and the pixel is considered to be a qualified pixel.
In the embodiment of the invention, the global evaluation process based on ensemble learning is as follows:
in the global evaluation stage, the PDI map of the IC solder joint sample obtained by comparing the classifier with the local statistical model is used for training or detecting the classifier, and the PDI map of the IC solder joint sample includes a result of detecting pixel points in the IC solder joint sample picture by using the local statistical model, that is, a result of processing the detail features of the IC solder joint sample.
Firstly, inputting an IC welding spot sample picture into a local statistical model to obtain a PDI picture of the IC welding spot sample as a training set, wherein the training set comprises a PDI picture of a qualified IC welding spot sample (called a qualified sample) and a PDI picture of an unqualified IC welding spot sample (called an unqualified sample), and a balanced training subset consisting of a plurality of qualified samples and unqualified samples is obtained through downsampling; then, training the KNN classifier by using each training subset, verifying the trained KNN classifier in the whole training set, and calculating the weight of each KNN classifier; and finally, calculating the MUP of each sample in the training set, and solving the evaluation threshold value of the training set.
Down-sampling: because the number of unqualified samples in the training set is far smaller than that of qualified samples, the KNN classifier is trained by using the training set, so that the detection result is biased to the qualified samples, and the unqualified samples are difficult to detect. Therefore, qualified samples are randomly selected from the training set, and the qualified samples and the unqualified samples form balanced training subsets, each obtained balanced training subset may be different, and the number of the qualified samples and the number of the unqualified samples in each balanced training subset are the same.
Training a KNN classifier: and training each KNN classifier respectively by using the balance training subset obtained in the downsampling process, wherein each KNN classifier obtained by the training is different. The KNN classifier trained by the balance training subset obtained in the downsampling process can well discriminate unqualified samples, so that the problem of sample imbalance is solved.
Adaptively determining a weight for each KNN classifier: because each KNN classifier is different, the detection effect of the KNN classifiers needs to be evaluated, the KNN classifier with better detection capability is endowed with relatively higher weight, and the KNN classifier with poorer detection capability is endowed with lower weight. By the method, the IC welding spot sample is more reasonably evaluated and has more robustness. And (4) evaluating each sample in the whole training set once by using the trained KNN classifier, and counting the accuracy of evaluation. Calculating the weight of each KNN classifier according to formula (1):
wherein ,wiWeight of the ith KNN classifier, eiRepresenting the accuracy of the ith KNN classifier, emaxRepresenting the highest accuracy in KNN classifiers, eminRepresenting the lowest accuracy in the KNN classifier.
The weight of each KNN classifier is determined in a self-adaptive mode, the KNN classifier with good detection capability is endowed with relatively high weight, and the KNN classifier with poor detection capability is endowed with low weight, so that the adverse effect on the accuracy evaluation of the IC welding spot sample is reduced.
Global evaluation: after each sample in the training set is detected by using a plurality of KNN classifiers, the average failure probability (MUP) of each sample can be calculated according to the formula (2):
wherein ,MUPXRepresenting the mean probability of failure, KNN, of sample Xi(X) represents the detection result of the sample X by the ith KNN classifier, and N represents the number of KNN classifiers.
Evaluation threshold value: each sample in the training set is evaluated, and an Unqualified Probability Threshold (UPT) of the training set is calculated according to formula (3), namely the evaluation Threshold:
UPT＝αUmin+(1-α)Qmax；(3)
wherein α represents the proportion of qualified samples in the training set, and is calculated by formula (4),Uminrepresents the minimum value of MUP in the failed sample, QmaxRepresents the maximum value of MUP in the qualified sample, so that the boundary of the qualified sample and the unqualified sample can be comprehensively considered, and the evaluation threshold is more robust.
Q in equation (4) represents a qualified sample, U represents an unqualified sample, num ({ Q }) represents the number of qualified samples in the training set, and num ({ Q } + { U }) represents the number of samples in the training set. And adjusting the proportion of the qualified sample limit and the unqualified sample limit in the evaluation threshold according to the proportion alpha of the qualified samples in the training set, wherein the evaluation threshold is changed along with the composition proportion of the qualified samples and the unqualified samples in the training set, so that the evaluation result is more reasonable.
The IC welding spot defect detection method based on the learning mechanism provided by the embodiment of the invention is based on a mechanism from local statistical modeling to global evaluation detection, not only can well grasp the characteristic information in the training set samples, but also can reveal the relation among the training set samples to globally evaluate the quality of the IC welding spot samples, so that the method can complete the detection task of the IC welding spot samples under the conditions of insufficient and unbalanced training samples.
In the embodiment of the invention, the weight of the KNN classifier is set in a self-adaptive mode, the self-adaptive weight can improve the influence of the KNN classifier with good detection effect, and reduce the influence of the KNN classifier with poor detection effect on global evaluation, so that the robustness of the whole detection scheme can be improved, and the detection accuracy is also improved.
