CN116415660A - Construction and retrieval method, device and medium based on wafer defect knowledge base - Google Patents
Construction and retrieval method, device and medium based on wafer defect knowledge base Download PDFInfo
- Publication number
- CN116415660A CN116415660A CN202310684277.XA CN202310684277A CN116415660A CN 116415660 A CN116415660 A CN 116415660A CN 202310684277 A CN202310684277 A CN 202310684277A CN 116415660 A CN116415660 A CN 116415660A
- Authority
- CN
- China
- Prior art keywords
- knowledge base
- wafer
- data
- samples
- defect
- 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.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000010276 construction Methods 0.000 title claims description 8
- 239000013598 vector Substances 0.000 claims abstract description 38
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 238000003062 neural network model Methods 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 3
- 238000004590 computer program Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 15
- 238000004891 communication Methods 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 238000009411 base construction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
The invention relates to a method, equipment and medium for constructing and retrieving a wafer defect knowledge base, which comprises the steps of obtaining wafer electrical property test results; setting failure and effective areas, and marking pictures; training the processed data into an image recognition neural network, and extracting the characteristics of the processed data by using the obtained neural network model to obtain vector representation of the training image data; storing all training data feature vectors into a full database; using a clustering algorithm to obtain representative samples and storing the representative samples into a knowledge base; and extracting the characteristics of the data to be detected by using the neural network model to obtain vector representation of the image data to be detected, and calculating the similarity between the characteristic vector of the current data to be detected and the sample characteristic vector in the knowledge base. The invention screens out the sample with typical characteristics in each class through the characteristic vector of the wafer map, and compared with the process of directly searching the full data, the process provided by the invention greatly reduces the searching complexity and improves the searching efficiency.
Description
Technical Field
The invention relates to the technical field of semiconductor manufacturing, in particular to a method, equipment and medium for constructing and retrieving a wafer defect knowledge base.
Background
In the process of manufacturing semiconductor chips, along with the continuous improvement of the chip manufacturing level, the size of the semiconductor device is increasingly biased to be miniaturized, and the miniaturization process has quite a large number of technical indexes and requirements. However, in the production process, various wafer defects cannot be avoided, a wafer map can be obtained after the wafer is subjected to an electrical test, an experienced engineer can determine the defect type of the wafer by detecting and identifying the wafer map, and further analysis and estimation are performed to estimate the mechanism of the defect in the production process.
At present, computer technology is rapidly developed, and more enterprises automatically identify wafer defects by using methods such as neural networks, deep learning and the like. The classification-based method is simple to realize, but the method is difficult to judge new defects; the new category identification can be realized based on the retrieval method, but the wafer yield is huge, the monthly productivity of manufacturers is in units of ten thousand (hundred thousand/million) pieces, and the retrieval efficiency is affected due to the increasing number of samples in the retrieval library along with the iterative updating of the retrieval library.
Therefore, the invention provides a new defect knowledge base construction method and a corresponding search analysis method, so that the problem to be solved in the field is to ensure the identification efficiency while accurately identifying the defect type of the wafer.
Disclosure of Invention
The invention provides a method, equipment and medium for constructing and searching a wafer defect knowledge base, which are used for solving the technical problems of low searching efficiency and low new category searching accuracy in the existing defect searching technology.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a construction and retrieval method based on a wafer defect knowledge base comprises the following steps,
s01: acquiring a wafer electrical property test result;
s02: counting die in a failure state and die in a normal state on a wafer according to an electrical test result, setting a gray value of the position where the failure die is located as 255, setting a gray value of the position where the normal die is located as 127, setting a gray value of the wafer area not to be 0, generating a wafer map, and marking the picture;
s03: marking the existing wafer map data according to expert knowledge, using the existing wafer map data as training data for training a neural network, and storing current network parameters after the neural network is converged to obtain a final network model after the neural network is subjected to multiple rounds of iterative training and learning defects;
s04: extracting features of the data processed in the step S02 by using the neural network model obtained in the step S03, and obtaining vector representation of training image data;
s05, storing all training data feature vectors into a full database, and simultaneously recording the category of the corresponding wafer defect;
s06: using a clustering algorithm to obtain representative samples in the same type of defect data in the full database, and storing the representative samples in a knowledge base;
the number of representative samples to be extracted can be adjusted and set according to the number of clustering centers;
if the total number of samples of a certain type of defects is less than the set clustering number at the beginning, all the samples of the type are stored into a knowledge base;
s07: and extracting the characteristics of the data to be detected by using the neural network model of the S03 to obtain vector representation of the image data to be detected, and calculating the similarity between the characteristic vector of the current data to be detected and the sample characteristic vector in the knowledge base.
Further, if the similarity between the current sample and any sample in the knowledge base is higher than the threshold value of 0.5, the current sample is considered to belong to the category to which the knowledge base sample with the highest similarity belongs, and the current sample is stored in the full-quantity base.
Further, if the similarity between the current sample and all samples in the knowledge base is lower than 0.5, switching the search base to a full database, and calculating the similarity between the feature vector of the current data to be detected and the feature vector of the sample in the full database.
Further, if the similarity between the current sample and all samples in the full database is lower than 0.5, judging that the current sample is of a new defect type, and storing the current sample into the knowledge base and the full database.
Further, in step S03, the dimension of the output feature vector of the neural network is 1024;
further, in step S05, the specific flow of knowledge base construction is screening of typical feature samples: and clustering the same type of data by utilizing kmeans to obtain n clustering centers, selecting a sample closest to the Euclidean distance of the clustering centers as a typical sample, and storing the sample in a knowledge base.
In yet another aspect, the invention also discloses a computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method as described above.
In yet another aspect, the invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the method as above.
According to the technical scheme, a knowledge base-full database two-stage retrieval structure is constructed for wafer image defect retrieval, samples with typical characteristics in each class are screened out through the characteristic vector of the wafer map, and compared with direct full data retrieval, the process provided by the invention greatly reduces the retrieval complexity and improves the retrieval efficiency; compared with a fixed search library mode, the process provided by the invention can update the knowledge library and the full-quantity library in real time, so that the search accuracy is ensured.
Specifically, the method for constructing and searching the wafer defect knowledge base has the following beneficial effects:
(1) The method for constructing the defect search library constructs a two-stage search structure of a knowledge library-full-quantity library, wherein typical samples of wafer defects are contained in the knowledge library, and in the first-stage search process, the search speed can be improved by a plurality of times compared with the search in the full-quantity library.
(2) The retrieval flow provided by the invention simultaneously considers the typical sample and the common sample, classifies the new defect characteristics as new defect types when the difference between the new defect characteristics and the existing sample characteristics is larger than the set threshold value, updates the knowledge base and the full database in real time, and then encounters the defects of the same category, so that the defects can be well identified, and high-quality data can be provided for subsequent root cause analysis.
Drawings
FIG. 1 is a schematic diagram of a method according to an embodiment of the present invention;
FIG. 2 is a flowchart of neural network extraction of image features according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of knowledge base construction in an embodiment of the invention;
fig. 4 is a schematic diagram of a search rule in an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, 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, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention.
As shown in fig. 1, the method for constructing and retrieving a wafer defect knowledge base according to the present embodiment includes the following steps:
s1: and obtaining the wafer electrical property test result.
S2: and counting die in a failure state and die in a normal state on the wafer according to an electrical test result, setting the gray value of the position where the failure die is located as 255, setting the gray value of the position where the normal die is located as 127, and setting the gray value of the wafer area not to be 0.
Specifically, after the gray level map of the wafer map is obtained through conversion, scaling is carried out on the wafer map, the sizes of all the images are guaranteed to be the same, and normalization processing is carried out on the scaled images, wherein the normalization mode is shown in the following formula.
Wherein,,
representing gray values at image coordinates (x, y), for example>
Representing the values at the normalized image coordinates (x, y).
S3: and marking the existing wafer map data according to expert knowledge, using the existing wafer map data as training data for training a neural network, and storing current network parameters after the neural network is converged to obtain a final network model after the neural network is subjected to multiple rounds of iterative training and learning of the defect characteristics.
S4: the flow of extracting features of the neural network is shown in fig. 2, the input image is sent to the neural network after being preprocessed in S2, and the input of the final classification layer is taken as the features of the wafer map after a series of convolution, normalization and activation layers.
Specifically, the features of the wafer image extracted in S4 are 1024 dimensions, and the normalization operation is performed on the feature vectors of 1024 dimensions. The best dimensions for different data, different scenes, are different.
S5: and storing the feature vectors of all the known data into a full database, and recording the category of the corresponding wafer defect.
S6: the construction flow of the knowledge base corresponds to fig. 3, and the flow comprises similar data screening, similar defect feature clustering, counting all clustering centers after clustering is finished, screening samples closest to the clustering centers, taking the samples as typical samples, and storing feature vectors and corresponding defect categories of the typical samples into the knowledge base.
The distance between the cluster center and the sample can be calculated through Euclidean distance and cosine distance.
If the number of defect data in a certain category is less than the number of clustering centers in the step S6 at the beginning, all samples in the category are stored in a knowledge base.
S7: the search rule corresponds to fig. 4, wherein the preprocessing mode and flow of the wafer sample to be detected are consistent with those described in S2, and the flow of the mode of extracting the feature vector of the wafer image by the neural network is consistent with that described in S3, and the description is omitted.
Further, extracting feature vectors of the wafer samples to be detected, calculating the similarity between the current feature vector and all feature vectors in the knowledge base through a similarity calculation formula, if the similarity between the current sample and any sample in the knowledge base is higher than a threshold value of 0.5, considering that the current sample belongs to the category of the knowledge base sample with the highest similarity, and storing the current sample in the full-scale base.
Further, if the similarity between the current sample and all samples in the knowledge base is lower than a threshold value of 0.5, switching the search base to a full-quantity database, and calculating the similarity between the feature vector of the current data to be detected and the feature vector of all samples in the full-quantity database; if the similarity between the current sample and any sample in the full database is higher than the threshold value of 0.5, the current sample is considered to belong to the category of the sample in the full database with the highest similarity, and the current sample is stored in the full database.
Further, if no matching item with the similarity higher than the threshold value of 0.5 exists, the current sample is judged to be a new defect type, and the new defect type is stored in a knowledge base and a full-quantity database.
In the production identification process, the knowledge base and the full database can be updated continuously and iterated, and the change of products can be followed, preferably, the knowledge base can be updated periodically according to specific production yield and process change, so that the representativeness of samples in the knowledge base is ensured.
The invention currently comprises about 10 ten thousand pieces of data in one project, the whole retrieval time is about 0.19s, and after a knowledge base is constructed by extracting a typical sample, the knowledge base comprises about 2000 pieces of data, and the retrieval time is about 0.003s.
In summary, the embodiment of the invention constructs a knowledge base-full database two-stage retrieval structure aiming at wafer image defect retrieval, and samples with typical characteristics in each class are screened out through the characteristic vector of the wafer map, so that compared with the process of direct full data retrieval, the process provided by the invention greatly reduces the complexity of retrieval and improves the retrieval efficiency; compared with a fixed search library mode, the process provided by the invention can update the knowledge library and the full-quantity library in real time, so that the search accuracy is ensured.
In yet another aspect, the invention also discloses a computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method as described above.
In yet another aspect, the invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the method as above.
In yet another embodiment provided herein, a computer program product comprising instructions that, when executed on a computer, cause the computer to perform any of the wafer defect knowledge library-based construction and retrieval methods of the above embodiments is also provided.
It may be understood that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and explanation, examples and beneficial effects of the related content may refer to corresponding parts in the above method.
The embodiment of the application also provides an electronic device, which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus,
a memory for storing a computer program;
and the processor is used for realizing the construction and retrieval method based on the wafer defect knowledge base when executing the program stored in the memory.
The communication bus mentioned by the above electronic device may be a peripheral component interconnect standard (english: peripheral Component Interconnect, abbreviated: PCI) bus or an extended industry standard architecture (english: extended Industry Standard Architecture, abbreviated: EISA) bus, or the like. The communication bus may be classified as an address bus, a data bus, a control bus, or the like.
The communication interface is used for communication between the electronic device and other devices.
The Memory may include random access Memory (Random Access Memory, abbreviated as RAM) or nonvolatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; it may also be a digital signal processor (English: digital Signal Processing; DSP; for short), an application specific integrated circuit (English: application Specific Integrated Circuit; ASIC; for short), a Field programmable gate array (English: field-Programmable Gate Array; FPGA; for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), etc.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, 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.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the 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 scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The construction and retrieval method based on the wafer defect knowledge base is characterized by comprising the following steps,
s01: acquiring a wafer electrical property test result;
s02: counting die in a failure state and die in a normal state on a wafer according to an electrical test result, setting a gray value of the position where the failure die is located as 255, setting a gray value of the position where the normal die is located as 127, setting a gray value of the wafer area not to be 0, generating a wafer map, and marking the picture;
s03: marking the existing wafer map data according to expert knowledge, using the existing wafer map data as training data for training a neural network, and storing current network parameters after the neural network is converged to obtain a final network model after the neural network is subjected to multiple rounds of iterative training and learning defects;
s04: extracting features of the data processed in the step S02 by using the neural network model obtained in the step S03, and obtaining vector representation of training image data;
s05, storing all training data feature vectors into a full database, and simultaneously recording the category of the corresponding wafer defect;
s06: using a clustering algorithm to obtain representative samples in the same type of defect data in the full database, and storing the representative samples in a knowledge base;
the number of representative samples to be extracted can be adjusted and set according to the number of clustering centers;
if the total number of samples of a certain type of defects is less than the set clustering number at the beginning, all the samples of the type are stored into a knowledge base;
s07: and extracting the characteristics of the data to be detected by using the neural network model of the S03 to obtain vector representation of the image data to be detected, and calculating the similarity between the characteristic vector of the current data to be detected and the sample characteristic vector in the knowledge base.
2. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 1, wherein:
extracting feature vectors of the wafer samples to be detected, calculating the similarity between the current feature vector and all feature vectors in the knowledge base through a similarity calculation formula, if the similarity between the current sample and any sample in the knowledge base is higher than a threshold value of 0.5, considering that the current sample belongs to the category of the knowledge base sample with the highest similarity, and storing the current sample in the full-quantity base.
3. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 1, wherein:
if the similarity between the current sample and all samples in the knowledge base is lower than 0.5, switching the search base to a full database, and calculating the similarity between the feature vector of the current data to be detected and the feature vector of the sample in the full database.
4. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 3, wherein:
and if the similarity between the current sample and all samples in the full database is lower than 0.5, judging that the current sample is of a new defect type, and storing the current sample into a knowledge base and the full database.
5. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 1, wherein: after converting in step S02 to obtain the gray scale map of the wafer map, scaling the wafer map to ensure that all the images have the same size, and normalizing the scaled images, where the normalization mode is as follows:
wherein,,
representing gray values at image coordinates (x, y), for example>
Representing the values at the normalized image coordinates (x, y).
6. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 1, wherein: in step S04, the neural network model obtained in the above step S03 is used to extract the features of the data processed in step S02, to obtain a vector representation of the training image data, which specifically includes,
the input image is sent to a neural network after being preprocessed by S02, and the input of a final classification layer is taken as the characteristic of a wafer map after a series of convolution, normalization and activation layers.
7. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 6, wherein: and S04, the extracted wafer image features are 1024 dimensions, and normalization operation is carried out on the 1024-dimension feature vectors.
8. The method for constructing and retrieving wafer defect-based knowledge base as claimed in claim 1, wherein: the construction of the knowledge base comprises similar data screening and similar defect feature clustering, counting all clustering centers after the clustering is finished, screening samples closest to the clustering centers, taking the samples as typical samples, and storing feature vectors and corresponding defect categories of the typical samples into the knowledge base;
the distance between the cluster center and the sample can be calculated through Euclidean distance and cosine distance.
9. A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method of any one of claims 1 to 8.
10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310684277.XA CN116415660A (en) | 2023-06-12 | 2023-06-12 | Construction and retrieval method, device and medium based on wafer defect knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310684277.XA CN116415660A (en) | 2023-06-12 | 2023-06-12 | Construction and retrieval method, device and medium based on wafer defect knowledge base |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116415660A true CN116415660A (en) | 2023-07-11 |
Family
ID=87049580
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310684277.XA Pending CN116415660A (en) | 2023-06-12 | 2023-06-12 | Construction and retrieval method, device and medium based on wafer defect knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116415660A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060215902A1 (en) * | 2005-03-24 | 2006-09-28 | Hisae Shibuya | Method and apparatus for detecting pattern defects |
| CN109034262A (en) * | 2018-08-13 | 2018-12-18 | 东北大学 | A kind of batch processing method of X-ray orientation device defect recognition |
| 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 |
| KR20190081843A (en) * | 2017-12-29 | 2019-07-09 | 주식회사 비스텔 | Method and apparatus for processing wafer data |
| CN111858990A (en) * | 2020-07-28 | 2020-10-30 | 上海喆塔信息科技有限公司 | Wafer map failure mode similarity retrieval method based on convolution classification network |
| CN113077462A (en) * | 2021-04-30 | 2021-07-06 | 上海众壹云计算科技有限公司 | Wafer defect classification method, device, system, electronic equipment and storage medium |
| US20210232872A1 (en) * | 2020-01-27 | 2021-07-29 | Kla Corporation | Characterization System and Method With Guided Defect Discovery |
-
2023
- 2023-06-12 CN CN202310684277.XA patent/CN116415660A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060215902A1 (en) * | 2005-03-24 | 2006-09-28 | Hisae Shibuya | Method and apparatus for detecting pattern defects |
| KR20190081843A (en) * | 2017-12-29 | 2019-07-09 | 주식회사 비스텔 | Method and apparatus for processing wafer data |
| CN109034262A (en) * | 2018-08-13 | 2018-12-18 | 东北大学 | A kind of batch processing method of X-ray orientation device defect recognition |
| 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 |
| US20210232872A1 (en) * | 2020-01-27 | 2021-07-29 | Kla Corporation | Characterization System and Method With Guided Defect Discovery |
| CN111858990A (en) * | 2020-07-28 | 2020-10-30 | 上海喆塔信息科技有限公司 | Wafer map failure mode similarity retrieval method based on convolution classification network |
| CN113077462A (en) * | 2021-04-30 | 2021-07-06 | 上海众壹云计算科技有限公司 | Wafer defect classification method, device, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112990432B (en) | Target recognition model training method and device and electronic equipment | |
| CN109471938B (en) | Text classification method and terminal | |
| CN110969066B (en) | Live video identification method and device and electronic equipment | |
| CN110197205B (en) | Image identification method of multi-feature-source residual error network | |
| CA3066029A1 (en) | Image feature acquisition | |
| CN111368636B (en) | Object classification method, device, computer equipment and storage medium | |
| CN110717881A (en) | Wafer defect identification method and device, storage medium and terminal equipment | |
| CN112818162B (en) | Image retrieval method, device, storage medium and electronic equipment | |
| CN112766218A (en) | Cross-domain pedestrian re-identification method and device based on asymmetric joint teaching network | |
| CN111340213B (en) | Neural network training method, electronic device, and storage medium | |
| CN115171125A (en) | Data anomaly detection method | |
| CN112115996B (en) | Image data processing method, device, equipment and storage medium | |
| CN116258873A (en) | Position information determining method, training method and device of object recognition model | |
| CN116258906A (en) | Object recognition method, training method and device of feature extraction model | |
| CN116415660A (en) | Construction and retrieval method, device and medium based on wafer defect knowledge base | |
| CN116579980A (en) | Printed circuit board defect detection method, medium and equipment based on small sample learning | |
| CN114463746A (en) | Target recognition model training and cell recognition method and device and electronic equipment | |
| CN114595352A (en) | Image identification method and device, electronic equipment and readable storage medium | |
| CN113298166A (en) | Defect classifier, defect classification method, device, equipment and storage medium | |
| CN112446311A (en) | Object re-recognition method, electronic device, storage medium and device | |
| CN110879821A (en) | Method, device, equipment and storage medium for generating rating card model derivative label | |
| CN116610806B (en) | AI-based RPA digital service processing method and computer equipment | |
| CN111625672B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN116758524A (en) | Object identification method and device and electronic equipment | |
| CN114333022B (en) | Training method of character feature extraction model, character recognition method and related equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230711 |