CN116612075A - Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium - Google Patents
Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium Download PDFInfo
- Publication number
- CN116612075A CN116612075A CN202310468168.4A CN202310468168A CN116612075A CN 116612075 A CN116612075 A CN 116612075A CN 202310468168 A CN202310468168 A CN 202310468168A CN 116612075 A CN116612075 A CN 116612075A
- Authority
- CN
- China
- Prior art keywords
- defect
- image
- defect list
- defect detection
- channel
- 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 339
- 238000001514 detection method Methods 0.000 title claims abstract description 162
- 238000010894 electron beam technology Methods 0.000 title claims abstract description 17
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000005259 measurement Methods 0.000 title claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 19
- 238000011156 evaluation Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 230000008901 benefit Effects 0.000 abstract description 9
- 238000000034 method Methods 0.000 description 23
- 238000004422 calculation algorithm Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000007665 sagging Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/22—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by measuring secondary emission from the material
- G01N23/225—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by measuring secondary emission from the material using electron or ion
- G01N23/2251—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by measuring secondary emission from the material using electron or ion using incident electron beams, e.g. scanning electron microscopy [SEM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
The application provides a defect detection method, a device, electron beam measurement equipment and a computer storage medium, wherein the defect detection method comprises the following steps: acquiring a multichannel image; respectively carrying out image registration on each channel image in the multi-channel images to extract a target image; performing defect detection on each target image to obtain a corresponding defect list unit; splicing all the defect list units to obtain a total defect list; screening defects in the total defect list to obtain target defects; the defect detection method can exert the advantages of detection results of all channels, avoid rechecking conflicts when the detection results of all channels are different, ensure the consistency of the detection results of all channels, and further improve the detection precision of multi-channel defect detection.
Description
Technical Field
The present application relates to the field of defect detection technologies, and in particular, to a defect detection method, a defect detection device, an electron beam measurement apparatus, and a computer storage medium.
Background
The image recognition technology has important application value in the aspect of surface defect detection, has wide development prospect in the fields of industrial manufacture, quality detection and the like, can realize automatic operation by utilizing the image recognition technology to detect the surface defect, does not need manual intervention, and reduces the possibility of artificial omission and misjudgment; and the detection efficiency can be greatly improved, defect information with different scales, forms and colors can be adaptively processed, and the defect information can be automatically classified and identified, so that the cognitive limit on the defect form in manual detection can be overcome.
Some to-be-detected pieces have high difficulty in image recognition, and because of special surface materials, fine structures, complex and various defect characteristics, more background interference noise and the like, the wafer is taken as an example, the surface defects are more in variety and can be influenced by ambient illumination to present different appearances, the characteristics are complex and various, accurate classification and recognition are difficult to directly carry out through single image recognition, and negative influence is generated on detection results.
Therefore, when the type of to-be-detected piece adopts different channel images to detect defects or adopts the same channel image to detect different defects, the results may have larger difference (as shown in fig. 1, the detected results of the channel images are different), and thus, the defect false detection or missing detection is caused, so that the yield is reduced.
Disclosure of Invention
The application provides a defect detection method, a defect detection device, electron beam measurement equipment and a computer storage medium, which are used for solving the technical problem that the reliability of a detection result is lacking due to low universal consistency of multichannel defect detection in the prior art.
According to a first aspect of the present application, there is provided a defect detection method comprising: acquiring a multichannel image; respectively carrying out image registration on each channel image in the multi-channel images to extract a target image; performing defect detection on each target image to obtain a corresponding defect list unit; splicing all the defect list units to obtain a total defect list; and screening the defects in the total defect list to obtain target defects.
In a further aspect of the present application, image registration is performed on each of the multi-channel images to extract a target image, including: carrying out image registration on each channel image and the reference image to obtain a corresponding registration score; comparing the registration score with a preset registration threshold; and when the registration score is higher than the registration threshold value, extracting the corresponding channel image as a target image.
In a further aspect of the present application, the splicing of all defect list units to obtain a total defect list includes: taking one defect list unit as a reference defect unit of the total defect list; comparing the defects in each defect list unit with the existing defects in the total defect list in a traversing manner; and splicing the defects which are different from the existing defects in each defect list unit with the existing defects to obtain a total defect list.
In a further aspect of the present application, screening defects in the total defect list to obtain target defects includes: acquiring an evaluation score of each defect in the total defect list; sorting all defects according to the evaluation score; the top n defect is extracted as the target defect.
In a further aspect of the present application, obtaining an evaluation score for each defect in the total defect list includes: performing confidence evaluation on each defect in the total defect list; respectively carrying out weighted assignment on each defect according to the confidence evaluation index to obtain an evaluation score; the confidence evaluation index comprises at least one of quality, stability, importance, defect occurrence number and scene characteristics of the channel image.
In a further aspect of the present application, performing defect detection on each target image to obtain a corresponding defect list unit, including: acquiring a stability index of a target image and a contrast index between the target images; acquiring a defect list unit by adopting a preset first defect detection method under the condition that the contrast index between target images is larger than or equal to a preset first similarity threshold and the stability index is larger than or equal to a first stability threshold; and under the condition that the contrast is smaller than the first similarity threshold and/or the stability index is smaller than the first stability threshold, acquiring the defect list unit by adopting a preset second defect detection method.
In an alternative aspect of the present application, the defect detection method further includes: acquiring position information corresponding to the target defect; and transmitting the position information to the detection channel corresponding to each channel image, and marking on the corresponding channel image.
In a second aspect of the present application, there is also provided a defect detection system comprising: the reading module is used for reading the multichannel images in the preset database; the extraction module is used for carrying out image registration on each channel image in the multi-channel images so as to extract a target image; the defect detection system is used for respectively carrying out defect detection on the target images to obtain corresponding defect list units; the splicing module is used for splicing the defect list units to obtain a total defect list; and the screening module is used for screening the total defect list to obtain target defects.
The application further provides electron beam measuring equipment, which comprises an emitting device, a measuring device and a measuring device, wherein the emitting device is used for emitting electron beams to a wafer; the image acquisition device is used for acquiring and processing the electronic reflection signals on the wafer and outputting a multichannel image; the upper computer is provided with the defect detection system.
Finally, the present application also provides a computer storage medium, where the computer storage medium includes a stored program, and when the program runs, the device where the computer storage medium is controlled to execute the defect detection method described above.
In summary, the application has at least the following beneficial effects:
compared with the method for independently detecting each channel, the method for detecting the multi-channel image comprises the steps of firstly carrying out image registration on each channel image of the multi-channel image to extract a target image, and avoiding false detection caused by matching failure of one or more channel images; meanwhile, the calculation force can be reduced, and the response speed can be improved. The application acquires an independent defect list unit by carrying out defect detection on the target image, utilizes the imaging characteristics of each channel and the advantages of combining detection of different channels, carries out splicing and summarizing on the defect list unit, carries out secondary screening on the spliced total defect list, and captures accurate target defects. The defect detection method can exert the advantages of detection results of all channels, avoid rechecking conflicts when the detection results of all channels are different, ensure the consistency of the detection results of all channels, and further improve the detection precision of multi-channel defect detection.
Additional features and advantages of embodiments of the application will be set forth in the detailed description which follows.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are some embodiments of the application and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a defect marked on a corresponding channel image obtained by performing defect identification by using different channel images in the prior art;
FIG. 2 is a general flowchart of a defect detection method according to an embodiment of the present application;
FIG. 3 is a flowchart showing a step S2 in the defect detecting method according to the embodiment of the present application;
FIG. 4 is a flowchart showing a step S3 in the defect detecting method according to the embodiment of the present application;
FIG. 5 is a flowchart showing a step S4 in the defect detecting method according to the embodiment of the present application;
FIG. 6 is a flowchart showing a step S5 in the defect detecting method according to the embodiment of the present application;
FIG. 7 is a flowchart showing a step S51 in the defect detecting method according to the embodiment of the present application;
FIG. 8 is a flowchart illustrating a defect detection method according to an embodiment of the present application in step S6;
FIG. 9 is a schematic diagram of labeling a defect obtained by the defect detection method according to the embodiment of the present application in a corresponding channel image;
fig. 10 is a schematic block diagram of a defect detection system according to an embodiment of the present application.
Detailed Description
To further clarify the above and other features and advantages of the present application, a further description of the application will be rendered by reference to the appended drawings. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not limiting, as to those skilled in the art.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the specific details need not be employed to practice the present application. In other instances, well-known steps or operations have not been described in detail in order to avoid obscuring the application.
As noted above, the same defect varies in different channels, such as, for example: in some low gray levels, defects may not be accurately identified resulting in missed detection; or in some high gray levels, the defect may be too pronounced, leading to a false positive: the results of channel image defect detection in different channels show different performances, so that for some objects to be detected with complex shapes and tiny geometric features, the defect detection by adopting image recognition can cause the reduction of the production yield and quality of products.
The present application provides a defect detection method, a defect detection device, an electron beam measurement apparatus, and a computer storage medium, which aim to solve the above-mentioned technical problems.
The defect detection method provided by the embodiment of the application can be packaged in an upper computer, and is matched with hardware equipment to execute control, so that the defect automatic detection of the object to be detected is realized in the production process.
Referring to fig. 2, fig. 2 is a general flowchart of a defect detection method according to an embodiment of the application. The defect detection method at least comprises the following steps:
s1, acquiring a multi-channel image;
in step S1, the "multi-channel image" includes a plurality of channel images, each of which has respective information, such as Alpha channel, depth channel, heat map channel, etc., and may describe different physical quantities, and by means of the multi-channel image, the information quantity and dimension of the image may be improved.
Step S1 may use a plurality of detection channels to perform image acquisition to obtain a plurality of channel images. The image acquisition devices in each detection channel may be the same or different.
It will be appreciated that the image acquisition device may be at least one of a camera or a detector.
The multichannel image has the advantage that when certain channels are interfered or noisy, the information of other channels can play a complementary role, so that the anti-interference capability of the image is improved, and higher accuracy can be obtained in subsequent defect detection.
Step S2 is further executed, and each channel image in the multi-channel images is subjected to image registration respectively so as to extract a target image;
the images of the individual channels, and the "channel images" mentioned above, are first separated using code or functions in image processing software or programming language. Exemplary are: the original multi-channel image is extracted into a red channel image, a green channel image, a blue channel image, a depth channel image, a gray channel image, etc. Each channel image can provide different information, for example, a gray channel can provide more comprehensive image information, and the method is suitable for detecting defects without color difference, a blue channel can provide better detection effect, and the like, which are not exemplified herein.
By carrying out image registration on each channel image, discarding images with poor quality from a plurality of channel images and screening out target images, the image quality can be improved, and the influence caused by factors such as noise, distortion and the like can be eliminated in the subsequent detection link, so that the definition of the images and the accuracy of defect detection are improved. Image registration refers to the process of accurately aligning the channel images. The aim is to establish the correspondence between the channel images so that the channel images overlap spatially for comparison and analysis.
In the step, after each channel image is registered, the extracted target image continues to perform defect detection in the step S3, and other images do not need to perform defect detection, so that the number of channel images for performing defect detection is reduced, the corresponding speed of detection can be improved, and the detection task can be completed in a shorter time to improve the efficiency.
Step S3 is further executed, defect detection is carried out on each target image, and a corresponding defect list unit is obtained;
in step S3, the target image may be subjected to defect detection using a model based on threshold segmentation, edge detection, texture analysis, or the like. And selecting a matching algorithm according to the detection requirement of the object to be detected, processing each channel image to obtain a corresponding defect binary image, and carrying out classification numbering on defect information mapped by the defect binary image to obtain a defect list unit.
In an embodiment of the present application, the defect detection method further includes:
step S4, splicing all the defect list units to obtain a total defect list;
and S5, screening defects in the total defect list to obtain target defects.
In step S4, for the same kind or different defect list units, the defect list units may be combined into a total defect list and arranged according to a certain ordering rule, so that the number and importance degree of different kinds of defects can be conveniently checked and compared, the defect condition existing in the image can be more intuitively understood, and the corresponding processing decision is matched.
And then, in step S5, the target defects which can represent the real defects are directly screened from the total defect list, so that the consistency of defect detection is realized.
In summary, the general inventive concept of the embodiment of the application provides a defect detection method, which fully utilizes imaging characteristics of an image acquisition device in different channels to judge image registration results of all channels, performs defect detection on images of all channels and splices detection results on the premise of ensuring accurate image registration, and then performs unified screening to obtain accurate and unified detection results among all detection channels. Compared with the prior art, the method has higher detection accuracy and stronger detection robustness, thereby being beneficial to improving the detection accuracy of multi-channel defect detection and improving the production efficiency and the product quality of products.
The following specifically describes each step in the general inventive concept:
with continued reference to fig. 3, fig. 3 is a specific flowchart of step S2 in the defect detection method according to the embodiment of the present application. Step S2 of image registration of each of the multi-channel images to extract a target image includes:
step S21, performing image registration on each channel image and a reference image to obtain a corresponding registration score;
step S22, comparing the registration score with a preset registration threshold;
and S23, when the registration score is higher than the registration threshold, extracting the corresponding channel image as a target image.
It will be appreciated that the channel image in one of the channels may be selected as a reference image (reference image) and the other channel images as test images (test images), with the test images being in a pass registration with the reference image by a matching algorithm model. The channel images and the reference images are sequentially led into a registration algorithm model according to a preset sequence to perform image registration.
In step S21, a channel image which is suitable for defect detection of the product, is clear and has high quality can be selected as a reference image, and the reference image usually contains more details and has obvious contrast when corresponding to the defect detection of the product, thereby being beneficial to improving registration accuracy. As for the surface defects of the wafer, secondary electron image (SE) channel images or reflected electron (BEL) channel images may be preferred as reference images.
The registration algorithm model may not be limited, for example, a registration algorithm based on an image histogram, a regularization registration algorithm based on mutual information, a registration algorithm based on a peak signal-to-noise ratio, and the like, and specifically needs to consider similarity and difference between channels, and types and parameters of image transformation. Flexibly set according to the application scene. At the same time of image registration, the selection of the registration algorithm also needs to combine the characteristics of the channels to ensure the accuracy and efficiency of registration.
By carrying out image registration on each channel image, discarding images with poor quality from a plurality of channel images and screening out target images, the image quality can be improved, and the influence caused by factors such as noise, distortion and the like can be eliminated in the subsequent detection link, so that the definition of the images and the accuracy of defect detection are improved.
The registration score is a quantitative indicator that evaluates the degree of similarity between the reference image and other channel images to calculate the similarity between the channel images and the reference image to determine whether the channel images are available for defect detection.
In one specific example, to first read the reference image and the channel image; extracting characteristic points of the channel image and the reference image respectively through extraction codes; then using an image matching method, such as a FLANN algorithm to match the characteristic points of the reference image and the channel image, and screening out the best matching point pair of the reference image and the channel image; in an alternative, the matching score=matching point pair/feature point number, where "feature point number" is the minimum value of the feature point number of either of the reference image and the channel image.
A target image repository may be preset, and when the registration score is higher than a preset first threshold, for example, 0.75, the target image is extracted and stored in the repository to facilitate the defect detection in the next step, otherwise, the next instruction is abandoned below the first threshold. The method is used for providing an image foundation with higher quality for subsequent defect detection so as to improve the accuracy of image identification, reduce the calculation force of defect detection and accelerate the calculation response.
With continued reference to fig. 4, fig. 4 is a flowchart illustrating a specific step S3 in the defect detection method according to the embodiment of the present application. Performing defect detection on the preprocessed target image in step S3 to obtain defects includes:
step S31, acquiring a stability index of a target image and a contrast index between the target images;
step S32, acquiring a defect list unit by adopting a preset first defect detection method under the condition that the contrast index between target images is larger than or equal to a preset first similarity threshold and the stability index is larger than or equal to a first stability threshold;
step S33, acquiring a defect list unit by adopting a preset second defect detection method under the condition that the contrast is smaller than a first similarity threshold and/or the stability index is smaller than a first stability threshold;
step S34, classifying and numbering the defects to form corresponding defect list units.
It should be noted that the above-mentioned "first defect detection method" and "second defect detection method" are only used for descriptive purposes, and are merely for characterizing that the defect methods adopted in the two cases of step S32 and step S33 are different, and do not imply that only two defect detection methods are included, that is, different types of defect detection methods may be included in the first defect detection method.
In step S31, the stability index of the image may be determined by taking the variance, standard deviation, root mean square error, peak signal to noise ratio, etc., and the variance is calculated by, for example, calculating the difference between each pixel in the target image and the average value of the whole image, and then summing the squares of the differences. The larger the variance is, the worse the stability index for the channel image, and vice versa.
The contrast index can be understood as the difference between different colors or gray levels in the image, and the contrast index can use gray level histogram, construction of Laplacian pyramid and other algorithms.
In an alternative, a gray level histogram of the target image may be calculated, with the difference between the maximum and minimum values in the histogram being used as a calculated calibration for the contrast index.
In step S32 to step S33, the same defect is used for the target image with higher and more stable contrast index, whereas different defects are used when the contrast index is lower or the stability index is worse.
In an actual application example, the first defect detection method may adopt feature extraction, then perform threshold segmentation, morphological operation, edge detection, and the like, and finally perform morphological analysis on the detected defect, and identify and classify the detected defect according to the shape, size, position, and other features of the different defects.
Under the condition that the contrast is lower than the first similarity threshold and/or the stability index is lower than the first stability threshold, the second defect detection method can be used for preprocessing the condition of the target image, improving the quality of the target image, such as noise removal, contrast enhancement and the like, and then performing feature extraction, morphological analysis and the like, so that the detection result is more true.
The efficiency and the accuracy of defect detection can be greatly improved by steps S31 to S33.
In step S34, the corresponding defect list unit is formed by classifying and numbering the defects.
In a specific embodiment, firstly, defining defect types, such as weld sagging, air holes, cracks, long holes, uneven penetration and the like, which are generated by a wafer, and establishing a mapping relation between the defect types and related forms in advance;
each defect type is numbered classified for it. For example, weld sagging may be divided into a 1-start numbering sequence, air pockets may be divided into a 2-start numbering sequence.
After the defect detection in step S32 or step S33 is completed, the defects and the corresponding target images are numbered, and Channel 1-1, channel 2-1, channel 5-2. The number sequences are summarized to form a corresponding defect list unit, and the defect list unit can represent a specific defect and can comprise related information such as a file, a line number, a defect list unit and the like where the defect list unit is located by configuring the defect list unit.
With continued reference to fig. 5, fig. 5 is a flowchart illustrating a specific step S4 in the defect detection method according to the embodiment of the present application. Splicing the defect list unit to obtain a total defect list in step S4 includes:
step S41, taking one of the defect list units as a reference defect unit of the total defect list;
step S42, comparing the defects in each defect list unit with the existing defects in the total defect list in a traversing way;
step S43, the defect which is different from the existing defect in each defect list unit is spliced with the existing defect to obtain a total defect list.
Step S41 may preferably use the first defect list unit as a reference defect unit, and it is understood that the first defect list unit does not necessarily correspond to the channel image with the sequence 1 since the channel image is filtered in step S2.
And (3) comparing the other defect list units after the reference defect unit with the reference defect unit in sequence, and adding the total defect list when the same defects exist, namely merging, and otherwise, until all the defect list units are traversed, so as to finally form the total defect list.
It can be understood that the number of merging is counted in the summarizing process, or the frequency of occurrence of each defect result is directly counted, and the higher the relative frequency of occurrence of the same defect is, the higher the authenticity of the fed-back defect is. In the process, the authenticity of the detection result can be greatly optimized, the advantages of the detection results of all channels are exerted, the recheck conflict existing when the detection results of all channels are different is avoided, and the consistency of the detection results of all channels is ensured.
With continued reference to fig. 6 and fig. 7, fig. 6 and fig. 7 are specific flowcharts of step S5 and step S51 in the defect detection method according to the embodiment of the present application, respectively. Screening the total defect list to obtain the target defect in step S5 includes:
step S51, obtaining an evaluation score of each defect in the total defect list;
step S52, sorting all defects according to the evaluation score;
step S53, extracting the defect of the n rank as the target defect.
Step S5 may be understood as the subsequent adjustment and optimization of the final output structure after step S4, because each defect is different in type, the reliability of the characterization of the final target defect by singly using the number of defects is not high; and S5, weighting the total list detection results in the evaluation, finally summarizing, calculating and screening the defects of the previous n as target defects, and evaluating and optimizing the defect detection and identification results to improve the accuracy and reliability of detection and identification.
The size of n should be determined according to the specific situation of the wafer produced in the process, if a plurality of surface defects may be generated simultaneously in the process, n may be set to be more than 2, otherwise n may be set to be 1.
In a specific embodiment, assigning the defects in the total defect list in step S51 includes:
step S511, performing confidence evaluation on each defect in the total defect list, wherein confidence evaluation indexes comprise at least one of quality, stability, importance, defect occurrence number and scene characteristics of the channel image;
and S512, respectively carrying out weighted assignment on each defect according to the confidence evaluation index to obtain an evaluation score.
It is understood that each defect is numerically calibrated by assigning a confidence assessment to each defect. The value of the final confidence may feed back the importance, severity and authenticity of the defect, generally the higher the score, the more the defect needs to be focused and addressed.
And then carrying out weighted assignment on all defects according to the confidence level in sequence. Different weighting factors can be defined respectively for the quality, the stability, the importance, the number of times of occurrence of detection defects, the scene characteristics and the like of the images of each channel, and the size of the weighting factors can be determined according to specific situations and can be adjusted through experiments or experience. The score is adjusted according to the magnitude of the confidence value. For example, a defect with a higher confidence may be multiplied by a larger weighting factor and a defect with a lower confidence may be multiplied by a smaller weighting factor.
With continued reference to fig. 8, fig. 8 is a flowchart illustrating a specific step S6 in the defect detection method according to the embodiment of the present application. In a further aspect of the embodiment of the present application, the defect detection method further includes:
step S61, obtaining position information corresponding to the target defect;
step S62, position information is issued to a detection channel corresponding to each channel image, and identification is carried out on the corresponding channel image;
and acquiring position information of the defect through a defect detection algorithm, and converting the position information into a coordinate form in the image. Different conversion methods, such as pixel coordinates, actual coordinates, etc., may be employed according to different image coordinate systems. The converted image coordinate information is then issued to an image processing module of the plurality of inspection channels to identify corresponding defect areas in the images of the respective channels.
It will be appreciated that the manner of identification may be any of target framing, color filling, and contouring, specifically adjusted as desired, as ultimately shown in fig. 9.
And step S63, outputting and storing the identified channel image.
The channel images are stored and named, the naming can be performed by time stamping or specific defects, and the problem tracing is conveniently performed after the function call is used subsequently.
In summary, compared with the method for separately detecting each channel, the general inventive concept of the embodiment of the application provides a defect detection method, which registers channel images to extract a target image, avoids the occurrence of false detection caused by failure of matching of one or more channel images, and can reduce calculation force and improve response speed. The application subsequently carries out defect detection on the target image to obtain a winning defect list unit, utilizes the imaging characteristics of each channel and the advantages of combining detection of different channels, carries out splicing and summarizing on the defect list unit, carries out secondary screening on the spliced total defect list, and captures accurate target defects. The defect detection method can play the advantages of detection results of all channels, avoid rechecking conflicts when the detection results of all channels are different, ensure the consistency of the detection results of all channels, and further improve the detection precision of multi-channel image defect detection.
[ Defect detection System ]
Referring to fig. 10, fig. 10 is a schematic block diagram of a defect detection system according to an embodiment of the application; the defect detection system 100 includes:
a reading module 101, configured to read a multi-channel image in a preset database;
an extracting module 102, configured to perform image registration on each channel image in the multi-channel images, so as to extract a target image;
a defect detection module 103, configured to perform defect detection on each target image to obtain a corresponding defect list unit;
a splicing module 104, configured to splice all the defect list units to obtain a total defect list;
and the screening module 105 is configured to screen the defects in the total defect list to obtain target defects.
It is to be understood that the various modules of the system of the present application may be implemented in whole or in part in software, hardware, firmware, or a combination thereof. The modules may each be embedded in a processor of the electronic device in hardware or firmware or may be independent of the processor, or may be stored in a memory of the electronic device in software for the processor to call to perform the operations of the modules. Each module may be implemented as a separate component or module, or two or more modules may be implemented as a single component or module.
Electron beam measuring apparatus
An embodiment of the present application further provides an electron beam measuring apparatus (not shown), which is applicable to wafer inspection, and includes:
an emission device for emitting an electron beam to the wafer;
the image acquisition device is used for acquiring and processing the electronic reflection signals on the wafer and outputting a multichannel image;
the host computer includes the defect detection system 100 described above.
The defect detection equipment can be arranged at the periphery of a detection station of the wafer and is used for detecting the wafer in real time.
It can be understood that the electron beam measuring device can emit an electron beam through the emitting device, the emitting device can control parameters such as the direction and the energy of the electron beam, the image collecting device can collect three-dimensional non-contact by scanning the surface of an object of a wafer, can measure electronic reflection signals such as complex shapes, tiny geometric features and the like, and can process the collected electronic reflection signals, so that a multichannel image of an object to be detected is obtained.
The upper computer reads the multichannel image of the image acquisition device through a serial port or wireless transmission mode, and the defect detection method is carried out on the multichannel image, so that the detection precision of the electron beam measuring equipment is improved.
It is to be understood that the specific features, operations and details described herein before with respect to the method of the application may also be similarly applied to the apparatus and system of the application, or vice versa. In addition, each step of the method of the present application described above may be performed by a corresponding component or unit of the apparatus or system of the present application.
The present application provides a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the above-described defect detection method.
Those skilled in the art will appreciate that the method steps of the present application may be performed by a computer program to instruct related hardware such as an electronic device or a processor, and that a computer program implementing the above-mentioned defect detection method may be stored in a non-transitory computer readable storage medium, which when executed causes the steps of the present application to be performed. Any reference herein to memory, storage, or other medium may include non-volatile or volatile memory, as the case may be. Examples of nonvolatile memory include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage, hard disk, solid state disk, and the like. Examples of volatile memory include Random Access Memory (RAM), external cache memory, and the like.
The technical features described above may be arbitrarily combined. Although not all possible combinations of features are described, any combination of features should be considered to be covered by the description provided that such combinations are not inconsistent.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application.
Claims (10)
1. A defect detection method, comprising:
acquiring a multichannel image;
respectively carrying out image registration on each channel image in the multi-channel images so as to extract a target image;
performing defect detection on each target image to obtain a corresponding defect list unit;
splicing all the defect list units to obtain a total defect list;
and screening the defects in the total defect list to obtain target defects.
2. The defect detection method of claim 1, wherein the performing image registration on each of the multi-channel images to extract the target image comprises:
carrying out image registration on each channel image and a reference image to obtain a corresponding registration score;
comparing the registration score with a preset registration threshold;
and when the registration score is higher than the registration threshold value, extracting the corresponding channel image as a target image.
3. The defect detection method of claim 1, wherein the splicing all the defect list units to obtain a total defect list comprises:
taking one defect list unit as a reference defect unit of the total defect list;
comparing the defects in each defect list unit with the existing defects in the total defect list in a traversing manner;
and splicing the defects which are different from the existing defects in each defect list unit with the existing defects to obtain a total defect list.
4. The defect detection method of claim 1, wherein the screening defects in the total defect list to obtain target defects comprises:
obtaining an evaluation score of each defect in the total defect list;
sorting all defects according to the evaluation score;
the top n defect is extracted as the target defect.
5. The defect detection method of claim 4, wherein the obtaining an evaluation score for each defect in the total defect list comprises:
performing confidence evaluation on each defect in the total defect list;
respectively carrying out weighted assignment on each defect according to the confidence evaluation index to obtain an evaluation score;
wherein the confidence evaluation index comprises at least one of quality, stability, importance, defect occurrence number and scene characteristics of the channel image.
6. The defect detection method according to claim 1, wherein performing defect detection on each of the target images to obtain a corresponding defect list unit comprises:
acquiring a stability index of the target image and a contrast index between the target images;
acquiring a defect list unit by adopting a preset first defect detection method under the condition that the contrast index between target images is larger than or equal to a preset first similarity threshold and the stability index is larger than or equal to a first stability threshold;
and under the condition that the contrast is smaller than the first similarity threshold and/or the stability index is smaller than the first stability threshold, acquiring a defect list unit by adopting a preset second defect detection method.
7. The defect detection method according to any one of claims 1 to 6, further comprising:
acquiring position information corresponding to the target defect;
and transmitting the position information to a detection channel corresponding to each channel image, and marking the corresponding channel image.
8. A defect detection system, comprising:
the reading module is used for reading the multichannel images in the preset database;
the extraction module is used for carrying out image registration on each channel image in the multi-channel images respectively so as to extract a target image;
the defect detection system is used for carrying out defect detection on each target image to obtain a corresponding defect list unit;
the splicing module is used for splicing all the defect list units to obtain a total defect list;
and the screening module is used for screening the defects in the total defect list to obtain target defects.
9. An electron beam measurement apparatus, comprising:
an emission device for emitting an electron beam to the wafer;
the image acquisition device is used for acquiring and processing the electronic reflection signals on the wafer and outputting a multichannel image;
a host computer equipped with the defect detection system according to claim 8.
10. A computer storage medium, characterized in that the computer storage medium comprises a stored program, wherein the program, when run, controls a device in which the computer storage medium is located to perform the defect detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310468168.4A CN116612075A (en) | 2023-04-27 | 2023-04-27 | Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310468168.4A CN116612075A (en) | 2023-04-27 | 2023-04-27 | Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116612075A true CN116612075A (en) | 2023-08-18 |
Family
ID=87677263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310468168.4A Pending CN116612075A (en) | 2023-04-27 | 2023-04-27 | Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116612075A (en) |
-
2023
- 2023-04-27 CN CN202310468168.4A patent/CN116612075A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110349126B (en) | Convolutional neural network-based marked steel plate surface defect detection method | |
| CN111462225B (en) | Centroid identification and positioning method of infrared light spot image | |
| US10861145B2 (en) | Defect inspection device and defect inspection method | |
| CN107014294B (en) | Contact net geometric parameter detection method and system based on infrared image | |
| CN108711148B (en) | Tire defect intelligent detection method based on deep learning | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN116721107B (en) | Intelligent monitoring system for cable production quality | |
| CN116754484B (en) | Nondestructive testing method for nonmetallic liner fiber winding container | |
| CN113758932A (en) | Lithium battery diaphragm defect vision system based on deep learning | |
| CN113820333B (en) | Battery pole piece abnormality detection method, device, upper computer and detection system | |
| CN117173640B (en) | Beef cattle temperature anomaly detection method | |
| CN115841488A (en) | Hole checking method of PCB (printed Circuit Board) based on computer vision | |
| CN113533372B (en) | Paper defect detection method, system, device and computer storage medium | |
| CN117195439B (en) | Intelligent optimization method for large-shaft-diameter full-section mechanical sealing performance | |
| CN100353368C (en) | Particle extraction for automatic flow microscope | |
| CN115984283B (en) | Intelligent detection method for welding quality of reinforcement cage | |
| CN116612075A (en) | Defect detection method, defect detection device, electron beam measurement equipment and computer storage medium | |
| CN112036391A (en) | Electronic inspection method and system based on shale gas field production station | |
| CN108363958B (en) | Oil tank detection method based on high-resolution optical remote sensing image | |
| CN116612461A (en) | Target detection-based pointer instrument whole-process automatic reading method | |
| CN115830302A (en) | Multi-scale feature extraction and fusion power distribution network equipment positioning identification method | |
| CN111507177B (en) | Identification method and device for metering turnover cabinet | |
| CN114998194A (en) | Product defect detection method, system and storage medium | |
| CN113822104A (en) | Artificial intelligence water surface detection system based on virtual scale of many candidates | |
| CN113344858B (en) | Feature detection method, device and computer storage medium |
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 | ||
| CB02 | Change of applicant information |
Address after: 100176 building 12, yard 156, Jinghai 4th Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing Applicant after: Dongfang Jingyuan Microelectronics Technology (Beijing) Co.,Ltd. Address before: 100176 building 12, yard 156, Jinghai 4th Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing Applicant before: DONGFANG JINGYUAN ELECTRON Ltd. |
|
| CB02 | Change of applicant information |