In the embodiment of the invention, the evaluation threshold is set in a self-adaptive mode, and because the unqualified sample is added into the training set, the evaluation threshold in the embodiment of the invention can simultaneously consider the limit of the qualified sample and the limit of the unqualified sample, so that the evaluation is more reasonable. Meanwhile, the proportion of the qualified sample limit and the unqualified sample limit in the evaluation threshold is adjusted according to the proportion of the qualified samples in the training set, and the evaluation threshold is changed along with the composition proportion of the qualified samples and the unqualified samples in the training set, so that the evaluation result is more reasonable.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An IC welding spot defect detection method based on a learning mechanism is characterized by comprising the following steps:
establishing a local statistical model according to a plurality of qualified IC welding spot samples;
collecting pictures of an IC welding spot sample, inputting the pictures into the local statistical model for comparison, and obtaining a potential defect image of the IC welding spot sample as a training set; the training set comprises qualified samples and unqualified samples, wherein the qualified samples are potential defect images of the qualified IC welding spot samples, and the unqualified samples are potential defect images of the unqualified IC welding spot samples;
respectively training a plurality of classifiers by using different training subsets; wherein the training subset consists of all unqualified samples and the same number of qualified samples in the training set;
respectively evaluating each sample in the training set by using the trained classifiers, determining the weight of each classifier, and calculating the average disqualification probability of each sample so as to obtain the evaluation threshold of the training set sample;
and acquiring an IC welding spot sample picture to be detected, obtaining the average unqualified probability of the IC welding spot sample by using the local statistical model and the plurality of classifiers, and comparing the average unqualified probability of the IC welding spot sample with the evaluation threshold value to obtain the detection result of the IC welding spot sample.
2. The learning mechanism-based IC solder joint defect inspection method of claim 1, wherein determining the average failure probability of the IC solder joint sample using the local statistical model and the plurality of classifiers further comprises: inputting the IC welding spot sample picture into the local statistical model to obtain a potential defect image of the IC welding spot sample, and evaluating the potential defect image by utilizing a plurality of classifiers to obtain the average failure probability of the IC welding spot sample.
3. The IC solder joint defect detection method based on the learning mechanism as claimed in claim 2, wherein comparing the average failure probability of the IC solder joint sample with the evaluation threshold to obtain the detection result of the IC solder joint sample further comprises: and judging whether the average disqualification probability of the IC welding spot sample is smaller than the evaluation threshold value, if so, determining that the IC welding spot sample is qualified, otherwise, determining that the IC welding spot sample is disqualified.
4. The learning mechanism-based IC solder joint defect inspection method of claim 1 or 3, wherein determining the weight of each classifier further comprises calculating the weight of each classifier using the following formula:
wherein ,wiWeight of the ith classifier, eiIndicating the accuracy of the ith classifier, emaxRepresenting the highest accuracy in the classifier, eminRepresenting the lowest accuracy in the classifier.
5. The IC solder joint defect detection method based on the learning mechanism as claimed in claim 4, wherein the classifier is a KNN classifier.
6. The learning mechanism-based IC solder joint defect inspection method of claim 5, wherein calculating the average failure probability for each sample further comprises calculating the average failure probability for each sample using the following equation:
wherein ,MUPXRepresenting the mean probability of failure, KNN, of sample Xi(X) represents the probability of the ith KNN classifier failing the sample X, and N represents the number of KNN classifiers.
7. The learning mechanism-based IC solder joint defect detection method of claim 6, wherein the evaluation threshold is calculated using the following formula:
UPT＝αUmin+(1-α)Qmax；
wherein α represents the proportion of qualified samples in the training set, UminRepresents the minimum value of MUP in the failed sample, QmaxRepresents the maximum value of MUP in the qualified samples, Q represents the qualified samples, U represents the unqualified samples, num ({ Q }) represents the number of qualified samples in the training set, and num ({ Q } + { U }) represents the number of samples in the training set.
8. The learning mechanism-based IC solder joint defect inspection method of claim 1, wherein building a local statistical model from a plurality of qualified IC solder joint samples further comprises: and establishing a local statistical model, initializing the local statistical model by utilizing a first qualified IC welding spot sample, and updating the local statistical model according to a subsequent qualified IC welding spot sample.
9. The IC solder joint defect detection method based on learning mechanism of claim 8, wherein the local statistical model is a VIBE model.
