CN113095438B - Wafer defect classification method, device and system thereof, electronic equipment and storage medium - Google Patents
Wafer defect classification method, device and system thereof, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113095438B CN113095438B CN202110478016.3A CN202110478016A CN113095438B CN 113095438 B CN113095438 B CN 113095438B CN 202110478016 A CN202110478016 A CN 202110478016A CN 113095438 B CN113095438 B CN 113095438B
- Authority
- CN
- China
- Prior art keywords
- defect
- sample
- classification
- wafer
- detected
- 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.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 298
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003860 storage Methods 0.000 title claims description 11
- 238000001914 filtration Methods 0.000 claims description 24
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000004891 communication Methods 0.000 claims description 11
- 238000013145 classification model Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 8
- 238000004519 manufacturing process Methods 0.000 abstract description 8
- 235000012431 wafers Nutrition 0.000 description 110
- 230000008569 process Effects 0.000 description 12
- 238000013527 convolutional neural network Methods 0.000 description 10
- 230000000875 corresponding effect Effects 0.000 description 8
- 238000012549 training Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 3
- 230000002146 bilateral effect Effects 0.000 description 2
- 239000000470 constituent Substances 0.000 description 2
- 238000010894 electron beam technology Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011982 device technology Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- 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
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30148—Semiconductor; IC; Wafer
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a wafer defect classification method, which comprises the following steps: obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values; acquiring a sample image of the wafer sample to be detected, which is acquired in advance; and carrying out defect classification according to the sample image and the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected. The classification method reduces the dependence on manual operation, ensures the accuracy and the reliability of classification, reduces the production cost to a certain extent, and can be applied to mass product detection.
Description
Technical Field
The present invention relates to the field of semiconductor defect classification, and in particular, to a wafer defect classification method, and an apparatus, a system, an electronic device, and a computer readable storage medium thereof.
Background
With the development of semiconductor device technology, there are more and more processes for manufacturing semiconductor devices, and each process has a certain complexity, and each process may generate some undesirable structures for processing a wafer, where the wafer defects may cause on-chip circuits to fail to operate properly. In the chip manufacturing process, the wafer defect detection step is generally arranged after a plurality of key processes, so as to monitor the key processes and ensure the accuracy of the key processes.
Because of the extremely complex process flow of chip manufacturing, wafer defect types are numerous, and no unified classification mode exists at present. Generally, an engineer classifies wafer defects according to practical situations, such as a process performed before wafer inspection and an inspection method.
Currently, the most commonly used detection methods include two types: the first is manual detection, namely firstly scanning a wafer to be detected through a defect scanning device to obtain a position of the wafer to be detected, which is possibly provided with a defect, marking the position correspondingly, then automatically photographing through equipment such as an electron microscope (SEM), an optical microscope or an electron beam microscope according to the position marked by the defect scanning device to obtain one or more images of the defect of the wafer, and then classifying the defect by a worker based on the one or more images of the defect of the wafer; the second is classification based on various wafer defect classification models, such as CNN classifier, etc., that is, wafer defect images captured by SEM equipment, etc., are input into a classifier trained in advance, and then automatically classified by the classifier. However, each of the above two detection methods has a certain problem:
1) The worker detection method can only aim at the situation that wafer defects to be detected are few in types of defects on each wafer; in addition, the accuracy or reliability of classifying the defects by the staff is positively correlated with the experience of identifying the defects by the staff, namely, the more the experience is, the more the accuracy and reliability are correspondingly increased, namely, the classifying method has high professional requirements on the staff. On the other hand, if the worker performs defect recognition for a long time, accumulated fatigue may be caused, thereby reducing accuracy and reliability thereof, and the method is not widely applicable to real-time monitoring and classification of wafer defects in the wafer production process because of manual classification.
2) For the case of a large number of wafer defects, the defects are classified by adopting a Convolutional Neural Network (CNN) algorithm, wherein the conventional CNN algorithm is to extract defect characteristic parameters based on a defect region to classify the defects, such as texture characteristics, gray characteristics, morphological characteristics and the like to form characteristic vectors, and then input the characteristic vectors into a classifier to be processed to obtain classification results. Typical classification methods mainly comprise supervised classification and unsupervised classification, and compared with common unsupervised classification methods, an ISODATA method and a t-average method are adopted; the supervised classification method comprises a minimum distance method, a mahalanobis distance method and a maximum likelihood method. Since defect classification is directly performed according to images obtained by automatic photographing through a Scanning Electron Microscope (SEM), an optical microscope, or an electron beam microscope, such algorithms are mainly limited to whether defect feature parameters directly extracted from photographed images can effectively express differences of different defect types, good defect feature expression is important, but defect feature parameters extracted for specific defect types are difficult to be widely applied to classification of other defect types, which makes defect detection and classification more difficult.
In view of the above, the present invention provides a new wafer defect classification method and apparatus thereof.
Disclosure of Invention
In view of the above technical problems, an object of the present invention is to provide a wafer defect classification method, a device, a system, an electronic device and a storage medium thereof, which overcome or alleviate the above defects in the prior art to a certain extent.
In a first aspect of the present invention, a method for classifying wafer defects is provided, including the steps of:
obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values;
acquiring a sample image of the wafer sample to be detected, which is acquired in advance;
and carrying out defect classification according to the sample image and the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected.
Further, in some exemplary embodiments of the present invention, the step of classifying the defects according to the sample image and the defect feature parameter specifically includes:
performing first classification on the sample images according to the defect characteristic parameters to obtain at least one group of sample images to be classified;
and performing secondary classification on the at least one group of sample images to be classified according to a pre-trained classification model to obtain the type of the defect.
Further, in some exemplary embodiments of the present invention, before the step of classifying the defects, in order to ensure accuracy of classification, the method further includes the steps of: and preprocessing the sample image.
Further, in some exemplary embodiments of the invention, the preprocessing includes: and (5) filtering.
A second aspect of the present invention provides a defect classification apparatus, comprising:
the first data acquisition module is used for acquiring defect characteristic parameters of each defect in the wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values;
the second data acquisition module is used for acquiring a sample image of the wafer sample to be detected, which is acquired in advance; and the defect classification module is used for performing defect classification according to the sample image and the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected.
Further, in some exemplary embodiments of the present invention, the defect classification module specifically includes:
the first classification unit is used for classifying the sample images for the first time according to the defect characteristic parameters to obtain at least one group of sample images to be classified;
and the second classification unit is used for secondarily classifying the at least one group of sample images to be classified according to a pre-trained classification model to obtain the type of the defect.
Further, in some exemplary embodiments of the present invention, the apparatus further comprises: and the image preprocessing module is used for preprocessing the sample image.
Further, in some exemplary embodiments of the present invention, the preprocessing module specifically includes: and the filtering unit is used for carrying out filtering processing on the sample image.
A third aspect of the present invention provides a defect classification system, comprising:
the scanning equipment is used for scanning the wafer sample to be detected to obtain defect characteristic parameters of each defect in the wafer sample to be detected; the defect characteristic parameters include: defect size and signal strength values;
the photographing device is used for photographing the wafer sample to be detected so as to obtain a sample image of the wafer sample to be detected;
and any one of the defect classification devices is configured to classify defects according to the sample image and the defect characteristic parameter, so as to obtain a type of each defect in the wafer sample to be detected.
A fourth aspect of the present invention is to provide an electronic device usable for defect classification, the device comprising at least one processor, at least one memory, a communication interface and a bus; the processor, the memory and the communication interface complete communication with each other through the bus; the memory is used for storing a program for executing any one of the methods; the processor is configured to execute a program stored in the memory.
A fifth aspect of the present invention provides a computer readable storage medium storing a computer program for defect classification, which when executed by a processor implements the steps of any of the methods described above.
Advantageous effects
The defect classification method based on the defect characteristic parameters comprises the following steps: obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values; acquiring a sample image of the wafer sample to be detected, which is acquired in advance; and then, carrying out defect classification according to the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected. Compared with the traditional classification method, the defect classification is performed based on the defect characteristic parameters, so that the dependence on manpower can be reduced, on one hand, the defect classification efficiency is improved, and meanwhile, the industrial production detection cost of the wafer is reduced; on the other hand, the accuracy and the reliability of defect classification are improved; the technical scheme provided by the invention has higher efficiency and classification result precision, so that the method can be used for monitoring and detecting the wafer defects on the production line in real time.
Further, the wafer defects are classified for the first time based on defect attributes (namely defect characteristic parameters such as defect size and signal intensity value), and then the sample images to be classified are further classified accurately by utilizing a pre-trained classification model; the method solves the problem that the defect characteristic parameters extracted for specific defect types are difficult to be widely applied to classification of other defect types by first classification to a certain extent (in other words, the classification of the parameters of the defect characteristics obtained by wafer scanning enables the classification of the wafer to be more standardized, the application scene of classification standards to be wider and more universal, avoids that manual classification standards are not easy to unify and are easily influenced by manual subjective judgment); and reduces the dependence on manpower, thereby improving the classification efficiency and reducing the production cost.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale. It will be apparent to those of ordinary skill in the art that the drawings in the following description are of some embodiments of the invention and that other drawings may be derived from these drawings without inventive faculty.
FIG. 1 is a flow chart of a wafer defect classification method according to an exemplary embodiment of the invention;
FIG. 2 is a schematic diagram of partial classification of wafers according to an exemplary embodiment of the invention;
FIG. 3 is a schematic diagram of a wafer defect classification apparatus according to an example embodiment of the invention;
fig. 4 is a block diagram showing the constituent structure of an electronic device according to an exemplary embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In this document, suffixes such as "module", "component", or "unit" used to represent elements are used only for facilitating the description of the present invention, and have no particular meaning in themselves. Thus, "module," "component," or "unit" may be used in combination.
Example 1
Referring to fig. 1, a flow chart of a wafer defect classification method according to an exemplary embodiment of the invention, specifically, the wafer defect classification method according to the exemplary embodiment includes the following steps:
s101, obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which is obtained by scanning the wafer sample to be detected in advance.
In some embodiments, the defect characterization parameters include: defect size and signal strength values.
In some embodiments, a wafer defect is first identified by a defect scanning device, specifically, a wafer sample to be detected is divided into small pixels by the defect scanning device, then gray values or brightness values of pixels at the same position of different wafers are compared, some pixels with larger differences are identified as the wafer defect, then the identified wafer defect is marked, and defect characteristic parameters of the wafer defect are recorded by the defect scanning device.
Of course, in other embodiments, some standard wafer samples (i.e., standard wafer samples without any defect) may be selected, then the standard wafer samples are scanned to obtain a gray value or a brightness value of each pixel point in the standard wafer samples, the gray value or the brightness value of each pixel point in the standard wafer samples is referred to (for example, according to an average value of the gray values or the brightness values of the pixel points at the same position in the selected standard wafer samples), a gray value threshold range or a brightness value threshold range of each pixel point in the standard case (i.e., in the absence of a defect) is set, then the gray value threshold range or the brightness value threshold range is set as a standard for identifying the defect by the defect scanning device, for example, when the defect is scanned, the pixel points exceeding the gray value threshold range or the brightness value threshold range are identified as the wafer defect, then the identified wafer defect is marked, and the defect characteristic parameters of the wafer defect are recorded by the defect scanning device.
In some embodiments, the defect characteristic parameters of each defect are obtained by scanning the respective wafers through various defect scanning devices, such as a laser scanning device, an infrared scanning device, an ultrasonic scanning device, and the like, and preferably, the defect characteristic parameters of a plurality of wafer samples to be detected can be obtained directly from the defect scanning devices by using the laser scanning deviceThe line communication or wireless communication mode obtains a plurality of defect characteristic parameters from the defect scanning equipment, and specifically, the defect characteristic parameters comprise: the size of the defect and the signal strength value, and the identity of the wafer to be tested where the defect is located (e.g., wafer number Ip to be tested, wherein p=1, 2, ··, obtaining defect characteristic parameters of each defect on each wafer to be tested: defect parameters for N1 (n1=1, 2,3·) defects on wafer sample I1 to be tested: defect (size) size: a is that I1，1 ，A I1，2 ，A I1，3 ，…A I1，N1 The method comprises the steps of carrying out a first treatment on the surface of the Signal intensity value: s is S I1，1 ，S I1，2 ，S I1，3 ···S I1，N1 The method comprises the steps of carrying out a first treatment on the surface of the Defect parameters for N2 (n1=1, 2,3·) defects on wafer sample I2 to be tested: defect size: a is that I2，1 ，A I2，2 ，A I2，3 ，…A I2，N2 The method comprises the steps of carrying out a first treatment on the surface of the Signal intensity value: s is S I2，1 ，S I2，2 ，S I2，3 ···S I2，N2 ···。
S103, acquiring a sample image of a wafer sample to be detected, which is acquired in advance.
In some embodiments, the sample image is obtained in advance by various defect photographing devices, in particular, scanning each marked defect with a scanning electron microscope to obtain a sample image, and accordingly, a plurality of sample images may be obtained directly from these defect photographing devices. For example, a plurality of sample images can be acquired from a scanning electron microscope through wired communication or wireless communication, namely, the acquired sample image set comprises: b1, B2, B3, B4 … BN.
S105, performing defect classification according to the defect characteristic parameters acquired in the step S101 and the sample images acquired in the step S103 to obtain the type of each defect in the wafer sample to be detected.
In some embodiments, the step S105 specifically includes the steps of:
firstly, classifying sample images for the first time according to the acquired defect characteristic parameters to obtain at least one group of sample images to be classified.
In some embodiments, the wafer defects may be first classified by defect size (i.e., the size of the defect), the wafer defects may be classified into one or more types by defect size, for example, classifying the identified wafer defects by the size characteristic values of existing defects in the defect scanning apparatus, into large defects and small defects, the classification result is shown in fig. 2, where fig. a is the small defect and fig. b is the large defect.
Of course, in other embodiments, the wafer defects may be classified for the first time by signal intensity values that are related to the imaging principles of the defect scanning device used, and the signal intensity values collected by the defect scanning device are reflected light intensities, and some are scattered light intensities. Specifically, a classification standard using the signal strength value as a main parameter may be pre-constructed, the signal strength value is divided into different numerical intervals, and the professional performs defect type identification (such as defect class name or codes for identifying various defect classes) on the different numerical intervals, for example, the strength signal value may be divided into different numerical intervals: c1, C2, C3 … CX, C1 being labeled as a first defect class identifier, i.e. corresponding to a first wafer defect; c2 is labeled as a second defect class identifier, i.e., corresponding to a second wafer defect; c3 is labeled as a third defect class identifier, i.e., corresponding to a third wafer defect; … are collectively expected to be identified by X wafer defects, and each wafer defect corresponds to a defect class. Then, the signal intensity value of the defect in the sample image is matched with the numerical interval of the classification standard by using defect classification equipment, so that the identification and classification of the defect are carried out on the sample image. Of course, the signal intensity values herein may be replaced by gray or luminance values.
In other embodiments, a plurality of preset size threshold ranges are preset according to the defect sizes (i.e., the sizes of the defects), each preset size threshold range corresponds to at least one preset signal strength threshold range, then the obtained defect sizes of the defects on each wafer to be tested are compared with the preset size threshold ranges, and the signal strength values of the defects are compared with the preset signal strength threshold ranges, and if the defect sizes are determined to belong to the corresponding preset size threshold ranges and the signal strength values are determined to belong to the corresponding preset size threshold ranges, the defects are classified into a group of defects to be classified.
For example, a first type of defect corresponds to a preset defect size threshold range of X1-X2 and a preset signal strength threshold range of Y1-Y2; the second type of defects corresponds to a preset defect size threshold range of X3-X4, and a preset signal intensity threshold range of Y3-Y4; then, comparing the defect characteristic parameters of N1 defects on the wafer I1 to be tested obtained in the step S101 with the preset defect size threshold range and the preset signal strength threshold range, if A is determined I1，1 Belonging to X3-X4 and S I1，1 Belongs to Y3-Y4; and judge A I5，7 Belonging to X3-X4 and S I5，7 Belonging to Y3-Y4, dividing the defect on the wafer I1 to be detected and the defect on the wafer I5 to be detected into the same group of defects to be classified; similarly, a plurality of groups of defects to be classified are obtained: m1, M2, M3.
And secondly, performing secondary classification on at least one group of sample images to be classified obtained in the step S1051 according to a pre-trained classification model to obtain the types of all defects.
In some embodiments, the classification model may employ common classifiers such as CNN, SVM, etc., and the selected classifier is trained in advance, specifically by pre-marking sample images with corresponding defect types as training samples.
In a specific embodiment, the convolutional neural network (Convolutional Neural Networks, CNN) method is used to perform the secondary classification, and the convolutional neural network algorithm can be: alexNet networks, ZFNet networks based on AlexNet improvements, VGG networks, ***net networks, etc., for example, in this embodiment ZFNet may be employed for defect classification. Before secondary classification, a scanning electron microscope is used for collecting sample images of a plurality of wafer materials, and corresponding defect types are marked in advance to obtain a training set. The defect types can be specifically classified into three types of redundant defects, crystal defects, mechanical damages and the like according to morphological characteristics of the defect types in combination with actual conditions (of course, the defect types can also be classified into types of defects such as offset, fragmentation, unfilled corner, broken edge, incomplete, protrusion and the like, and the defect types can also be classified into types of point defects, dislocation, original defects, impurities and the like, and the classification method of the defect types is not limited to the classification in actual production work, and can be flexibly classified according to actual application scenes), further, the positions and defect types of sample defects of a training set are marked, and meanwhile, in order to improve the generation efficiency of the training set, the data set can be expanded by performing methods such as horizontal, vertical overturn, random contrast change and the like on sample images. Then, training the patch-based ZFNet detector, wherein when the patch-based ZFNet classifier is trained, the data set is a sample image of the wafer materials, the sample image contains defect positions and types, a series of data expansion operations are performed to obtain a plurality of groups of data, and 60% of the data are randomly selected as a training set and 40% of the data are used as a test set. On the basis of the trained patch-based ZFNet detector, sample images of wafers to be detected are sequentially sent to the ZFNet detector, and the ZFNet detector classifies defects according to set defect types.
In some embodiments, a sample image to be classified may be randomly selected from each group, and then classified by the classifier, so as to obtain a defect type of the sample image to be classified, and all sample images to be classified in the group of images to be classified in which the sample image to be classified is currently located are marked as the defect type. Of course, further, in order to improve the accuracy of classification, at least two images to be classified may be selected from each group for secondary classification.
In other embodiments, a sample image to be classified having the largest defect (size) and/or the largest signal strength value may be selected from each group for secondary classification. Further, in order to improve the accuracy of classification, all the sample images to be classified in each group can be sorted according to the order of the sizes of the defects (sizes), and then the first two bits or the first three bits of sample images to be classified of the defect (size) size are selected from each group for secondary classification; or, sorting all the sample images to be classified in each group according to the sequence of the signal intensity values from large to small, and then selecting the first two or the first three sample images to be classified in the signal intensity value sorting from each group for secondary classification; alternatively, at least one sample image to be classified having the defect (size) and signal intensity values ordered in the first few bits (e.g., 3 bits) is selected from each group for secondary classification.
Further, in some embodiments, to remove noise interference in the sample image, to improve accuracy of defect classification, the sample image is further preprocessed before performing the step of defect classification.
In some embodiments, the preprocessing is filtering; specifically, the filtering process includes: median filtering, mean filtering, square filtering, gaussian filtering and bilateral filtering. In this embodiment, median filtering may be used to perform filtering processing on a sample image, where pixel points in the sample image are first divided into isolated noise points, edge details and flat noise, where the isolated noise points and flat noise are processed by median filtering, the edge details are directly output without processing, and then a sample image with available accurate classification is obtained.
Example two
Referring to fig. 3, a schematic diagram of a wafer defect classification apparatus according to an exemplary embodiment of the invention is shown. Specifically, the wafer defect classification apparatus of the present exemplary embodiment includes:
a first data obtaining module 14, configured to obtain a defect characteristic parameter of each defect in the wafer sample to be detected, where the defect characteristic parameter is obtained by scanning the wafer sample to be detected in advance; wherein the defect characteristic parameters include: defect size and signal strength values; specifically, the first data acquisition module 14 may be a laser scanning device, an infrared scanning device, an ultrasonic scanning device;
a second data acquisition module 16, configured to acquire a sample image of the wafer sample to be detected acquired in advance; specifically, the second data acquisition module 16 employs a scanning electron microscope;
and the defect classification module 18 is configured to classify defects according to the sample image and the defect characteristic parameters, so as to obtain the type of each defect in the wafer sample to be detected.
In some embodiments, the defect classification module 18 specifically includes: the first classification unit is used for classifying the acquired sample images for the first time according to the defect characteristic parameters to obtain at least one group of sample images to be classified; and the second classification unit is used for secondarily classifying the at least one group of sample images to be classified according to a pre-trained classification model to obtain the type of each defect.
In some embodiments, the second classifying unit may classify the wafer defect using a CNN algorithm.
In some embodiments, the apparatus further comprises an image preprocessing module, configured to preprocess the sample image. Specifically, the preprocessing module specifically includes:
a filtering unit for filtering the sample image subjected to the background processing; specifically, the filtering unit may perform denoising processing on the sample image by using any one or more of median filtering, mean filtering, block filtering, gaussian filtering, bilateral filtering, and the like.
Accordingly, based on the above-mentioned wafer defect classification device, a wafer defect classification system is provided, and the system includes:
the scanning equipment is used for scanning the wafer sample to be detected to obtain a sample image of the wafer sample to be detected and defect characteristic parameters of each defect in the wafer sample to be detected; wherein the defect characteristic parameters include: defect size and signal strength values;
the photographing device is used for photographing the wafer sample to be detected so as to obtain a sample image of the wafer sample to be detected;
and the defect classification device in the second embodiment is configured to classify defects according to the sample image and the defect feature parameters, so as to obtain a type of each defect in the wafer sample to be detected.
Example III
In a third aspect of the invention, an electronic device is provided, comprising a memory 502, a processor 501 and a computer program stored on the memory 502 and executable on the processor 501, said processor 501 implementing the steps of the method described hereinbefore when executing said program. For convenience of description, only those parts related to the embodiments of the present specification are shown, and specific technical details are not disclosed, please refer to the method parts of the embodiments of the present specification. The electronic device may be any electronic device including various electronic devices, such as a PC computer, a network cloud server, a mobile phone, a tablet PC, a PDA (Personal Digital Assistant ), a POS (Point of Sales), a vehicle-mounted computer, and a desktop computer.
Specifically, fig. 4 is a block diagram of the constituent structure of an electronic device according to an exemplary embodiment of the present invention. Bus 500 may include any number of interconnected buses 500 and bridges linking together various circuits, including one or more processors, represented by processor 501, and memory, represented by memory. Bus 500 may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., as are well known in the art and, therefore, will not be described further herein. The communication interface 503 provides an interface between the bus and a receiver and/or transmitter 504, which receiver and/or transmitter 504 may be separate and independent receivers or transmitters 504 may be the same element, such as a transceiver, providing a means for communicating with various other apparatus over a transmission medium. The processor 501 is responsible for managing the bus 500 and general processing, while the memory 502 may be used to store data used by the processor 501 in performing operations.
From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a computer readable storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, or a network device, etc.) to perform the above-described method according to the embodiments of the present disclosure.
The computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable storage medium may also be any readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).
The computer-readable medium carries one or more programs, which when executed by one of the devices, cause the computer-readable medium to perform the functions of: obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which is obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values; acquiring a sample image of the wafer sample of the chip to be detected, which is acquired in advance; and carrying out defect classification according to the sample image and the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected.
Those skilled in the art will appreciate that the modules may be distributed throughout several devices as described in the embodiments, and that corresponding variations may be implemented in one or more devices that are unique to the embodiments. The modules of the above embodiments may be combined into one module, or may be further split into a plurality of sub-modules.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising several instructions for causing a computer terminal (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are to be protected by the present invention.
Claims (9)
1. The defect classification method based on the defect characteristic parameters is characterized by comprising the following steps:
obtaining defect characteristic parameters of each defect in a wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values;
acquiring a sample image of the wafer sample to be detected, which is acquired in advance;
performing defect classification according to the sample image and the defect characteristic parameters to obtain the type of each defect in the wafer sample to be detected, wherein the step of performing defect classification according to the sample image and the defect characteristic parameters specifically comprises the following steps:
performing first classification on the sample images according to the defect characteristic parameters to obtain at least one group of sample images to be classified;
and performing secondary classification on the at least one group of sample images to be classified according to a pre-trained classification model to obtain the type of the defect.
2. The method of claim 1, further comprising, prior to the step of classifying the defect, the step of:
and preprocessing the sample image.
3. The method of claim 2, wherein the preprocessing comprises: and (5) filtering.
4. A defect classification apparatus, comprising:
the first data acquisition module is used for acquiring defect characteristic parameters of each defect in the wafer sample to be detected, which are obtained by scanning the wafer sample to be detected in advance; the defect characteristic parameters include: defect size and signal strength values;
the second data acquisition module is used for acquiring a sample image of the wafer sample to be detected, which is acquired in advance; the defect classification module is configured to perform defect classification according to the sample image and the defect feature parameter, and obtain a type of each defect in the wafer sample to be detected, where the defect classification module specifically includes:
the first classification unit is used for classifying the sample images for the first time according to the defect characteristic parameters to obtain at least one group of sample images to be classified;
and the second classification unit is used for secondarily classifying the at least one group of sample images to be classified according to a pre-trained classification model to obtain the type of the defect.
5. The apparatus as recited in claim 4, further comprising: and the image preprocessing module is used for preprocessing the sample image.
6. The device according to claim 5, wherein the preprocessing module specifically comprises:
and the filtering unit is used for carrying out filtering processing on the sample image.
7. A defect classification system, comprising:
the scanning equipment is used for scanning the wafer sample to be detected to obtain defect characteristic parameters of each defect in the wafer sample to be detected; the defect characteristic parameters include: defect size and signal strength values;
the photographing device is used for photographing the wafer sample to be detected so as to obtain a sample image of the wafer sample to be detected; and
the defect classification apparatus according to any one of claims 4 to 6, configured to perform defect classification according to the sample image and the defect feature parameter, so as to obtain a type of each defect in the wafer sample to be detected.
8. An electronic device comprising at least one processor, at least one memory, a communication interface, and a bus; wherein,
the processor, the memory and the communication interface complete the communication with each other through the bus;
the memory is used for storing a program for executing the method of any one of claims 1 to 3;
the processor is configured to execute a program stored in the memory.
9. A computer-readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method of any one of claims 1 to 3.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110478016.3A CN113095438B (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
| CN202410073784.4A CN117893817A (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110478016.3A CN113095438B (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410073784.4A Division CN117893817A (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113095438A CN113095438A (en) | 2021-07-09 |
| CN113095438B true CN113095438B (en) | 2024-03-15 |
Family
ID=76681461
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410073784.4A Pending CN117893817A (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
| CN202110478016.3A Active CN113095438B (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410073784.4A Pending CN117893817A (en) | 2021-04-30 | 2021-04-30 | Wafer defect classification method, device and system thereof, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN117893817A (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113222967A (en) * | 2021-05-28 | 2021-08-06 | 长江存储科技有限责任公司 | Wafer detection method and system |
| CN113610788B (en) * | 2021-07-27 | 2023-03-07 | 上海众壹云计算科技有限公司 | Fault monitoring method and device for image acquisition device, electronic equipment and storage medium |
| CN114119472A (en) * | 2021-10-21 | 2022-03-01 | 东方晶源微电子科技（北京）有限公司 | Defect classification method and device, equipment and storage medium |
| JP2023076992A (en) * | 2021-11-24 | 2023-06-05 | 東レエンジニアリング株式会社 | Automatic defect classification device |
| CN115360116B (en) * | 2022-10-21 | 2023-01-31 | 合肥晶合集成电路股份有限公司 | Wafer defect detection method and system |
| CN115753813B (en) * | 2022-11-01 | 2023-10-31 | 太原国科半导体光电研究院有限公司 | Method, device and system for detecting wafer defects, storage medium and electronic equipment |
| CN115661136B (en) * | 2022-12-12 | 2023-03-07 | 深圳宝铭微电子有限公司 | Semiconductor defect detection method for silicon carbide material |
| CN115876823B (en) * | 2023-01-19 | 2023-07-14 | 合肥晶合集成电路股份有限公司 | Film defect detection method, film defect detection device and detection system |
| CN116130377B (en) * | 2023-04-17 | 2023-09-29 | 西安奕斯伟材料科技股份有限公司 | Method, device and system for detecting defects of epitaxial wafer and manufacturing method thereof |
| CN117197617B (en) * | 2023-09-19 | 2024-05-28 | 芯率智能科技(苏州)有限公司 | Defect classification method and system for repeated defects |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090053274A (en) * | 2007-11-23 | 2009-05-27 | 삼성전자주식회사 | Reviewing apparatus of wafer defect and method thereof |
| CN101996855A (en) * | 2009-08-21 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Wafer defect analysis method |
| CN104062305A (en) * | 2014-07-28 | 2014-09-24 | 上海华力微电子有限公司 | Defect analysis method for integrated circuit |
| CN108133900A (en) * | 2017-12-21 | 2018-06-08 | 上海华力微电子有限公司 | A kind of Defect Scanning board and its automatic defect classification method |
| CN109977808A (en) * | 2019-03-11 | 2019-07-05 | 北京工业大学 | A kind of wafer surface defects mode detection and analysis method |
| WO2020057644A1 (en) * | 2018-09-21 | 2020-03-26 | Changxin Memory Technologies, Inc. | Method and apparatus for classification of wafer defect patterns as well as storage medium and electronic device |
| CN110969598A (en) * | 2018-09-28 | 2020-04-07 | 台湾积体电路制造股份有限公司 | Wafer inspection method and wafer inspection system |
| CN111819676A (en) * | 2018-03-28 | 2020-10-23 | 科磊股份有限公司 | Training neural networks for defect detection in low resolution images |
| CN112129772A (en) * | 2019-06-24 | 2020-12-25 | 杭州元色科技有限公司 | Defect detection system and method |
| CN112529873A (en) * | 2020-12-09 | 2021-03-19 | 深圳市芯汇群微电子技术有限公司 | Wafer defect detection method based on ART neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW535207B (en) * | 2002-05-08 | 2003-06-01 | Powerchip Semiconductor Corp | Method for automatically controlling defect-specification in manufacturing process of semiconductors |
-
2021
- 2021-04-30 CN CN202410073784.4A patent/CN117893817A/en active Pending
- 2021-04-30 CN CN202110478016.3A patent/CN113095438B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090053274A (en) * | 2007-11-23 | 2009-05-27 | 삼성전자주식회사 | Reviewing apparatus of wafer defect and method thereof |
| CN101996855A (en) * | 2009-08-21 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Wafer defect analysis method |
| CN104062305A (en) * | 2014-07-28 | 2014-09-24 | 上海华力微电子有限公司 | Defect analysis method for integrated circuit |
| CN108133900A (en) * | 2017-12-21 | 2018-06-08 | 上海华力微电子有限公司 | A kind of Defect Scanning board and its automatic defect classification method |
| CN111819676A (en) * | 2018-03-28 | 2020-10-23 | 科磊股份有限公司 | Training neural networks for defect detection in low resolution images |
| WO2020057644A1 (en) * | 2018-09-21 | 2020-03-26 | Changxin Memory Technologies, Inc. | Method and apparatus for classification of wafer defect patterns as well as storage medium and electronic device |
| CN110969598A (en) * | 2018-09-28 | 2020-04-07 | 台湾积体电路制造股份有限公司 | Wafer inspection method and wafer inspection system |
| CN109977808A (en) * | 2019-03-11 | 2019-07-05 | 北京工业大学 | A kind of wafer surface defects mode detection and analysis method |
| CN112129772A (en) * | 2019-06-24 | 2020-12-25 | 杭州元色科技有限公司 | Defect detection system and method |
| CN112529873A (en) * | 2020-12-09 | 2021-03-19 | 深圳市芯汇群微电子技术有限公司 | Wafer defect detection method based on ART neural network |
Non-Patent Citations (4)
| Title |
|---|
| IC晶圆表面缺陷检测技术研究;马磊;《中国优秀硕士学位论文全文数据库信息科技辑》(第03期);I138-7398 * |
| Wafer Defect Detection and Classification Algorithms Based on Convolutional Neural Network;Fang Xin 等;《Computer Engineering》;第44卷(第08期);218-423 * |
| 基于光散射理论的玻璃晶圆表面缺陷检测方法研究;涂政乾 等;《光散射学报》;第32卷(第03期);245-250 * |
| 基于机器视觉的晶圆缺陷检测***分析与设计;陈治杉;《中国优秀硕士学位论文全文数据库信息科技辑》(第09期);I135-73 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113095438A (en) | 2021-07-09 |
| CN117893817A (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113095438B (en) | Wafer defect classification method, device and system thereof, electronic equipment and storage medium | |
| CN110148130B (en) | Method and device for detecting part defects | |
| US9098891B2 (en) | Adaptive sampling for semiconductor inspection recipe creation, defect review, and metrology | |
| CN109598698B (en) | System, method, and non-transitory computer readable medium for classifying a plurality of items | |
| TWI833854B (en) | Optical-mode selection for multi-mode semiconductor inspection | |
| CN113096119B (en) | Method and device for classifying wafer defects, electronic equipment and storage medium | |
| US11686689B2 (en) | Automatic optimization of an examination recipe | |
| TW202044067A (en) | Machine learning-based classification of defects in a semiconductor specimen | |
| CN112805719A (en) | Classifying defects in semiconductor samples | |
| Tang et al. | Anomaly detection of core failures in die casting X-ray inspection images using a convolutional autoencoder | |
| CN111179261A (en) | Defect detection method, system, terminal device and storage medium | |
| KR102174424B1 (en) | Method for Inspecting Component basesd Server and system and apparatus therefor | |
| CN116978834A (en) | Intelligent monitoring and early warning system for wafer production | |
| KR20190098271A (en) | Diagnostic Methods for Defects and Classifiers Captured by Optical Tools | |
| KR20210067606A (en) | Apparatus and method for defect classification using feature generation based on machine-learning | |
| JP7146857B2 (en) | Method and system for classifying wafer defects using wafer defect images based on deep learning | |
| JP7404817B2 (en) | Learning device, detection device, learning method, and learning program | |
| KR102230559B1 (en) | Method and Apparatus for Creating Labeling Model with Data Programming | |
| US11151706B2 (en) | Method of classifying defects in a semiconductor specimen and system thereof | |
| KR102355997B1 (en) | method for monitoring the soundness of concrete structure based on smart glass | |
| CN113610155B (en) | Wafer defect classification method and device based on similarity comparison model, electronic equipment and storage medium | |
| CN113837173A (en) | Target object detection method and device, computer equipment and storage medium | |
| CN111724352B (en) | Patch LED flaw labeling method based on kernel density estimation | |
| US12020417B2 (en) | Method and system for classifying defects in wafer using wafer-defect images, based on deep learning | |
| CN118229658A (en) | Method and device for determining fault photovoltaic module, storage medium and electronic device |
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 |