10. The method of claim 1, wherein the training subset consists of all failed samples and the same number of qualified samples in the training set, and further comprises: the qualified samples are randomly selected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010529781.9A CN111681235B (en) | 2020-06-11 | 2020-06-11 | IC welding spot defect detection method based on learning mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010529781.9A CN111681235B (en) | 2020-06-11 | 2020-06-11 | IC welding spot defect detection method based on learning mechanism |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111681235A true CN111681235A (en) | 2020-09-18 |
| CN111681235B CN111681235B (en) | 2023-05-09 |
Family
ID=72454550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010529781.9A Active CN111681235B (en) | 2020-06-11 | 2020-06-11 | IC welding spot defect detection method based on learning mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111681235B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113255778A (en) * | 2021-05-28 | 2021-08-13 | 广汽本田汽车有限公司 | Welding spot quality detection method and device based on multi-model fusion and storage medium |
| CN113313704A (en) * | 2021-06-18 | 2021-08-27 | 广东工业大学 | Integrated circuit pin detection method, device, terminal and storage medium |
| CN113378957A (en) * | 2021-06-23 | 2021-09-10 | 广东工业大学 | Adaptive statistical model training method, welding spot defect detection method and system |
| CN115017862A (en) * | 2022-07-01 | 2022-09-06 | 广东赛昉科技有限公司 | Integrated circuit image matching method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104867145A (en) * | 2015-05-15 | 2015-08-26 | 广东工业大学 | IC element solder joint defect detection method based on VIBE model |
| CN107123117A (en) * | 2017-04-26 | 2017-09-01 | 广东工业大学 | A kind of IC pin quality of welding spot detection method and device based on deep learning |
| CN109014544A (en) * | 2018-08-17 | 2018-12-18 | 龙岩学院 | Miniature resistance spot welding quality on-line monitoring method |
| CN109615609A (en) * | 2018-11-15 | 2019-04-12 | 北京航天自动控制研究所 | A kind of solder joint flaw detection method based on deep learning |
| CN111145175A (en) * | 2020-01-10 | 2020-05-12 | 惠州光弘科技股份有限公司 | SMT welding spot defect detection method based on iForest model verification |
-
2020
- 2020-06-11 CN CN202010529781.9A patent/CN111681235B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104867145A (en) * | 2015-05-15 | 2015-08-26 | 广东工业大学 | IC element solder joint defect detection method based on VIBE model |
| CN107123117A (en) * | 2017-04-26 | 2017-09-01 | 广东工业大学 | A kind of IC pin quality of welding spot detection method and device based on deep learning |
| CN109014544A (en) * | 2018-08-17 | 2018-12-18 | 龙岩学院 | Miniature resistance spot welding quality on-line monitoring method |
| CN109615609A (en) * | 2018-11-15 | 2019-04-12 | 北京航天自动控制研究所 | A kind of solder joint flaw detection method based on deep learning |
| CN111145175A (en) * | 2020-01-10 | 2020-05-12 | 惠州光弘科技股份有限公司 | SMT welding spot defect detection method based on iForest model verification |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113255778A (en) * | 2021-05-28 | 2021-08-13 | 广汽本田汽车有限公司 | Welding spot quality detection method and device based on multi-model fusion and storage medium |
| CN113313704A (en) * | 2021-06-18 | 2021-08-27 | 广东工业大学 | Integrated circuit pin detection method, device, terminal and storage medium |
| CN113378957A (en) * | 2021-06-23 | 2021-09-10 | 广东工业大学 | Adaptive statistical model training method, welding spot defect detection method and system |
| CN115017862A (en) * | 2022-07-01 | 2022-09-06 | 广东赛昉科技有限公司 | Integrated circuit image matching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111681235B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111681235A (en) | IC welding spot defect detection method based on learning mechanism | |
| CN111145175B (en) | SMT welding spot defect detection method based on iForest model verification | |
| CN111899241B (en) | Quantitative on-line detection method and system for defects of PCB (printed Circuit Board) patches in front of furnace | |
| CN104867144B (en) | IC element welding point defect detection methods based on mixed Gauss model | |
| CN108827969A (en) | Metal parts surface defects detection and recognition methods and device | |
| CN110880175B (en) | Welding spot defect detection method, system and equipment | |
| CN109918523B (en) | Circuit board component detection method based on YOLO9000 algorithm | |
| CN114926463A (en) | Production quality detection method suitable for chip circuit board | |
| CN110415214A (en) | Appearance detecting method, device, electronic equipment and the storage medium of camera module | |
| CN113628179B (en) | PCB surface defect real-time detection method, device and readable medium | |
| TWI765442B (en) | Method for defect level determination and computer readable storage medium thereof | |
| CN113971654A (en) | Welding spot detection model training method, welding spot detection method and welding spot detection device | |
| CN115439411A (en) | Method and device for detecting polarity of circuit board component, medium and electronic equipment | |
| CN116228741A (en) | PCBA (printed circuit board assembly) component defect detection method and device | |
| CN115294114A (en) | Quality detection method based on ECU circuit welding | |
| CN113674260A (en) | SMT welding spot defect detection method | |
| CN114387230A (en) | PCB defect detection method based on re-verification detection | |
| CN113176275A (en) | Method, device and system for retesting display panel | |
| CN113313704B (en) | Integrated circuit pin detection method, device, terminal and storage medium | |
| JP4814116B2 (en) | Mounting board appearance inspection method | |
| TW202113599A (en) | System for generating detection model according to standard data to confirm soldering state and method thereof | |
| CN114092448B (en) | Plug-in electrolytic capacitor mixed detection method based on deep learning | |
| CN115170476A (en) | Printed circuit board defect detection method based on image processing | |
| TWI758134B (en) | System for using image features corresponding to component identification for secondary inspection and method thereof | |
| Lei et al. | A precise convolutional neural network-based classification and pose prediction method for PCB component quality control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |